Energy Department Paper No. 11 E1 Power System Load Management Technologies November 1983 World Bank Energy Department POWER SYSTEM LOAD MANAGEMENT TECHNOLOGIES Final Report of Research Project No. R648 Prepared by Resource Dynamics Corp. (USA) for the World Bank Energy Department November 1983 Copyright (c) 1983 The World Bank 1818 H Street, N. W. Washington, D. C. 20433 U.S.A. This paper is one of a series issued by the Energy Department for the information and guidance of Bank staff. The paper may not be published or quoted as representing the views of the Bank Group, nor does the Bank Group accept responsibility for its accuracy or completeness. POWER SYSTEM LOAD MANAGEMENT TECHNOLOGIES Abstract In recent years, techniques referred to as load management have begun to play an important role in shaping the patterns of elec- tricity consumption in industrialized countries. Along with pricing, a variety of hardware is used to control loads directly and save on energy and peak capacity. This study reviews the state-of-the-art of these so- called "hard" techniques in light of recent technological advances, provides data on cost and manufacturers of this equipment, and identi- fies controllable loads in developing countries. ei EXECUTIVE SUMMARY AND CONCLUSIONS In many power systems of developing countries, peak demand is accommodated in two ways: costly and underutilized additional invest- ments along with the use of the most expensive fuels, or as a last resort, forced and often haphazard power cuts which can be economically costly. As the industrialized countries have learned there exists a third alternative of load management whereby demand patterns can be changed into a more tractable shape in the interest of both the customer and the utility. ' Cement factories, refineries, chemical industries, irrigation networks exhibit flexibility in cutting down their demand for a short time. The first way to achieve this result is through pricing incentives such as systematic discounts on off-peak rates, or low cost interruptible service if critical situations are harder to predict. But in many cases these incentives do not exist or they must be complemented with some kind of direct load control ranging-from the telephone call to the remote switching of thousands of appliances. This report focuses on the technical features and cost of these so-called "hard" techniques: hardware at the point of control such as timers, relays, storage, back-up facilities and in the case of remote control, communications and control center. The study goes beyond component analysis and hypothesizes some typical control system configurations in order to assess the cost per control point and its sensitivity to some key parameters. Cost estimates generally exclude installation and maintenance; these correspond to country specific labor costs generally less than the spread of prices quoted for each equip- ment. The report includes qualitative assessments about the type of power system and demand pattern that are most conducive to load manage- ment and on the contractual arrangements which can be worked out. As the example of Kenya shows, short curtailments of power loads are man- ageable in the industrial and agricultural sectors of the developing world and, for the more advanced countries, in the commercial and resi- dential sectors. In that respect, thermoelectric systems with marked daily peaks show better prospects than hydroelectric ones, which tend to be more constrained by energy availability over a long dry period than by capacity limitations. But even in that case, load management may have some value in deferring hydropower capacity additions when these are costly or do not bring about significant fuel savings. iii Generally a big share of the load pertaining to a few large users can be managed through existing telephone lines and other lesser ones with simple local controls (timers, priority relays, etc.). The least favorable case occurs when remote control is required for a low density and small size population; but even then, a solution involving radio links would be paid back in less than a year by a peak "shaving" of ten kilowatts. For longer interruptions, the advantages of load management are not so clear-cut except in the many instances where investments in stand-by or storage capacity have been already made for other reasons. Clearly, a cost-benefit evaluation of load management strate- gies must consider not only the hardware costs quantified in this report but also the broader economic costs of demand curtailment as against the savings in fuel and investment anticipated by the utility. These sav- ings are all reflected in the system marginal costs, and the extent to which they are passed along to consumers in their power bills will be a key factor in developing a lasting and sound load management strategy. Insofar as the marginal cost and load analysis as well as the adopted pricing postures are country-specific, the evaluation and design of a load management project should be conducted on a case-by-case basis and coordinated with the utility's planning and rate making processes. iv TABLE OF CONTENTS Page I. INTRODUCTION AND SUMMARY ...................... 1: Applicability of Load Management Systems .... 1 to Developing Nations Customer Considerations ................. 3 Utility Considerations .................... 4 Organization of Report ................ 4 Load Management System Summary .............. 5 II. LOAD MANAGEMENT TECHNOLOGIES .................. 13 Types of Load Management Technologies .....** 14 Local Control and Communication ...........15 Systems Thermal Energy Storage Systems ............ 16 Supplemental Energy Systems ............... 1.7 Component Costs of Load Management Systems ** 18 Local Controllers .......................... 19 Remote Control Systems .................... . 20 .Communications and Information Systems *** 21 Thermal Storage Systems ................... 23 Active Solar Energy Systems ............... 25 Passive Solar Energy Systems ............... 26 - Cogeneration Systems ....................... 27 Dual Heating Systems ....................... 28 III. SYSTEM DESIGN CONSIDERATIONS .................. . 38 Point of Control ............................ 38 Local Controllers ......................... 38 Remote Control Systems .................... 41 Communications and Information Systems .... 42 Thermal Energy Storage Systems ............ 43 Supplemental Energy Systems ............... 46 Communications and Central Control .......... 49 Total Costs for Remote Control Systems ...... 54 Cost Estimates ............................ 55 Results ................................... 58 Base Case Comparison ...................... 58 Economies of Scale ......................... 59 Dispersed Case ............................. 59 Adverse Conditions ........................ 61- Bidirectional ..........*.................... 62 Sophistication ............................ 62 High Cost Case ............................ 62 TABLE OF CONTENTS (CONTINUED) Pa ge IV. UTILITY AND CUSTOMER CONSIDERATIONS ...... 63 Load Management Potential in Developing Nations 63 Supply Adequacy.............................. 63 System Load Considerations .................. 63 Generation Mix............................. 64 Management Capabilities .................... 65 Customer Acceptance........................ 65 Industrial Customers........................... 66 General Principles......................... 67 Industrial Electricity End Uses ............ 68 Load Management in Specific Industries ..... 69 Institutional Considerations............... 70 Other Customers ............................... 75 Commercial/Institutional ................... 75 Residential ................................ 76 Agriculture . Matching Load Management Systems to Types ....* of Loads Local Controllers.......................... 79 Remote Control Systems ...................... 79 Communications and Information (C&I) ....... Systems Thermal Energy Storage Systems ............. 80 Supplemental Energy Storage Systems ........ 80 Contractual Arrangements ........................ 4 Conclusions......................................88 APPENDICES Appendix A Vendors of Load Management Equipment .... 90 Appendix B Approach to Equipment Cost Assumptions .... LZ29 for Remote Control Technologies TABLE OF CONTENTS (CONTINUED) LIST OF FIGURES AND TABLES Page Figure 1-1 Load Management Considerations . 2 Table 1-1 Description of Load Management ........ 6 Systems Table 2-1 Load Management Equipment and Systems. 28 Table 2-2 Time Controllers ..................... 29 Table 2-3 Demand Limiters ...................... 30 Table 2-4 Load Management Thermostats .......... 30 Table 2-5 Ripple System Component Costs ....... 31 Table 2-6 Power Line Carrier Component Costs ... 32 Table 2-7 Radio (Direct) Component Costs ....... 33 Table 2-8 Hybrid (Radio/PLC Systems) Component 34 Costs Table 2-9 Telephone (Signal) Systems Component 35 Costs Table 2-10 Coaxial Cable Component Costs ........ 36 Table 2-11 Solar Systems for Space Heating ....... 37 Table 2-12 Typical Solar Domestic Hot Water ..... 37 Heater Costs Table 3-1 Geographic Considerations Related.... 51 to Communication Systems Table 3-2 Other Communication System .......... 52 Considerations Table 3-3 Definition of Cases .................. 56 Table 3-4 Assumed Quantity Discounts ........... 57 Table 3-5 Equipment Costs of Remote Control .... 59 Systems Table 3-6 Base Case Cost Per Point for Different 60 Size Systems Table 4-1 Applicability of Load Management ...... 81 Technologies to Customer Loads Table 4-2 Considerations for Agreements . 84 on Load Management Services -1- Chapter 1 INTRODUCTION AND SUMMARY This report evaluates electric load management systems and their ability to control loads in developing nations. Advantages, disadvan- tages and costs of available load management systems are described. Customer and utility characteristics which determine the need for, and usefulness of, load management systems are also reviewed. APPLICABILITY OF LOAD MANAGEMENT SYSTEMS TO DEVELOPING NATIONS Electric utilities in developing nations strive to bring the bene- fits of electric service to new customers and to expand the benefits realized by existing customers. These utilities often face great difficulties in achieving these objectives. Many are growing quickly at a time when new capacity and energy' costs are rapidly increasing. New load is added, but load factors may be initially low because of the limited numbers and high coincidence of new customer end uses. Low load factors increase electric power costs because significant portions of the generation, transmission and distribution capacity needed to meet peak loads are idle for much of the time. Increased load factors improve plant utilization and limit investments in new capacity. Whether increased load factors. can be achieved is deter- mined by how load varies over time. Load management is used by electric utilities to change patterns of electricity consumption over time. -The primary objectives of load management are usually either to reduce utility system peak demands or to shift consumption at the time of peak system demand to .other periods. These actions are intended to bring about improvements in load factor. Sometimes, additional objectives--such as energy conser- vation-also may be achieved through the use of load management techniques. Many technologies for load management are currently available and a variety of- new ones are currently under development. Not all are equally useful in every circumstance. Some work better than others, given specific conditions of load, utility characteristics and service area. This report describes some of the tradeoffs which must be considered to properly select a load management system for a specific application. Figure 1-1. Load Management Considerations F-~~~~ ~ . -7- o____ r____ As Figure 1-1 illustrates, the evaluation of A load .management strategy must consider the characteristics of customers, of the utility, and of the load management systems': * Load,manazement system considerations relate to the capabili- . ties and - limitations of the load management technologies-. These may be further categorized according to type of compo- nent (e.g., point of control, communication systems, and central control) Appropriate to the.specific technology. * Customer considerations relate to the types of loads -which may be controlled, and to the institutional and business relationships between the customer and the utility. A parti- cularly important component of this relationship is the rate which a customer pays for electric services. Utility rate A Fstructures may provide varying levels of economic incentives or disincentives to customers to participate in a load management program. * Utility considerations relate to the technical and economic need for load management and to the management and financial capability of the utility to support a load management program. In general, load management technology is available, tested, and often flexible enough to be applied under a variety of circumstances. While utility conditions differ from country to country, the basic principles are essentially the same. The major differences among countries are in customer considerations. -3- Customer Considerations The economic and social infrastructures of developing nations vary considerably-differing both from those of developed countries and among themselves. Loads which are the focus of control in developed countries, with load management experience (e.g., winter electric space heating in Europe) may not be significantly present in many developing nations. Other loads, generally insignificant in a developed country, may be crucial in developing nations (e.g., irriga- tion pumping). Further, the relationship between utility and customer may be different in areas where electricity has long been supplied and areas which are only now being or have only recently been electrified. The presence of specific loads is influenced by both climate and status of development. Load management techniques were initially developed in Europe, where utilities were winter-peaking because of high space heating loads. Utilities in other countries may have loads which peak in different seasons. Some developing nations, for instance, experience summer peaks because of increased fan use or irrigation pumping. These different loads and seasonality patterns imply the need for different load management strategies. There are four basic electricity-consuming sectors in which load management may be used: industrial, residential, commercial and agricultural. The indistrial sector consists of a diverse set of electricity consumers, although the industrial base of some countries may be quite limited. In many cases, a single industrial facility may represent .a significant portion of a country's electric load. The degree of flexibility that industrial customers have to absorb power cuts varies dramatically. In almost all cases, the primary concern of. plant management will be maintaining production, not conducting load manage- ment. Utility industrial load management practice throughout the world has been to design tariff incentives and to let the industrial customers respond to these incentives. Sometimes, direct- communica- tions with a small number of large industrial customers is part of the program. In most cases, this tariff approach has been effective. Residential loads, *b.y contrast, are frequently controlled by specific local control, remote control and energy storage devices or by combinations of these devices. Many residential customers in some developing nations, however, may have few controllable loads. Commercial loads vary and sometimes have characteristics of either industrial or residential customers. Some types of commercial institutional customers may have fixed hours of operation and thus may find it difficult to shift loads. The types of load management systems which may be applied to commercial facilities vary accordingly. -4- The primary agricultural load which may be subject to load manage- ment in many countries is irrigation pumping. Management of this load must be carefully controlled in order not to disrupt crop growing cycles. Utility Considerations Several considerations affect the usefulness of load management to an electric utility. These include supply adequacy, load factor, generation mix, management and financial strength. . The significance of each of these considerations to a specific developing country depends upon the unique situation in that country. The relationship between supply and demand requirements is a major consideration in assessing the potential for load management. If load growth is greater than planned generating capacity additions, then reserve margins and reliability fall, and the utility.has difficulties in meeting load. Load management may be a particularly attractive option under these circumstances. The attractiveness of load management systems as an alternative to conventional generation depends upon the pattern of loads on the system. Highly peaked system loads, as -defined by low annual and daily load-factors, may also indicate that load management would be helpful. Organization of Report Chapter 2 characterizes the different systems for load management which are currently available and describes the component equipment costs of these systems. Chapter 3 discusses the applicability of load management systems and provides estimates of the total system equipment costs. Chapter 4 reviews utility and customer considerations in establi- shing the applicability of a load management system. Appendix A identifies vendors of load management equipment. Appendix B presents system cost assumptions for remote control technologies. -5- LOAD MANAGEMENT SYSTEM SUMMARY Each of the load management technologies reviewed in this report is summarized in Table 1-1. The operation of the technology is discussed and basic components identified. Typical end-use applica- tions of the technology are described and suitability of. these applications is evaluated. The technologies listed in this table are divided into three general categories: * Local control and communication systems are based on the local or remote logical control of electricity end uses. * Thermal energy storage systems consume energy from the electric utility during off-peak periods and release this energy to the end use during peaks. * Supplemental energy systems provide alternative sources of energy during times of utility system demand peaks. Table 1-1 DESCRIPTION OF LOAD KANAGENVNT SYSTENS LOAD ISANAGENENT SVSTEMS OPERATIONAL DESCRIPTION BASIC COMPONENTS APPLICATIONS EVALUATION LACAL CONTROL AND C01INUMICATION SYSTENS * Local Controllers - Priority Relay A simple operating deiice designed Complete enit withl To dae. priority re- Effectiveness of the priority to limit tilemaximum electrical - current anxisig lays lave led limited relay Is highly dependent upon tIe demand in a facility fly preventing ievice, experimentation In demand coincidence of the loads two or more high load appliances from - small logic circuit residential applies- that are Interconnected, being used at the same time. - electrical relay. tioas with an elec- tric sryer interlocked with an electric water henater. or air coelij- t doner. Priority relays may also be used in similar industrial Or commercial appli- cations. - Time Controllers Time controllers are electronically A completely Oack4gc4 Timers will cycle The tims controller has* been used driven load control devices to auto- motor dlriven. solid electrical loads such oe many years within recent atically turn on or off external atte,.Or micro- as IIVAC systems. years has leeen equipped wit circuits at a predetermined time, processor based time fighting , PUMPS microprocessor technOlogy in switco (progrannaa~ motors, refrigeration, attempts to increase its le&xi- controller). send water heating. bility. - Demand Limiter ies type of control allows a utility Demand limiters range Coemmonly used in Com- In addition to he utility's ban- or a consumer to limit electrical in size from small mercial and industrial fitig from possible peak load demand at a particular location by microprocessor& to, *ite* and capable reductions, coazaercial and indus- cycling or deferring loads in a pro- large computer con- of controlling IIVAC trial users benefit also becase programmed manner. trol led en,ergy man- systems. thermal stor- this system wilt minimize monthly agement systats withe age systems, space electricity demand charges and sairsk Peripheral baesting. refrigeration, reduce the swichleear capacity re- devices. wirtng is water heating, and quired at tie Installation. usually eitheer Some industrial. pro- twisted pair wire or Ceases. eisting electrical wiring. Table 1-1 (Continued) LOAD MANAGEMENT SYSTEMS OPERATIONAL DESCRIPTION BASIC COMPONENTS APPLICATIONS EVALUATION - Load Management Thermo- This type of thermostat will control A completely packaged The thermostat is Host thermostats preheat or pre- stats an air conditioner or heating system unit with only simple typically ultilized in cool prior to peak conditions. As based on outdoor temperature, indoor wiring involved. 'residential and small such. some level of consumer dis- temperature, time of day, or a signal commercial facilities comfort may be experienced. from a utility. The thermostat con- to control the heat- trols the duty cycle of a heating or ing or cooling system. cooling system by directly control- If interfaced with a ling the temperature settings. '* remote control system, the control of the ter- mostat can be initiated by the utility. t Remote Control Systems Radio Control System Radio systems generally use one or a iCentral control Through a switching Radio systems are widely used and more FM transmitters strategicaatr system operation the radio the technology is well-advanced. situated within the service area to * Transmitter facil- system will cycle or Coat, performance, and reliability transmit coded commands from the ities defer are predictable. The cost per utility to radio receiver switches at * Radio Recever/ - water heaters point controlled tends to be lower customers' service location. The Switch - air conditioners than for other cofmunication and switches are used to remotely turn Package * - electric space load control systems. The flexi- on or off major hppliances or etec- - FM receiver heating systems bility of the radio is limited in tricity consuming devices. As an - Tone or digital - thermal storage terms of receiver service range. alternative to utility operated radio decoder systems Signal propagation is affected by transmitters, sidebands on commercial - Electrical relay - irrigation pumps terrain or man-made objects. stations may be used r switch - commercial refriger- Another limitatc, is that systemy ation. compatibility across vendors is at best minimal. Tse of comer- cil radio transmitter oideband allows lower capital cost and higher power transmission. Ripple Control System Ripple control systems use the elec- a Central control Through a switching The ripple system has been widely tric utility's transmission and dis- system operation, by the used in European countries. The tribution network as a medium for a Transmitter Oo- receiver, the ripple ripple system offers increased transmitting low frequency signals. cated at substation) system will perform flexibility, increased signal Impulses generated by a transmitter, a Injection Equipment similar load control reliability and similar equipment are injected onto normal lirne voltages * Receiver or Tj-ans- functions as the radio reliability as compared to the in preset patterns of commands which ceiver systems radio aytem. However the coat are in turn received and decoudedby * Electrical Relay is generally higher.o ripple control receivers. These- Switchc receivers perform one or more switching functions at the point of control. Tale 1-1 (Continued) LOAD HAHAGEHEENr SYSTEMS OPERATIONAL DESCRIPTION BASIC COHP)NENTS APP11CATON1 EVA.UATION - tin- and hi-directional Tle power line carrier system is a Central control similar load control in a large power line carrier Power Line Carrier similar in principle to the ripple $yat*&a capabilitis as other application involving many trans- system in that both utilize exiting 0 Substal ion hardware remote control eye- Ponders or receivers, tile control- power linies as a tranlamission medium. - transmitter teama; and when used ling economic factor is the coat Nowever, because the power line - iniection equip- in a bidirectional of the point-of-control unit. carrier system utilites higher rs- ment mode. additional sowever, in am4lier application. quency, saignal attentuation is Communications capabilities suct$ such a system might not appear as greater. Power line carrier also ildr4ward as data acquisition. Cost-effective bOCAUAe the a4di- requires less signal power than does amplifiers remote meter reading tional costs of the coupling and ripple to achieve an acceptable lile raps and distribution alt amplifying equipment are more signal-to-noise ratio. For bI- repeaters possible. significant at lower levOld directional purposes, a transponder a Receiver or Trans- is used in place of a receiver, ceivers adding verification and monitoring e Optional hardware capabilities. for remote Peter reading * Electrical Relay switch - Telephone Control System The telephone control system is a utility central Thlis system has gili- Thle telephone control technology commsercilly available in two very control fer load control and lis been under going extensive different structural formats. The a Telephone lanter- bi-disectional cape- changes In an attempt to reduce I first uses a central computer. teld- face (this eqtlip- bilities as the bi- its depenidence on thle telephone phone interface, and an end point ment differs for directional power line company. Earlier systems required device which initiates all comimunica- the two systems) carrier system. communication equipment on tele- tion. The second system has tile same a End Power Receivers phone comany'sa premises. Current component make UP, but communication or Transponder verions of this technology oper- is initiated at central control. ates without any speciat 4sais- tence fcom the telephone company thereby eliminating telephone ytariff control. -casuist Cable Systemt This system uses a coaxial cable a central control May be used for num- but to the high coat Pt a coaxial to communicate to lad management a lleadend equipment erans two way commun- cable infrastructure this system devices at the point of control. a Cable system ication purpose. not is economically feasible only when - trunk cable and Just load management, applied to a exsing cahle system amplifier or Constructed jointly with - feeder cable and smother application such as a amplifier video cable network - above ground hardware and equipment a rep a point of control r Addressable tap Table 1-1 (Continued) LOAD MANAGEMENT SYSTEMS OPERATIONAL DESCRIPTION UASIC COMPONENTS APPLICATIONS EVALUATION - Hybrid Systems A hybrid system is a combination of . Primary system con- Hybrid systems have By taking advantage of superior two or more remote control systems. viata of central the same applications aspects of different systems a One type of hybrid system uses radio control front-end and load control func- hybrid system offers greater flex- to communicate to a PLC system located cosmmunication equip- tions as other remote ibility in both the terrain that close to the point of use. Another ment, and receivers control systems. can be crossed and the concentra- type of hybrid system involves two-way located at distri- tion of end-points that can be communication each way by a different bution points. served. method. S Secondary system(s) involfies interface with primary system and distribution transmitter and point-of-control receiver. a Communication and Informa- tion Systems - Hulti-Building Systems This system groups together the load Computerized infor- When instructed, Successfully operating multi- control efforts of multiple energy mation control operation managers of building systems have reduced users in an attempt to meet a util- center large buildings or utility peak loada by 25%. When ity's peak demand constraints. All Dedicated telephone facilities are able the peak demand level can be ade- users have communications terminals lines to reduce, cycle or quately reduced, utilities benefie from which they communicate back and Remote comsunica- defer lights, esca- because they are now provided with forth with a computerized control tion terminals lators, HVAC systems, a capacity that they could only center over a network of dedicated and thermal storage otherwise achieve by bringing into telephone lines. When the utility systems. play more expensive fuel and approaches their peak demand, the equipment. Building operators center instructs each user to benefit by sharing the risk of un- reduce a certain amount of its timely interruption of service. electrical load. THIERM4AL ENERGY STORAGE SYSTEM * Ceramic Brick Hteat Storage- Room electric storage heaters contain * Room heating unit Room units are most Utilities might achieve ant Room Units a magnsite brick storage core which * Outdoor temperature often found in resi- improvement in the system load is heated by electrical resistance sensor dential homes and factor, but at some level of wire coils at night for best during a Control panel - can apartment buildings, saturation the savings in the following day. be centrally located generation are offset by the for control of need to rebuild and upgrade numerous rooms distribution facilities. ost distribution circuits are capable of handling only a limited number of units. Table 1-1 (Ciantinsiod LOAD HHANACENT SYSTEMH OPERATIONAL DESCIfPTI(N BASic cimpouENs APPLICATIONS EVALUATION a Central Ceramic srick iIest Central storage heaters are similar to a Electric thermal Central ceramic Tie cost of a central ceramic Sturage room units in that ceramic bricks in storage unit heaters are most com- heater in addition to its impact tie core are heated at night for (an- e by-pass aid shut monly found in tie on the distribution system varies forced convection heating of tie entire aff damper section residential sector. directly with the atcrage capacity building the next day. a Night heating sec- The heater can be needed. tion interfaced with a * Coting cabinet remote control system (optionat) for direct control by a Inlet, intermediate, the utility. and untlet plenum sect ions * Thermostatic device a Pressurized Mater Ileat This storage system utilizee a sealed- A cosptptely packaged liManally installed in The pressurized water heat storage Storage package boiler in which electric pressurixed steel moderate-to-large com- system has liad some use in the immersion heaters heat a waterlike tank with electric mercisl buildings. The residential'sector., yet the cost fluid. A simple pipe loop through the elements, tempera- tank can replace the of the system tends to be high boiler retrieves the heat as required.. ture safety control, conventional boiler or for individual houses. pressure control and furnace and may be used relief valve, and with various distribu- drain valve. tion methods. Typi- cally, control is pro- vided by a time switch. I a Ice-Cool Storage The ice-cool storage system utilizes . fce storage tank ltse of time ice-cool The benefit of a cold storage an insulated tank containing coils - immersed storage system In system improves wit an increasing Immersed ift water. In operation, ice plastic tubing found in all aectorst coaling load. In comparison, to is frozen around the coila during oft- least exchamger residential, commeer- those systems using water only as peak hours. During peak houtrs thme - overrite and lebw cil. and Industrial. i storage medium, time ice-cool compressor of the freezing system is limit Flickmastats Control of the system system utilize& a smaller tank for turned off and the stored ice is used a mechanical package is usually provided le Same storage capacity. to provide cooling - evamparatter/hmeat by a time controller. ectanger -antifreeze pump - secontary water circulating pump a Control panel Table 1-1 (Continued) LOAD MANAGEMENT SYSTEMS OPERATIONAL DESCRIPTION BASIC COMPONENTS APPLICATIONS EVALUATION a In-Ground Heat Storage This storage system uses electric * Electric resistance Only applicable to The building must be well-insula- resistance mats laid beneath the mats one-floor buildings. ted and have proper ground cond- concrete floor of a building. The o Thermostat sensing Host of its use has itions otherwise the system is concrete and the earth beneath are bulb * been in the comunercial ineffective. The floor must be heated during off-peak hours. With And industrial sectors. covered with a dense material the system turned off, the stored The system can be (such as asphalt or tile) for heat keeps the building warm luring controlled by a therm- maximum heat transfer. the peak period. ostat, time controller, or a demand limiter. * Combined Heat and Cool Storage - Annual Cycle Energy The annual cycle energy system pumps * high efficiency A prototype system Since the ACES essentially pre- Storage (ACES) heat from a water reservoir in the water-to-air heat integrating space con- cludes the need for registance winter, gradually freezing some 80% pump ditioning and domestic heating (space and water) im the of the water. In the summer, the ice 9 Domestic hot water water heating systems. winter and can provide cooling is used for air conditioning. Dom- storage tank The system has been without the use of a compressor estic water heating is also provided. a Large, insulated, installed mostly in system in the summer, tile peak Water is heated via a water loop water storage residential and small daily and seasonal demand reduc- through a desuperheater in the heat reservoir with commercial buildings. tions brought about by an ACES pump assembly. imersed ice freez- could be substantial. However, coilIs. because of tle large storage reservoir required, the system might not be as practical for larger applicationL. - Daily Cycle Energy The daily cycle energy system was * Normal heat pump Space conditioning can Essentially, the DCES is all ice Storage (DCES) designed to charge and store beat system be provided for all coot storage system equipped with and/or cool on a daily basis. A well- o Thermal storage types of applications; a heat ing mode to provide space insulated storage tank is used for tank residenital, commercial, heating in tile winter. The com- storage of both non-preasurized hot e Prewired control and indutrial. bined thermal and cooling storage water in the winter and cool water panel systems may provide substantially and/or ice in the summser. reduced local daily peak loads during both surter and winter seasons. Table 1-1 lContinued) LOAD HANAGEHENt SYSTEHS OPERATIONAL DESCRIPTION AIC COHPONENTS APPLICATIONS EVALUATION SUPPLEMENTARY ENEGV SYSTEHS 0 Solar Energy Systems Solar systems utilise the sun's energy 0 Flat-plipte collec- Solar systems have The potential for solar energy in a active or passive manner for space tors teen installed in res- systems may appeat enormous but conditioning and domestic water hat- * Evacuated tube col- ideotial commercial. the developeent and use of solar Ing. The collection of solar energy lectors and industrial facil- technology has been plagued with for active use is most often performed a Concentrating col- ities usually to problems. Theae problems involve with flat plate, evacuated tube and lectors replace or supplement inexact engineering. excessively concentrating collectors. The most s Energy-storage existing heating complex system design. improper comason type of solar heating is a system systems aixing of coomponents. and ineffi- flat-plate warm water system. Solar a Energy-conversion cient backup equipment. cooling can be performed in three system ways& night radiation, compressive chilling, and absorptive chilling. * Cogeneration System$ Cogeneration systead involve tie e Topping system cogeneration systems Since a utility could acquire simultaneous production of electric prime moveral g are utilized in indus- electricity from an ipdustrial or and thermal energy and can be utilized turbine, steam trial and comuercial commercial cogeneration plant. to meet a facility's on-site process turbies, and diesel applications such as generating capacity.can be added needs. There are two types of cogen- engines dual-purpose power to thf utility with minimal or eration systems (topping cycle or * Bottoming system plants, some waste hoat taro capital Investment and in bottoming cycle) differentiated by prime movers utilization systems. sea time khan it would take to whether electric or thermal energy Organic Raaking and sums district heating construct a conveational or is produced first. low pressure systems, apace haating nuclear power plant. Also, a steam turbines and cooling, and total utility would have Increased energy systems. In- system reliability because ss dustries with flat dispersed power generation. ur actricl aeds and high thermal loadst may eis cogeneration to reduce demand on the utility during system peaks. a puat Ileating Systems Dual heating systems combine a fosail- a Fossil-fired forced A dual heating system A dual heating system is of parti- fired forced air beating system with air heating system io applicable to culr interest to those utilities an electric heating system. Heat Is a Electric heating al cast any type of with an lready high daily load provided by the electric heating aye- systems facility utilizing en factor where thaermal storage eye- tan until the utility reaches their at Rem4te control - electric or fossil- tows would be of little value. peak. When the peak period approaclaes. system or a local fired heating syatem Since talst system requires no of- a remote Control system or local con- controller peak chargipg excessive distribu- troller will switcha the heating lad d Ductwork system sion system loading and t he to the fostil-fired system. (optional) creation of "(lgo" peaks caused by charging cre eliminated. -13- Chapter 2 LOAD MANAGEMENT TECHNOLOGIES This chapter introduces the technologies which are currently available for electric load management and offers some insights to their design, operational and cost characteristics. This information is provided in two sections: * The first section summarizes the generic types of load management systems for which operational experience is available. * The second section provides data about the equipment costs of the various load management systems. Appendix A reviews the load management hardware, software, and systems offered by specific vendors. Although load management technology is advancing rapidly, fore- casts of technological development are beyond the scope of this effort. The following sections focus primarily on technologies which have evolved beyond the experimental stage and are now commercially available. While some innovative applications or modifications of existing load management hardware or software may be required to meet the specific needs of utilities and electricity users in developing nations, it is anticipated that utility systems in developing nations will not desire to experiment with previously untested equipment. The most -effective load management technologies tend to be those which provide the utility with remote control capability and those which control large commercial and industrial loads. Systems offering remote control or communication capability provide the greatest flexi- bility and assurance of coincident load reductions. Commercial and industrial applications usually contribute the greatest amount of load reduction per control point. The cost data provided in-this chapter are provided as a general reflection of the relative costs of the various technologies and their components. These data are estimates of representative costs based on contacts with manufacturers' representatives, available marketing literature, and a review of other load management equipment surveys. Cost variations among vendors may be significant and changing market conditions may require that these cost data be periodically updated. Costs for many of these systems, particularly those relying on solid-state electronic equipment, are likely to decline over time. Such cost decreases may result from three key factors. These are: * Increased competition in a growth market; -14- * Further technological advances; and * Improved production cost efficiency as the number and size of orders for load management equipment increase. Radio control equipment vendors, for example, frequently offer signi- ficant discounts to purchasers of large quantities of control switches. Costs for the various load management systems discussed in this chapter are not presented in, terms of dollars per kilowatt of control- led load because this measure is highly sensitive to the specifics of each apptication. The type(s) of load controlled, the coincidence of customer loads, the utility's load control strategy, and specific hardware and software configurations employed, all have direct influence on the costs per kilowatt of a load management technology. TYPES OF LOAD MANAGEMENT TECHNOLOGIES The number of load management systems for which operational experience is currently available is quite large. These systems represent a wide range of technological approaches to demand-side load management. They range from simple mechanical, time-controlled, devices to highly complex and sophisticated solid state electronic equipment - which interfaces through bi-directional communication systems. Load management systems may also incorporate-technologies for solar energy, energy storage, and cogeneration. The technologies for electric load management divide into three general categories which will be used as the basis for further discussion. These categories are as follows: * local control and communication systems, * thermal energy storage systems, and * supplemental energy systems. Some load management systems include elements of technology from two or more of these categories. Clearly, several of the available tech- nologies have the potential to play complementary roles within a single load management system. However, our discussions in this section will only address overlaps between these categories to the extent they reflect system designs for which actual operational experience is currently available. Table 2-1 outlines the major categories of load management technology alternatives. -15- Local Control and Communication Systems Developments in solid state electronics over recent years have led to considerable expansion of the number of load control and commu- nication systems which may serve electric load management functions. Generally, the least complex and sophisticated of these load control and communication systems are classified as local controllers. Remote control systems represent an intermediate level of sophistication and complexity., while communication and information systems with the capability of bi-directional or multi-diredtional communication tend to reflect the highest level of technological sophistication currently available. Local controllers include such devices as simple mechanical or electro-mechanical time-switches, priority relay devices, demand limiters, and thermostat controls. Each of these technologies operates in isolation and'generally interfaces with a specific end-use or a combination of end-uses within an individual customer facility. They can be highly effective in limiting individual customer peak requirements and customer requirements during prespecified hours of the day. They are not easily adaptable, however, to changes in the timing of utility requirements for load control. Their functions may also at times be overriden or subverted by customers who find their operation inconvenient or bothersome. Priority relays -and time controllers are most effective when applied to major electricity consuming devices. -Demand limiters may be applied either to indivi- dual end-uses which have variable electricity input requirements or to the overall electricity requirements of an -individual facility or home. Applications of thermostat controls are generally limited to electric air conditioning,-space heating and water heating systems. Remote control systems provide a utility with much greater ability to coordinate the timing and in some cases the magnitude of load control activities with utility system requirements. This class of technologies is .further subdivided on the basis of the communica- tions technology which is employed. Four types of communication technology alternatives exist to support these systems. These include: radio communication, telephone communication, power-line carriers and ripple controls. Ripple controls and powerline carrier systems utilize the electrical system itself to communicate load control signals. Hybrid systems which employ two or more of these communications technologies have also been tested. An example might be a system which employs radio communication technology to transmit a primary signal from a central location to a number of dispersed locations which then transmit a secondary signal via the electrical system (ripple or power line carrier) to control points for specific end-uses. A key concern of all such remote control systems is signal reliability. Each type of system has its own strengths and weaknesses in this regard. -16- Communications and information systems represent state-of-the-art technology. They are distinguished from remote control systems in that they provide for two-way communication between the utility and remote points of control. These systems can provide remote control capability and real-time verification and measurement of load reduc- tions. The most sophisticated of these systems are designed to collect, process, and analyze customer use data to assist the planning and implementation of load management activities on a real-time basis. Some of these systems may serve remote meter reading functions and may also be used in the future for implementing real-time pricing schemes-for large industrial and commercial customers. Two types of bi-directional communication system technology have demonstrated commercial applicability.. These are telephone and power line carrier systems. At present, telephone systems tend to provide greater operating flexibility and higher quality data transmission. Power line carrier load management systems, however, offer the poten- tial for direct interface with existing or planned central load dispatch and network control systems. Also, power line carriers are often preferred by utility system personnel because their operations are wholly within the control of the electric utility. Rowever, a number of existing bi-directional systems employ a combination of these technologies. Thermal Energy Storage Systems Load management through thermal energy storage focuses primarily on the management of loads associated with space conditioning and water heating requirements. In their most basic forms, these systems can provide load management benefits while operating on a stand-alone basis. Greater load management benefits from thermal energy storage are generally obtained by integrating these systems with either local controllers or remote control systems. Technologies for energy storage include ceramic, pressurized water, and in-ground heat storage, as well as cool storage using ice. Combinations of these technologies may also be employed in annual cycle and daily cycle energy systems. The design of an energy storage system is also generally such that its load management benefits are spread somewhat uniformly across the days or months or seasons. Without the use of remote control technology, it is generally not possible to focus load management activities to just those types of annual peak requirements for the electric utility system. Thermal energy storage systems must be carefully planned to avoid operating and installation pitfalls. High saturations of heat storage systems may produce greatly peaked demand requirements during previous off-peak periods. Thermal energy storage systems also tend to require significant physical space. This makes retrofit to an existing facility more difficult and costly. -17- Supplemental Energy Systems The third major category of load management technologies includes those systems which are designed to level electricity requirements by supplementing electricity use requirements with energy from other sources. Three types of technologies fall into this category: solar energy systems, dual heating systems, and cogeneration systems. Of these, dual heating and solar energy systems primarily satisfy heating requirements. Cogeneration systems may serve as either supply-side or demand- side technology. When cogeneration systems are operated in parallel and not connected to the utility system grid, their operation is generally considered to be demand limiting. The extent of their load management capability is measured by the degree to which their opera- tions serve to reduce utility system daily and annual peak electrical requirements. Solar energy technologies often involve some form of thermal energy storage and may be either active or passive in design. Solar photovoltaics (i.e., production of electrical power directly from solar energy) is generally still considered an experimental technology. Dual heating systems, on the other hand, represent a well-tested technology. It typically involves the substitution of fossil fuel for electric power during periods of electric' utility peak requirements. To be most effective, these systems once again require some form of local controll-er or remote control interface with the electric utility to ensure coordination of load management activities with utility requirements. COMPONENT COSTS OF LOAD MANAGEMENT SYSTEMS This section describes the magnitude of equipment costs for the various load management systems and their sensitivity to key para- meters. These cost estimates are intended as a general illustration of magnitude, trends and relationships. The components, design and cost of an individual system will vary with the specifics of the equipment, functions performed, and applications. Costs and capabili- ties are constantly changing as new products enter the market-place, as production costs change (especially for computerized central controls) and as production runs increase. Cost experience for several systems is derived from small-scale experimental installa- tions. The cost of large-scale commercial systems may be quite different in scale, component design, overall configuration and capabilities. Equipment costs present only part of the cost picture. Installa- tion costs may add significantly to capital expenditures. In addi- tion, operating and maintenance (O&M) costs may vary over the life -18- of the load management program. Both installation and O&M costs frequently depend on local labor costs. Moreover, delivered costs of equipment may vary in different parts of the world, as will the amount of support available from manufacturers of load management equipment. Local Controllers All of the equipment for local controllers is at the point of control. Priority Relays link a nonessential load (such as a water heater) to a priority or essential load (such as a pump). This interlock acts to switch off the nonessential load whenever the priority load is turned on, thereby reducing local peak demands. The cost of this device is $140 with an installation requirement of approximately 1/2 manhours (electrical technician). Time Controllers may be placed into two categories - mechanical clocks and microprocessor programmable clocks. Mechanical clocks usually have the capability to control only one circuit, yet their functional capabilities vary widely. They may be used, for instance, solely for time-of-use metering or for more direct control functions. The basic mechanical unit is a 24-hour clock that controls one pair of on/off operations. The advanced mechanical units offer a.seven-day clock with the capability to control multiple on/off operations and other options such as back-up power systems and an on/off pulse function. The mechanical units are self contained and vary in cost between $10 and $580.. (See Table 2-2.) The microprocessor programmable clock offers expanded capability of controlling up to sixteen circuits with multiple on/off opera- tions. The minimum microprocessor-based control unit provides one-circuit control with a 24-hour clock with eight on/off operations, and a skip-a-day function which suspends operations for weekends and holidays. This unit is self-contained and costs $100. The more powerful units 'control between 4 and 16 circuits and operate on a 365- day clock that allows programming of weekend and holiday schedules. Most advanced microprocessors offer back-up power systems and a safety option which gives the user the ability to set circuits to on or off in case of total system failure. The multiple circuit microprocessor control units range in cost between $250 and $3,800. Demand Limiters vary in cost and capability primarily by the number of circuits that are controlled and by the complexity of func- tions performed. This is shown in Table 2-3. Load Management Thermostats may be obtained with three levels of capabi.lity: preheating/cooling wall systems, optimal setting systems and optional setback systems with setback and restart capability. Costs for these three types of systems are shown in Table 2-4. -19- Programmable Controllers. The most significant development in the local control area in recent years has been the emergence of the "programmable controller," an application of which has been described earlier as a "microprocessor programmable clock." Rapid reductions in the price of microprocessors have allowed their use in a wide variety of applications, such as the management of commercial building and industrial process control. Energy management systems based on programmable controllers have proliferated and are currently offered by scores of vendors worldwide. Load management is clearly within the capabilities of these systems. Depending upon configuration and programming, they may perform any combination of the functions of any of the other local controllers. The central component of programmable controllers is the micro- processor and associated memory. This is connected to the controlled loads, other controlled devices and systems and to a programming unit. The programming unit, at its simplest, is a control panel which allows a limited number of options to be selected. More complex systems involve computer terminals and peripherals. A number of building energy management systems have even been developed using inexpensive personal computers. One capability, not yet exploited by utilities, is communications between customer programmable controllers and utility load management control systems, which may use any one of a variety of techniques (see below). This may afford both the utility and the customer with flexibility. The microprocessor chip itself- may cost as little as $10.00 (U.S.). Associated electronics and a programming interface unit may add several hundred dollars to' the cost. Switches, relays, ther- mostats and other sensors add further to the cost depending upon the installation. Remote Control Systems The components of remote control load management systems may be placed into up to three broad categories based upon their physical location and function. Central control components are located at a central facility, frequently (but not necessarily) operated by the utility. They perform operational 'decision-making and coordination functions. Only remote control load management technologies have central control components. Communications components provide the link between the central control component and the point of control. They may be located anywhere between the central control facility and the point of control. Point of control components are physically located on the premises of the load management program participants and may perform a wide variety of functions. Point of control is defined as the premises on which load management takes place, not the piece of equipment which is controlled. Each of these categories is discussed below. The prices and capabilities of specific technologies within each category, based on late 1982 contacts with vendors, are shown in Tables 2-5 through 2-10. -20- Central Control. Remote control systems centralize the control of the load management systems at a single point operated by the electric utility. In each of the remote control systems, the basic equipment used for central control is a computer coupled with equipment for electronic signaling. Typically, there is one point of microcomputer central control. Sometimes, however, the central control function may be dispersed in a set of microprocessors located at various points throughout the -transmission and distribution network, often coordi- nated by the central computer. Typically, there are two basic levels of sophistication (and therefore cost) to the microcomputer system. At the lowest level, much of the software for the operation is "hardwired" with relatively little opportunity for operator intervention. The microprocessor makes the key decisions. More sophisticated systems allow the operator to sit at a terminal, usually with a Cathode Ray Tube (CRT) display, in order to intervene actively. Equipment costs of these systems appear to depend primarily on the brand of microcomputer incorporated into the load management system and the precise equipment configuration. Software development may contribute substantially to the cost of the computer, -especially if significant effort is needed to adapt software to -the specific needs of the purchasing utility. Communications. The type and cost of equipment for communications depends primarily on the medium chosen for communications, the techno- logy and the amount of power required. Communications equipment injects load management signals into the appropriate media and, in some cases, carries the signals to its destinations. Four media have been used for load management communications: electrical transmission and distribution system, radio transmissions, telephone lines, and coaxial cable. Different communications equipment may be used in combination. This would occur when radio, telephone or coaxial cable is used to communicate between the central control microcomputer and a trans- mission transformer or distribution substation, and the transmission and/or distribution system used as the medium to link the substation to the point-of-control. Point of Control. Equipment on the premises of the facility to be controlled typically consists of a receiver for the signal and swit- ching mechanism. Frequently, receiver and switching mechanism are integrated into one package. Costs of point-of-control equipment may -21- vary widely with the technology, directionality (one-way versus two-way) and number of functions performed (e.g., number of circuits controlled.) Point-of-control equipment is based on a tone system or a micro- processor-controlled digital system. Until -recently, the less sophis- ticated tone systems enjoyed a significant cost advantage over the digital systems, which allow greater addressing capability and more flexibility to handle different functlions. Over the last two years, however, the cost of digital -systems has been dramatically reduced by the economies of large production runs and by competition. The cost of tone systems has not similarly dropped because the mechanical devices used to create the tones still require substantial fabrication effort. Currently, digital equipment is priced only very slightly (1-5%) above tone equipment and many vendors are phasing out their tone-signal equipment. Two-way systems add substantially to the cost of point of control equipment because additional components are required to take measure- ments and develop and transmit signals. Usually, two-way load manage- ment systems are used for other functions as well (e.g., load research, automatic meter reading). Adding these other functions helps to offset the additional cost of the two-way capability. The most basic point-of-use equipment for any of the systems will perform a switching activity on one circuit. Adding additional circuits increases costs, but generally no more than 5-15 percent of the- single circuit- equipment. The basic signal receiver remains essentially the same as does the equipment packaging. Generally, all that is added is an-additional switching relay. Installation costs (not considered in this report except for the supplemental energy systems) may be a substantial portion of the total cost of point-of-use equipment. Depending on the type of technology, substantial labor relative to the value of the equipment may be required for installation. Both labor and equipment costs may sometimes also be increased if the receiver for some reason cannot be located physically proximate to the equipment to be controlled. Most vendors of local management equipment offer substantial discounts for large purchases of point-of-control equipment, often in the range of 5 to 40 percent. Large orders allow the vendors to optimize and plan for large production runs. The exact pattern of price breaks depends upon the vendor (and, no doubt, the competitive situation), but generally, the largest price break occurs for purchases larger than about 10,000 units. Communications and Information Systems There are two communications and information systems which may be used for load management purposes: Systems Control and Data Acquisi- tion (SCADA) Systems and Multi-Building Systems. -22- SCADA. Many utilities already have or are planning Supervisory Control and Data Aquisition (SCADA) systems. SCADA systems allow electric utilities to remotely monitor and control their system operations. The difference between SCADA and remote-control load management is that the functions performed by SCADA systems are usually on the "supply side", while load management functions are on the "customer side". SCADA systems comnunicate commands from a central control point to remotely'located points of control. Status or performance information may then be returned from the remotely located points. SCADA communi- cations may use the same types of media as load management communica- tions (e.g., radiowave, telephone lines, power lines). Typically, digital signals are transmitted, but some SCADA systems. also transmit analog information. Remotely-collected data may be used in a real-time mode for system control or may be recorded for later analysis or forecasting. Similarities in tecinologies for both SCADA and load.management imply that the separate development of these two sysctms for the same utility may be duplicative. The integration or coordination of SCADA and load management will frequently increase the cost-effectiveness of both systems. If both are-of equal priority, duplication of effort may be avoided; if one is more important, utilization of its equipment may be- improved. Care, however, must be taken in integrating these two systems. There may be conflicting prioritios in system design or operation which must be resolved. More expensive equipment may also be required. Sometimes, for example, the increased complexity and number of functions mean that a single microcomputer is no longer a:dequate. A significantly larger-and more expensive-computer may be required (as may be redundancy). The added benefit of SCADA should be compared to these costs or the added benefit of Load management, if SCADA is of primary importance. Several load management system vendors offer the equipment required for a SCADA system (especially for the control of capacitor banks and transformers). Sometimes, however, distribution control equipment offered may only involve one-way communications. Multi-Building Systems. A multi-building system is, in essence, a miniature version of a utility's two-way remote load control system. It has central control, communications and point-of-control components of its own. A "central" microcomputer monitor's energy use in the controlled buildings and. sends control signals to the buildings at appropriate times. The microprocessor is connected by a communica- tions link to the utility's own control center. In the United States, where these systems originated, communication has been over dedicated telephone lines. There is, however, no reason why such communication has to be restricted to this medium. The message is .sent to a compu- ter cerminal at the point of control where the building operator -23- decides how to manually curtail load. (There is, however, no techni- cal reason why some loads could not be centrally controlled). The cost of the central computer is about $10,000-$50,000). Terminals at each point of end use may cost $200-500. Typically, 5 to 50 indepen- dent facilities are controlled. Much of the capital cost, however, is not equipment cost but the development of energy management activities and of the complex software required to monitor and signal the unique load of the specific multi-business system. The total investment may be a multiple of the equipment cost. Thermal Storage Systems A number of different thermal energy storage systems may be used for load management. Electric storage space heating is probably the most widespread load management practice in Europe. Electric utilities in Austria, Belgium, France, Great Britain, Germany and Switzerland all use these systems to shift electricity consumption during daily winter peaks to off-peak hours. Great Britain and Germany, the largest users, have over 20,000 MW each of controllable electric storage heating load. Most British systems are controlled by time clocks and thermostats while the German systems are controlled a combination of time clocks and ripple systems. This combination of control systems allows recharge time to -be distributed evenly through the daily demand valley. Most units are ceramic heat storage units, but in-ground heat storage units (and water -heaters) are also used. System costs for both European-style and recently-developed systems are discussed below: Ceramic Heat Storage--Room Units. There are three major colapo- nents of ceramic heat storage-room units: * Room heater units -- Heat storage bricks, blower, and regional heaters Units cost ianging in kWh storage capacity from 20-40 (power input 2 to 6 KW) $330 to $820 * Control Panels - Central control panel includes all time devices and unit control costs vary by number of units control capability (1-25) Costs $140 - $450 * -- Alternatively, a receiver for a utility remote control system may be installed to give the utility direct control over the load. (See Tables 2-5 through 2-10 for costs.) * Thermostats - Room thermostats - costs $15 - $25. Vary by precision of calibration and reliability of control. * Total equipment costs range from $495 to $1,300 -24- Ceramic Heat Storage--Central Units. These may be used for either residential or commercial applications: * Residential units include a central core of cast iron or refractory bricks, resistant heaters, blowers, and controls. Costs vary by heating capacity Capacity Cost 14 KY/90 kWh $1,900 30 KW/200 kWh $2,700 * Commercial units are essentially the same as residential, but with higher heating capacities and costs. Capacity Cost 21 KW/140 kwh $2,600 30 KW/200 kWh $2,750 Pressurized Water Heater. These units are primarily stand-alone with pressurized water tank heat exchange .coils and controls. Costs vary by size of tank. * Residential (250 gal) - $3,500 * Average Comercial (5,000 - 10,000 gallons) - $20,000 - $30,000 * Industrial (17,000 gal) - $40,000 - $55,000 Above 17,000 gallons requires multiple tanks Ice Cool Storage. Units primarily consist of cool storage tanks and controls. Costs and mechanical control package vary by size of cool storage. * Residential (36 ton-hour)1 $3,525 * Commercial and Industrial (480 - 540 metric ton-hour) $5,800 to $33,500 Costs do not include duct work, which varies by number of output points and by whether it is a retrofit or new installation. 1/ 1 ton-hour equals 1 stored ton of energy capable of cooling for one hour. One ton of cooling equals 12,000 Btus. A 36 ton-hour ice storage unit requires about 300 gallons of storage capacity. -_25- In-Ground Heat Storage. This system requires a solid ground foundation and concrete slab over heating mats. Costs vary by number and size of mats required (which is a function of building size and the level of heat loss). Costs for mats and all controls and thermostats is approximately $3.28 per square meter. Annual Cycle Energy Systems (ACES). The concept for this system was initially developed as an energy conservation system; however, because of its ability to. reduce weather-sensitive electrical demand, it has demonstrated significant potential as an energy management system. Basically the ACES utilizes a water-to-air heat pump and a water-filled thermal storage tank. During the winter months the heat pump is used to draw heat from the thermal water storage tank, thereby freezing the water in the thermal water reservoir. This reservoir is kept at a maximum of 80 percent ice through the use of solar collectors. During the suimmer months the fro.zen water is used as a cool storage for air conditioning system. Because of the experimental nature of this technology no hard cost data are available, but experimental equipment costs are 600 percent greater than systems using an electrical water heater and a standard air-to-air heat pump. Daily Cycle Energy Systems (DCES). This is an experimental energy management technology developed to improve daily load factors. - The system is designed to perform all the heating and cooling activities. during the off-peak periods. The DCES utilizes a air-to-air heat pump and a thermal storage tank for off-peak water heating. As with the ACES, this technology is highly experimental and commercial price quotes are unreliable. Cost estimates of complete DCES range from $4,160 to $23,100, respectively for 48-540 metric ton-hour systems. Active Solar Energy Systems Solar systems can vary widely. A great variety of equipment components have been developed which can be combined in many different ways to meet varying energy requirements in specific climates. Active solar systems require mechanical assistance for heat transfer from collector to storage to living space. Systems include a collector, storage and distribution components. The collector allows liquid (usually water) or air to be heated by sunlight. This heated transfer medium is then moved by gravity or a pump to a storage facility. When needed, the heat is transferred from -,storage to the living space. If a non-potable liquid is used as the transfer or storage medium, a heat exchanger will also be used. If solar space cooling is included in the system, an absorption chiller will be needed. -26- Solar collectors and photovoltaic cells are generally modular, without economies of scale in collector array design. Overall system cost is dominated by collector cost. Cost of delivered solar energy typically depends on temperature required and timing of energy demands. Active solar systems can add 20 to 30 percent to the cost of a medium-sized residential dwelling and achieve a 40 to 60 percent reduction in heating bills. (See Tables 2-11 and 2-12.) Passive Solar Energy Systems A passive solar system is based on an architectural design that considers the energy conservation value and thermal characteristics of building construction elements, and the interaction of these elements with the natural environment. The- passive solar heating system is part of the building and consists of no mechanical components. This system has a lower initial capital cost and less ongoing maintenafice than an active solar system requires. Passive solar systems include: * direct system - building is directly heated by sun, with sdlar heat -stored in the mass of the house; -* mass wall system - thermal storage mass covered with glass collects. and stores direct solar heat and transfers it to the building space; * attached sun space - greenhouse or atrium collects and stores solar heat. Conditioned living space draws heat from this space as needed. The windows or other collection medium of a passive solar system are generally south-facing and double or triple glazed to prevent heat loss. Additional reflective devices are sometimes used to concentrate solar radiation. Passive solar systems can increase the cost of a medium-size house by ten to fifteen percent and achieve a twenty to forty percent reduction in heating costs. (See Tables 2-12 and 2-13.) Cogeneration Systems Any heat engine can be combined with a waste heat recovery boiler to create a cogeneration system. The most important, and by far the most costly, component of a cogeneration system is the prime mover, the equipment or heat engine that converts the energy content of fuel to mechanical shaft energy. The mechanical energy is then used either to drive a generator and produce electricity, or directly as mechanical shaft horsepower. Other components of a cogeneration system include the thermal distribution system; electrical switchgear -27- and paralleling equipment; supplementary boilers (if needed); fuel shortage or pipeline interconnection; and controls and performance monitoring equipment. The size of the capital investment depends on the design and size of the cogeneration plant. Generally, cogeneration plants fired by more-difficult-to-use solid fuels (e.g., solid waste, biomass) involve a greater capital investment than do plants fired-by liquid fuels (oil or gas): TYPICAL RANGE OF CAPITAL COSTS FOR COGENERATION PLANTS Fuel Cost/KW Capacity Municipal Solid Waste $1,8OO - 3,000 Biomass 900 - 1,500 Oil or Gas 500 - 900 The higher capital cost of the equipment needed for a plant fired by solid fuels may, however, be offset by lower fuel costs, Economies of scale can often be realized in a cogeneration plant; larger plants cost less to build per unit of capacity than do smaller plants. For example, a 10 MW natural gas-fired plant typically costs about $9 million, or $900/KW, while a larger plant of 50 MW requires an investment of about $27.5 million, or $550/KW. Dual Heating Systems Dual heating systems are comprised of an electrical resistance central heating unit supplemented by a fossil-fuel system. Depending upon the relative energy source economics, the electric heating systems cab be operated normally. At the time of the utility's peak, the fossil-fuel system is activated, either manually or remotely by a utility communication system. Capital costs for retrofitting an existing fossil-fuel or electric heating system for dual use are estimated below: Dual Heating System Existing System Retrofit Cost Range Oil or Gas Furnace $750 - $1,500 Electric Furnace $1,200 - $2,600 -28-- Table 2-1 LOAD MANAGEMENT EQUIPMENT AND SYSTEMS I. Load Control and Communication Systems * Local Controllers - Priority relays - Time controllers - Demand limiters - Load management thermostats -- Programmable controllers * Remote Control Systems - Ripple systems - Power line carrier systems - tadio systems - Telephone systems -- Coaxial cable systems - Hybrid systems * Communication Information Systems - Multi-building systems - SCADA II. Thermal Energy Storage Systems * Ceramic Heat Storage - Room units - Central units * Pressurized Water Heat Storage * Ice Cool Storage * In-ground Heat Storage * Combined Heat and Cool Storage Systems - Annual cycle energy systems - Daily cycle energy systems III. Supplemental Energy System * Solar Energy Systems - Active systems - Passive Systems * Cogeneration Systems * Dual Heating Systems -29- Table 2-2 TIME CONTROLLERS DESCRIPTION 7-day Solid State or 24-hour Mechanical 7-day Solid State Microprocessor-Based 365-day Microprocessor Clock or Mechanical Clock Clock Based Clock o No back-up power e Back-up power * Back-up power * Manual override system system system * 1 on/off operation * Multiple on/off * Multiple on/off a Multiple on/off operations operations operations * Manual override * Back-up power system a Automatic power failure safety setting * Duty cycling * Optional add-ons - LM thermostats - Demand limiter COSTS $10 - $50 t75 - $200 t $250 - $1,200 $1,500 - $3,500 REASON FOR * Time interval pre- * Type of back-up * Type of clock . * Optional add-ons COST cision VARIATION a Internal/external * Skip-a-day function e Precision of time a Number of on/off interval operations * Tie to load * Pulse function * Number of on/off. * Manufacturer operations * Time interval * Type of back-up * Expandability precision * Internal/external e Sophistication of * Sophistication of programmability programmability * Number of on/off * Expandability # Number of circuits operations - * Number of circuits * Manufacturer -30- Table 2-3 ELUCTRO-MEC2AMICAL DEMAND LIXITERS CAPABILITY Low Intermediate Bigh DESCRIPTION * 2-8 Circuit * 10 Circuit e 20 Circuit Control Control Controt * Continuous * Continuous * Time Control- Demand onitor Demad onitor lar Micropro- With Sequential with Duty ceassor Based Load Shedding Cycling and Se-. quencial Load e Duty Cycling Shedding * Optional Cam- maitation Device COSTS $180 - *350 S800 - 91,700 1.800 - S2,500 REASON Fo& * Number of * Sophistication * Sophistication RANCZ Of Circuit Controls of Timing of Timing COST Device Device * Programmabil- * Level of Pro- ity grammability a Manufacturer * Optional Add- . one (coimuni- cations) NOTE: Demand limiters are primary offered as a qomponent to a larger system. Incremental component costs range from $350 --41,200 depending upon the sophistication and size of the base system. Table 2-4 LOAD MANAGEMENT TERMOSTATS Pre-Reating Optimal Start/ and Cooting Temperature Set CAPABILITT Wall Sstem optimal Start Back Reset DESCRIPTION * Preheats/ e Outdoor Temp- p Optimal Start of Cools Based on eracure Sesi-. Heacing and Cool- Time Schedule tive ing * Wall Thermo- * Controls Heac- * Outdoor Tempera- scats ing CooLing and ture Sensitive Compensate for Outdoor Temp- erature * Controls Water Reating Systems * Sec Back Tempers- ture Based on Time Schedule * Duty Cycling COSTS $60 $00 $1,200 -31- Table 2-5 RIPPLE SYSTEM COMPONENT COSTS (1982 U.S. *) Component Capabilities Costs Number/System CENTRAL CONTROL * Computer * time'and develop control instruc- One may be located at utility tions to communications system control center or at each sub- (if 2-way system, receive and pro- station. cess return information) - Clock Programming $10,000 - $25,000 - Demand Proportional Control S20,000 - $60,000 COMMUNICATIONS * Signal e injects low frequency command $ 2,000 - $10,000 one per injection point1 Generator signals into power bus. Generates (cost depends on power) signal at 0.1% of peak signal power * Coupling * injects signal into circuit t 1,000 - t2,000/MA one per injection point1 Circuit * Remote * receives messages from computer for $ 4,000 - $ 7,000 one per injection point1 Control Unit injection by coupling circuit (when computer is at control center) * Substation * included in 2-way systems, $5,000 one per pubstation Transceiver 4-function POINT OF CONTROL * Receiver/ ** receives signals, performs switching one per point of control Switch functions - 1-function $ 80 - $ 902 - 4-function $140 - $1802 * Tranoceiver/ * included in 2-way systems, $4,000 one per point of control Switch 4-function 1/ Injection points are frequently at distribution substations, but may be at higher voltage levels depending on the design of the ripple system. 2/ Volume discounts generally available. -32- Table 2-6 POWER LINE CARRIER COMPONENT COSTS (1982 U.s. t) Comoonent Cavabilities Costs Number/Svstem CENTRAL CONTROL * Computer a originates all control functions, I per system (2 for added receives and processes all opera- reliability) tional and measurement data. (Cost * varies with size of system and : sopbisticacion of software.) - 1-way $15,000 - $50,000 - 2-way $50,000 - $400.000 COMMUNICATIONS * Substation 0 receives signals from central I per substation Control Unit computer, receives signals for one transponders, retransmits to central compcer (if 2-way) - 1-way $ 5.000 - 2-way $15,000 e Discribu- control and monitor electric utility $ 700 - 5,000 1 per Substation tion Auto- distribution circuits maic Units (optional) @ Line Traps * traps insure signal propagation $ 400 -- $ 500 (If needed) one per capacitor (not always through capacitors and transformers (Volume Discounts) bank and transformer needed) e Repeaters * repeaters needed to boost signals _$ 2,000 *3,000 Generally, (if needed) (not always several kilometers from substation One every 5 to 15 miles but needed) some may be used for "dead" points in system. POINT OF CONTROL * Receiver/ * Receives signal sets switches Switch - I function - 90 - $110 1 per point of control - 4 function $120 - $160 e Transceiver/ * Required for 2-way system Switch - I function $170 - $210 1 per point of control - f function $340 - $300 -33- Table 2-7 RADIO (DIRECT) COMPONENT COSTS (1982 U.S. $) Component Capabilities Costs Number/System CENTRAL CONTROL * Control * originates load management commands Units - manual $10,000 - $15,000 1 (2 for added reliability) - computer $15,000 - $30,000 COMMUNICATIONS * Transmitter a broadcasts signal (cost depends on $ 7,000 - $10,000 depends on area to be signal strength) covered and geographyl * Repeater * rebroadcasts to distant or inac- $11,000 - $13,000 depends on area to bel cessible areas. (Same equipment as covered and geography transmitter plus receiver.) * Return a In two-way system, receives data $500 at least 1 per transmitter or Transceivers from points of control and re- repeater, depending on strength broadcasts to central control of signal from points of control. POINT OF CONTROL * Receiver/ * receives signal, sets switches 1 per point of control Switch -- Tone, 1 function $44 - $49 - Digital, 1 function $45'- $55 - Digital, 2 function $50 - $60 - Digital, 2 functions with $70 - $90 automatic reset * Trans- * receives and transmits information $90 - *95 .1 per point of control ceivers/ in 2-way system at point of control Switch 1/ A 100 watt for transmitter or repeater generally covers an area about 25-40 km in radius around the transmitter, depending on the terrain. -34- Table 2-4 ETRZD ("ADI/PLC SYSTEMS) CMPONENT COSTS (1982 U.S. S) Component Capabilities Costs Number/System CENTRAL CONTROL a Message * originates load managenent commands $12,500 - $25.000 1 per system (2 for added Generation reliability) Unit (com- puter) COnMCATIonS * Radio * receives signal from message Sener- 6 1,500 - * 9,000 depends on area to be Transmitter acor and broadcasts signal served. * Repeater * rebroadcasts to distant or inaccas- $10.000 - $15,000 depends on areas to be sible areas (same equipment as covered and geography transmitter plus receiver). * Radio Re- * receives signal from radio trans- $ 120 1 per 8 to 12 points of ceiver siter, injects PLC signal on low control Current - voltage distribution line Transmitter (RCX) POINT OF CONTROL * PLC Re- e receives PLC signal. control* ceiver/ switches Svitch - I function S50 - $55 1 per point - 2 function $80 * $95 -(volume discounts) -35- Table 2-9 TELEPHONE (SICNAL) SYSTEMS COMPONENT COSTS (1982 U.S. $) Component Capabilities Costs Number/System CENTRAL CONTROL * Central Computer e Receives, processes, and coor- dinates all load management - Main Frame Computer $75,000 - $100,000 1 per system (2 for (capable to process multiple added reliability) calls simultaneously at 3-5 seconds per call) - Micro Computer (capable to $15,000 - $35,000 1 per system (2 for process only one call at a time added reliability) at 1 minute per call. COMMUNICATION e Communication * Located at central control uses $ 11,000 1 per system provides 8 Substation (Card voice tone communication. No call instantaneous pro- cage) equipment necessary at telephone cessing. Additional units company (End point must initiate may be added together with call, then 2-way communication main frame computer to exists between central control increase processing capa- and end point) bilities. POINT OF CONTROL * Transponder * Initiates all communication. $250 - $260 1 per point of control - microprocesser Records total kWh consumption provides 2 function control device and one other function, such as: - second meter - TOD meter - demand meter - load control . * Load Control performed on each power line " PLC/Switch * One required for each load manage- $ 30 1 per load controlled ment function -36- Table 2-10 COAIAL CAMLE COMPONENT COSTS (1982 U.S. S) Comoonent Capabilities Costs Number/Sysees CENTRAL CONTROL * Central Computer * originates all control $17,500 - $70,000 1 per system (2 for added functions, receives and pro- reliability) ceasses all operational and measurement data. * if 2-vay system, computer receives and processes information * Headend Equipment * radio frequency modulators, $2,000 1 per system modmse, and multiplexers to inject signals into cable network COMMUNICATION * Rub toptional) * distribution substation $19,500 - $72,000 control substation required for larger systems requires computer and, headend equipment e Cable Trunk a primary distribution $8.000/km network a Trunk Amplifiers. * amplifier located on $9,300/ka .85 amplifier/mile trunk lines * feeder Network a secondary distribution S8,000/ka s-etvork, * Amplifiers (line * amplifier located on $9,300/ka 2.7 ampifier/nile extenders) feeder network * Filter (optional) e diplex filter required $200/ka for 2-way icmunication * Above ground- -e Infrastructufe for cable $9,600/km hardware and transmission-network equipment POINT OF CONTROL * Taps (coaxial * brings signal from feeder $250 *$260 usually L per point drop equipment) network to end-poinc receiver point of control * Addressable Taps * point-of-control receiver $90 - $100 1 per point of control located at point of control (one function) or centrally located to control $100 *$110 1 per point of control multiple loads (two functions) $8,000 1 per point of central control (multiple functions) * Power Line Control a load relay switch required $130 1 per load Switches for central control addres- (optional) sable taps. -37- Table 2-11 SOLAR SYSTEMS FOR SPACE HEATING (1982 U.S. $) Small Medium Large Systems Systems Systems DESCRIPTION Serves 120 Serves 180 Serves dwelling square meter square meter larger than 230 dwelling dwelling square meter ACTIVE $3,500 - $9,500 $11,500 - $21,500 $19,500 - $37,500 SYSTEM COSTS PASSIVE $2,500 - $7,500 $7,200 - $10,500 $8,500 - $15,500. SYSTEM COSTS COMMENTS: Estimated installed costs for residential solar space heating system. Solar domestic hot water system costs range from $750 to $5,500. Range of costs arise from range of solar systems and manufacturers in market today. Domestic hot water costs shown in Table 2-13. Table 2-12 TYPICAL SOLAR DOMESTIC HOT WATER BEATER COSTS (1982 U.S. $)- Collector (70 ft2) $1,185 Collector Supports .310 Solar Tank (Single 120 gallons) 660 Reat Exchanger 300 Piping 645 Miscellaneous 75 TOTAL (New) $3,175 NOTE: Installed cost of system designed for Boston, Massachusetts climate and providing one-half of the residential hot water heating demand. -38- Chapter 3 SYSTEM DESIGN CONSIDERATIONS This chapter discusses technical factors which a utility must consider to evaluate or design load management systems. The. chapter is divided into four major parts. The first part discusses considera- tions which relate to equipment installed at the point of control. The second part describes factors which affect the communications and control component of remote control systems. The third part discusses the equipment costs of alternative configurations of remote control systems. In the last part, conclusions are presented. POINT OF CONTROL Point of control considerations relate to that portion of the load management system located on the customer's premises. In many load management systems, all of the hardware is located on these premises. In others, point of control equipment may be only a part of the system. Five issues are considered for each of the load managerent tezhao- logies: * Effectiveness - the ability to control loads as intended. * Sophistication - inherent in technology (e.g., number of functions performed or level of technological sophistication) or required of the operators of the load management system (utility or customer). *- Flexibility - may be the ability to: - serve different types of loads, - operate in different modes, or - change functions/be programmable. * Reliability -- as illustrated by failure rates, consequences of failures, and operating or maintenance efforts required. *. Safey - potential for physical harm to people or property. Local Controllers There are four types of local controllers: priority relays, demand limiters, time controllers and load management thermostats. -39- Priority Relay. * Effectiveness: This technology has demonstrated limited success in reducing local (point-of-control) peak loads. Moreover, any savings in local load may not be coincident with the system peak periods. * Sophistication: The technology involved in the priority relay is a simple relay switch interlocking two separate loads. No operator training is necessary. * Flexibility: The operation of a priority relay requires that one non-essential load (such as a water heater) be inter- locked with an essential or priority load (such as a range or pump). This -characteristic of the priority relay requires that it be used with specific combinations of load types. Use of the priority relay in isolation cannot be tied to temperature -or time. This may limit its ability to be used alone as a daily operation temperature control device. * Reliability: No reliability problems have been identified. * Safety: The priority relay produces no safety-related problems, provided that curtailing the "nonessential" load to which it is connected produces no problems. Demand Limiter. * Effectiveness: Demand limiters are highly effective in reducing local peak loads; however, savings in local load may not be coincident with system peak periods. * Sophistication: Demand limiter technology involves conti- nuous monitorintg of demand in order to switch loads as required to keep demand below predetermined criteria. Most often, this is offered as a component of a larger energy management system. * Flexibility: This is a highly flexible technology which can be applied to many different kinds of loads. * Reliability: No problems have been identified. * Safety: No problems have been identified. Time Controller. * Effectiveness: Time controllers have been demonstrated to be highly effective in reducing local peak loads; however, these savings in local load may not be coincident with system peak loads. More advanced time controllers may be operated with minimal user discomfort or disruption. * Soohistication: The technology of time controllers encom- passes a broad range of sophisitication. At the low end,. there is a one-circuit, 24-hour mechanical clock with only one on/off operation. At the high end, there is a 16-circuit, programmable, microprocessor-based programmable controller with a one year schedule and multiple daily' on/off operations for each circuit. Programmable controllers are rapidly evolving. * Flexibility: Time controllers may be used with any load - capable of time scheduling or duty cycling operation. The more advanced microprocessor control units may be expanded with links to communication controls, load management thermtstats, or demand limiters. * Reliability: Time controllers have been used commercially a number of years and have demonstrated no reliability problems inherent to the technology. * Safety: Systems have varying capability to respond to safety-related problems. Both mechanical clocks and micro- processor-based controllers are available with back-up power systems. Load Management Thermostat. * Effectiveness; Load management thermostats are highly effective for reducing local weather-sensitive loads. User discomfort can be minimized with the more advanced revi- sions. Reduction of system peak requirements is dependent on the level of coincidence of weather sensitive load. * Sophistication: The technology ranges from basic duty- cycling units to time-controlled units with demand limiters which are sensitive to outdoor temperatures. Operator requirements range from mere setting of the thermostat to operation of programmable time controllers and/or demand limiters. -41- * Flexibility: The technology is limited to heating and air conditioning loads. Operational functions may vary from preheating or cooling to outdoor temperature sensitive systems operating in conjunction with time controllers and/or demand limiters. * Reliability: No reliability problems have been identified. * Safety: No safety problems have been identified. Remote Control Systems There is a basic similarity among each of the point-of- control devices for remote control systems. One-Way Systems. At the point of control, each technology involves a receiver and a set of one or more relay switches, which may or may not be integrated into a single unit. * Effectiveness: Each of these systems is capable of switching one or more circuits upon demand, depending upon the design of the receiver. Digital circuits allow more precise addressing of commands to specific end-users and uses. *- Sophistication: There are two levels of te'chnological sophistication: tone (which is fast becoming obsolete) and digital. Tone equipment sends or receives audio frequency signals (tones). Each frequency sent may represent a *different message. The amount of information that -can be carried by the tone signal is limited by the number of available frequencies. Digital signals allow more complex messages to be sent. Microprocessors at the point of control transform the signal -to switching commands after the signal has been received. * Flexibility: Digital receivers allow more specific addressing of points of control and switching functions. All remote control systems allow the utility to reset the timing of switching functions as required to adjust for changing conditions. * Reliability: The receivers for most of the technologies have, in large-scale production, recorded failure rates of between 0.5 and 2.0% per year. This is greatly affected by the utility's inspection and maintenance program. Many of the vendors provide simple tests for other receivers which can be used, for instance, by meter readers. The reliability of reception varies by technology. -- Ripple generally has a very high reception rate. -42- Power line carriers systems have experienced some difficulties as some operators have found it necessary to repeat signals. End-use equipment (especially appliance controllers and intercoms) may cause interference. - Radio reception (especially AM) is subject to interference. System operators generally report 90-95 percent reception. Reliability may be improved, at a cost, by overlapping transmitter reception zones. - Telephone signal reception is as good as the tele- phone system. Telephone systems also require close cooperation between the electric and telephone systems. - Coaxial cable generally produces the highest reliability because of the quality of the communications link. - Hybrid (radio and PLC) systems generally have higher reliability than radio because of the higher - locations (at the top of a distribution pole) of the receiver. * Sa fe: No specific problems- have beeh identified with any of these systems. Two-Way Systems. Both sophistication (number of functions capable of being performed) and the flexibility (ability to adjust to changing conditions) are improved by two-way systems. Reliability may also be increased because failed equipment may be rapidly identified. The performance of a two-way system depends to some extent on the type of system. Two-way ripple systems, for instance, are limited by the relatively slow data transmission rate-as slow.as 0.5 bits per second. By contrast, power line carrier systems allow data trans- mission at up to 128 bits per second and coaxial cable allows several thousand bits per second. Communications and Information Systems These include SCADA and multibuilding systems. SCADA systems rely on communications modes, such as those discussed above under remote control systems. Hence, SCADA is not discussed separately in this chapter. Multibuilding Systems. These coordinate the energy management activities of several buildings. -43- * Effectiveness: There is a need to have a balance of loads which is structured to foster cooperation among participants. * Sophistication: Control activities may vary with the building. Generally, these systems require a computer terminal and a trained end-use equipment operator in each building. Load control may be accomplished manually in each building independent of system technology. * Flexibility: Multi-building systems allow the response of each building to be specifically tailored to building condi- tions and to changes in those conditions. This may be a crucial advantage in managing such loads. * Reliability: The reliability of multibuilding systems depends primarily upon the willingness and ability of each of the building operators to actively participate in this cooperative program. Thermal Energy Storage Systems The characteristics of thermal energy -storage systems vary with the system. Six types are discussed below: Ceramic Heating Units (room and central). * Effectiveness: Ceramic heating units are highly efficient in reducing heating demands during local peak periods. As such, they are not appropriate for utilities in cold climates. These systems may be operated on a time device or may be connected to a utility remote control system. * Sophistication: Ceramic heating units depend on proven and relatively straightforward technology. No operator training is required. * Flexibility: The only use of ceramic heating units is as an off-peak heat source. Staggered -changing schedules provide the means of avoiding excessive demands as the units- are charged. Remote controlled units provide the greatest flexi- bility. * Reliability: Room units are in widespread use in Europe with no reported difficulties. * Safety: No safety-related problems have been identified. Pressurized Water Heater. * Effectiveness: - Pressurized water heaters are highly effective in reducing local peak loads. Load reductions, however, may not be coincident with system peak periods. * Sophistication: The components of pressurized water heaters have been in widespread use .for many years. Their configura- tion as pressurized water heaters in a thermal energy storage system is new. No operator training is necessary. * Flexibility: This thermal storage system is designed for application to space and water heating. User applications range from residential to large industrial. Tank storage systems may be designed to operate on short off-peak periods using a larger unit. * Reliability: This technology has been on the market for a number of years. No reliability problems have been identi- fied. Usually pressurized water heaters require less mainte- nance than !conventional heating systems because water is heated through heat exchangers, thereby avoiding the input of circulating water into the tank. This minimizes corrosion of tank and heating elements. * Safety: No safety-related .problems have been identified. Ice Cool Storaze. * Effectiveness: Poor results have been obtained for residen- tial applications- because of the evening and night cooling .requirements of residential users. These requirements limit the ability to charge the units during off-peak periods. Commercial and industrial applications have shown good load reductions in peak cooling - loads - because their energy consumption is primarily on-peak. Load reductions, however, may not be entirely coincident system peak periods. * Sophistication: Ice cool storage relies on .existing, generally-proven technologies. No operator training is required. * Flexibility: Application is limited to space cooling. * Reliabilitv: No reliability problems have been identified. * Safety: No safety-related problems have been identified. In-Ground Heat Storage. * Effectiveness: Effectiveness depends on building design and ground conditions. This technology can be highly effective under optimal conditions. Thermal storage savings vary with the thermal insulation and the vapor barrier of the building. (A building with larger heat loss would require a larger heat storage system). -45- * Sophistication: In-ground heat storage consists of resistant wire heating of slab and ground. No operator training is required. * Flexibility: This is a thermal storage heating system with limited operating flexibility. * Reliability: No reliability problems have been identified. * Safety: Heating mats must be covered by a solid, watertight slab in order to prevent electric shock. Annual Cycle Energy System (ACES). * Effectiveness: ACES is highly effective in reducing local weather-sensitive peak demands. Load x=ductions, however, may not be coincident with sys.tem peak periods. * Sophistication: This is highly sophisticated technology which is still in the developmental stage. ACES draws upon heat pump, solar, and thermal storage technologies. * Flexibility: Annual cycling technology which involves both winter heating-and summer cooling plus domestic water heating. * Reliability: Reliability cannot yet be determined because this technology is still under development. * Safety: No-safety-related problems have been identified. Daily Cycle Energy System (DCES). * Effectiveness: DCES is designed to reduce local- daily peak loads. Load reductions, however, may not be coincident with system peak periods. * Sophistication: This is a highly sophisticated technology which is still under development. DCES draws upon heat pump and thermal storage technologies. * Flexibility: DCES uses off-peak thermal storage as a means to improve the local load factor. No other application is available. * Reliability: Reliability cannot be determined because this technology is still under development. * Safety: No safety-related problems have been identified. -46- Suolemental Energy Systems .Supplementar energy systems provide an alternative source of energy to utility-generated electricity. Load management objectives may be achieved when the supplemental energy source is used at the time of system peak. Active Solar Energy Systems. * Effectiveness: Active solar energy systems are most attractive for meeting energy demands during daylight hours. Hence, load management objectives are best achieved when these daylight hours coincide with the -utility system peak. Fluids used for high-temperature applications are generally expensive to collect, transport and store. Broad application requires combination with economical energy storage. System effectiveness depends on amount of storage (volume of storage tank) and the schedule for charging the storage tank. Liquid * thermal transfer medium is more efficient than air transfer medium. * Sophistication: The technology for active solar energy system ranges from relatively highly unsophisticated, thermal system to highly sophisiticated. The primary use is for domestic hot water heating, space heating and. cooling. Thermal electric and photovoltaic technology are not yet futly commercial for most applications. Conventional electromechanical (thermostatic) controls are used with active-solar energy systems. * flexibility: The degree of flexibility depends on geographic location, degree days and storage capacity. Except for expensive photovoltaic systems, active solar energy systems can only handle domestic hot water and space heating loads. Variations in operating mode are limited; the schedule for charging the thermal energy storage component is the primary determinant of the operating mode. * Reliability: Reliability and maintainability are a function of design. Low maintenance -and high reliability may be realized at the expense of lower thermal transfer efficiency for systems with air as transfer medium. Maintenance problems associated with liquid-based systems result from freezing, corrosion and leaks. Condensation can also lower efficiency. Severe maintenance problems lead to loss of reliability. * Safety: No safety problems have been identified. Passive Solar Energy Systems. * Effectiveness: Passive solar systems may be designed for new buildings and, in some cases, retrofit to existing buildings. Their effectiveness depends on site and building orientation. Energy conservation is generally achieved more readily than load management because of the indirect link between operation and utility peak periods. These systems provide substantial load management benefits only when space heating is a significant portion of the utility's peak load. * Sophistication: This is generally an unsophisticated techno- logy. Passive solar systems, however require high degree of architectural design sophistication to control comfort condi- tioning. Space helating is derived directly from solar energy. Thermal energy flows naturally--there is no use of fans, pumps or blowers. Passive solar devices are not separate mechanical systems, but are a part of building. Some passive systems may be controlled by operation of blinds -and shutters requiring involvement of building inhabitants. There may* be optional exterior vents for summer cooling operation. * Flexibility: Such systems are limited to applications for space heating (and cooling) for single-family -homes and commercial facilities. There is no operating flexibility. * Reliability: Generally, passive solar systems require low maintenance and have .high reliability. One .retrofit option, adding attached sun space (greenhouse) to tan existing dwelling, requires a high level of maintenance. * Safety: No safety problems have been identified. Cogeneration Systems * Effectiveness: Used in applications with significant electric and thernial loads, cogeneration can improve the efficiency of on-site fuel use by jointly providing electri- city and thermal energy. Most commercial cogeneration units burn oil and gas in smaller sizes (less than 7 MW), and on oil, gas and coal in larger sizes. * Sophistication: Cogeneration systems can be used in a number of different operating modes that determine its peak load reduction capability. A cogeneration system can be operated to "follow" the customer electric or thermal loads, or to follow the utility's peak load. Only if the system is producing power during the time of the utility's peak load, can it be considered a load management option. Utility peak load reduction will also result if the customer's peak load is coincident with the utility system peak. -48- Industrial cogeneration systems generally produce electricity as a by-product of steam production. In many cases, advan- tages can be gained by first using the high-temperature combustion heat to generate electricity and using the by-product heat in the industrial process. Use of cogeneration requires experienced mechanical and electrical system operators. Control and monitoring equip- ment is relatively sophisticated and is more complex than standard building maintenance or boiler system operation. The interface between the cogeneration system and utility grid must also be carefully engineered, including appropriate protective devices. - * Flexibility: Cogeneration is best suited for operation in facilities with thermal demands close in magnitude to electrical demands. Operating options expand when the system is connected to the utility grid or when hot water thermal storage is added. For maximum load management impact, the systems must operate at the time of the utility peak. Operating flexibility is limited. There are two types of cogeneration systems: topping cycles and bottoming cycles. In a topping cycle, a prime mover burns fuel directly and is used to drive a generator. Waste heat from the prime mover is used to satisfy thermal require- ments of the industrial process. In a bottoming cycle, waste heat recovered from an industrial prqcess is used to generate electricity. Topping cycles are more commonly used. Topping and bottoming cycles each have certain operational . constraints. Topping cycles may be operated to follow either thermal or electric loads. Following either type of load may mean that, if the two are not fully coincident, the other is not always entirely satisfied by the cogeneration systems. Alternative sources of energy (e.g., backup boilers or electricity from the utility) may be required. Bottoming cycle operation depends on the availability of waste heat. This availability may not be coincident with electric demands. Careful balancing of prime mover and the loads and thermal energy storage may, in some cases, reduce these problems. * Reliability: The reliability of a cogeneration system is similar to the reliability of any on-site boiler/turbine or diesel engine power plant. It is a function of the redundancy designed into the system. With an isolated cogeneration system (not interconnected with the electric utility grid), redundant power generation capability is needed for satisfactory system reliability. Backup or -49- supplementary boilers are usually included as part of the system. Cogeneration system maintenance requirements depend on overhaul schedule, vendor repairs and spare parts support, and capability of site operating personnel. * Safety: Standard personnel safety precautions applicable to machinery and equipment operation should be followed. Special consideration should be given to any interconnection with the utility electrical grid. Dual Heating Systems. * Effectiveness: Dual heating systems offer load management advantages only if a significant portion of the peak load is attributable to electric heating. In such applications, it has proven highly effective. Implementation is dependent on fossil-fuel availability. * Sophistication: These systems are turned on or off by a standard thermostat. They switch from electric to fossil- fuel operation by either manual, local timer or remote microprocessor control. * Flexibility: Operating flexibility depends upon the degree of control exercised by the utility (through remote control communications) and on the ability of the specific device to rapidly change fuels. * Reliability: No reliability problems have been noted. * Safety: Standard precautions as applicable to electric resistance or direct combustion equipment must be taken. COMMUNICATIONS AND CENTRAL CONTROL Remote control load management systems involve communications between the control center (usually of the utility) and the customers whose loads are controlled. The applicability of remote control load management systems is determined both by geographic concerns and by other technical considerations. Geographic considerations related to each communication-based load management technology are summarized in Table 3-1 while other considerations are reviewed in Table 3-2. -50- Characteristics of the data to be communicated provide a starting point for designing a communications system. These data characteris- tics include the: * numbers and locations of data sources, * types of measurements communicated, * frequency of messages, and * required timeliness of receipt. The performance of the selected communications technology must meet specifications for data transmission based upon the required data characteristics. Design of the comminication system thus involves three major steps: 1. Identifying technologies which meet the specified data communications. requirements; 2. determining feasible system configurations, given available geographic, technical and inscitutional constraints; and 3. selecting the least-cost, adequately reliable system configuration. Each of the remote control load management systems allows centra- lized control of customer loads. Typically, this is managed by a computer capable of handling all of the intended functions of the load management system. Software also must be adequate to perform all of these functions. The development of such software frequently involves considerable effort on the part of both vendor and utility. Inadequate software may delay or reduce the effectiveness of a load management program. It is important for the utility to secure guaran- tees of software performance and usable documentation from the system vendor. Two of the load management technologies have unique central control considerations. * Telephone (signal) requires business and technical coopera- tion and favorable rate treatment by the operator of the telephone system if calls are to be centrally originated. * Radio (commercial) requires an available commercial radio station with appropriate area coverage and frequency allocations. Table 3-1 GEOGRAPHIC CONSIDERATIONS RELATED TO COMMUNICATION SYSTEMS System Remoteness Density (Urban/Rural) Terrain Ripple * Coverage of large areas requires a No upper limit on number of receivers per * Has no effect. higher-powered more extensive transmitter. equipment to overcome losses. Power Line Carrier (PLC) * Signals attenuate rapidly over a Significant primarily because distance and a lies no effect. long distances. changes caus,e attenuation. Economics improves as number of receivers per injection unit increases. Radio (Direct) * Unidirectional radio signal has a No upper limit on number of receivers per trane- 0 Sensitive to hilly areas and range of 5-25 miles depending mitter. man-made objects. on signal strength and terrain. May be farther if signal * Economics improves as number of receivers strength greater than 300 watts per transmitter increases. is allowed. Radio (Commercial) * Commercial transmitters are * Has no effect. * FM broadcasts are sensitive operated at higher powers to blockage by hilly areas (generally up to 100,000 watts) . and man-made objects. AM than are utility-operated broadcasts are less sensi- transmitters, giving greater tive. Un area coverage. Telephone (Signal) * Depends on existing telephone * Depends on existing telephone infrastructure. 4 Has no effect. infrastructure. Coaxial Cable * Marginally more expensive to * Lees dense areas increase coat of line per a .Difficult terrain may make serve remote areas. receiver. Density is a key determinant of costs. laying cable expensive unless there is a pre- existing utility right-of-way. Hybrid (Radio/PLC) a Hybrid systems may be used to a Radio/PLC hybrid systems are specifically 0 Combination of radio penetrate "pockets" of load in designed for dense areas. and PLC may overcome radio's remote areas. terrain problems. Multi-Building * Has no effect. * Close proximity reduces distribution system a Has no effect. Systems peak. Table 3-2 OnlER COMINICATIONS SYSTEM CuNSi PVllONS Remote Control Power System propa,educes/ System Design Infrastructure Relaisticataon Beability Flexibility Ripple a Able to be injected a Tr4nssion and Dim- 0 "'j.t applicat;4a are a Widespread use in a Nultiplo'control of each at any T4D voltage tributio.1 (T&D) system. unifirectional. Europe aad elsewhere location. levels with system- a Rid;rOctional *Yateml for over 30 years * Significant coamind wide propagation. are now emerging. no reported prob- flexibility. * No reported problems les with Injoc- Modular ependability. penetrating coverage tion and control Areas. equipment * d receiver tlllo are nase ths%I per year. a Utility a full con- trol over system. Power Line Carrier 0 May haveo problems * Transmission and Dis- a Higher frequency a Load management as- a In Gooms case. other (PLC) passing through tribution (TOD) system. than Ripple; requires pdrience only now lunctions of b14i- transformers or a Appropriate frequencies loas power-cou.su.alns being developed. Factional system may capacity banks. must be available to as and cheaper solid 0 Utility has fullt con- offet higher coes Possible corrections not to interfere with state devices. trot aver system. arising from power Include$ trapping radio. 0 ahi'lifectional systems a Sensitive to noise on system problems. capacitors, inductors, may allow multiple transmission line, a Modular expendlability.* Amplifiers, irapa- tunetiono. a Wi1th proper frequency tors, etc. selection multiple signals may be sent simultaneously. h Radio (Direct) a Independent of power a Appropriate radio a liniditectionmal is very * AN propogation in con- a Unidirectional digital system design. -frequencies must be simple -- no conlectia 4uctivo structures is switches offer con- ;vailable. h,ardwaria needed, limited. siderable flexibility [ Stes for transmitters a Bidirectional requires a Considerable experience for performing many ad repeaters must be much more sophisti- in U.S. functions. vailable and acres- catod equipment, a Receiver switches have a Modular expendability. sible. a Either Am (.5 - 1-32 per year failure a May be wsod as Part 2.0 llz or Fit (do0 - rate, prevrotive a - of systetid eYea 160 aily may e nmsed. tenlac and repair ad pretesting reduce in- field failures. * Risc of system tappering or other interference. a An -- most affected ry atmospheric conditions, especially night-time sky-wave.propagat ion. Also way be effected by electromagnetic inter- ference. a tr -- high signal to noise ratio gives trana- misuion reliability. a Fault location relatively easy because failure can occur at few sites. 0 utility lie lhll control over syptera. Table 3-2 (Continued) OTHER COMMUNICATIONS SYSTEM CONSIDERATIONS Remote Control Power System Preparednesh/ System Design Infrastructure Sophistication Reliability Flexibility Radio (Commercial) * Independent of power * Requires commercial a See Radio (Direct). * See Radio (Direct), but * See Radio (Direct). system design. AH or FH radio station , utility does not have with appropriate area full control over coverage and frequency systems. authorization. Telephone e Independent of power a Need access to tele- a Bidirectional ability a Control of some sys- a Ability to expand depends system design. phone system connected is readily available, tmes may be shared on coverage and quality to load sites. with telephone system of telephone service. * Data transmission rate operator. depends on degree of circuit conditioning. Coaxial Cable a Independent of power * Requires availability * Bidirectional service, a Usually highly reliable a Dedicated lines may allow system design. of dedicated communica- and numerous functions communications. non-load management uses. tion lines (e.g., cable are available. TV). Otherwise cost of * installation will be high. * Need right-of-way for line. Hybrid a Need appropriate in- * Appropriate radio fre- a Unidirectional involves * Radio segment may be * Allows system to be jection points for quencies must be avail- only marginal increase subject to interfer- tailored to service area change for communica- able. in complexity. ence. needs. tions methods. * * Bidirectional diffi- culty greatly in- creased. Multi-building * Independent of a Common building * Bidirectional commu- o Requires reliable a Can handle multiple Systems power system design. ownership or coopers- nications required. communications (e.g., functions of industrial tve organisation. coaxial cable or dedi- commercial loads. cated telephone lines). -54- TOTAL COSTS FOR REMOTE CONTROL SYSTEMS Component costs of load management technologies were discussed in the previous chapter. For point of control systems, the total system costs are just a multiple of the per-point cost, adjusted for any applicable volume discounts. By contrast, the cost of a remote control system depends upon the characteristics of both the load management system and the utility which it serves. Remote control system designs for specific applications can vary in numbers of points of control, capabilities, equipment configurations and geographic layout. * Each of the remote control systems has a computer-based central control component. Many systems have one computer at a single central control facility; some have a second computer at that facility to ensure reliability; others have microprocessors located with the signaling equipment at multiple remote sites; still others are designed for distributed computation with both central and remote computers. The number, locadion and power of signal transmitters also vary for each technology. The configuration of the signal transmitters may change with variations in the density of the points 'of control, terrain and design philosophy. There is* frequently a design tradeoff as to where to locate transmitters. Cheaper, smaller transmitters may usually be located close to the points of control, but many are required, especially where density is low. Fewer, higher-powers but more expensive units may be located at greater distances from the point of control. Frequently, two communications methods are used, a more reliable or expensive system (e.g., dedicated phone line, coaxial cable) between central control and signal -transmitters, and a less expensive or less reliable method to the points of control (e.g., ripple, PLC, or radio). Similarly, some two-way systems may use a different communications method for each way. Reliability may also be improved by redundancy in each of the system components, but only at a cost. The design of a remote control system depends on many factors unique to the application and to the performance and reliability requirements of the utility. Cost Estimates Total remote control system equipment costs are estimated for sets of assumed conditions. These conditions are selected to illustrate cost sensitivity to key determinants: * scale of application, * geographic density, * adverse conditions, -55- * signal directionality, and * sophistication. Hypothetical load management systems are defined. Calculations of total system equipment costs were performed for three load management program sizes: 1,000, 10,000, and 100,000 participants. Six sets of assumed conditions relating to the key determinants were developed and are summarized in Table 3-3. A key aspect of the cost specifications are assumed volume discounts for large purchase of equipment located at the point of control. These assumed volume discounts are summarized in Table 3-4 and generally reflect the magnitudes of discounts currently quoted by vendors. As will be seen, these discounts play an important role in economies of scale. The base case represents conditions which would normally yield low costs for implementation of the load management system. This is a dense urban setting with no adverse conditions. The system is assumed to be unidirectional with only one control function. The other five cases illustrate the effects of changing conditions in ways that may increase costs.. * The dispersed case shows the effect of moving from a dense urban setting to a less dense rural area (but with a lower per-point KVA load). * The adverse conditions case represents what happens when the system must be designed for geographic or power system condi- tions which create -problems other than those represented by other cases. * The bidirectional case indicates the-cost of two-way communi- cation for a single function. . * The sophisticated case shows. the effect of using the most technologically-advanced system. * The high cost case shows the effect of using the most expensive system under the most adverse conditions. The range of potential situations is effectively bounded by the base and high cost cases. -56- Table 3-3 DEFINITION OF CASES Parameters Adverse Signal Sophisti- Case Density1 Conditions2 Directions cation3 Base Urban None 1 Simple Dispersed Rural* None 1 Simple Adverse Conditions Urban Yes* I Simple Bidirectional Urban None 2* Simple Sophisticated Urban None 1 Complex* High Cost Rural* Yes* 2* Complex* Indicates parameter which is changed from base case. 1/ Urban - 1000 points/km2, Rural -50 points/km2 2/ The adverse condition varies by technology: Ripple - High load Power Line Carrier - Repeaters needed Radio - Hilly terrain Hybrid - Hilly Terrain Telephone - None defined Coaxial Cable - Hubs needed 31/ "Simple" systems have the least expensive equipment to perform only one load managment functions. "Complex" systems perform multiple functions with the most sophisticated systems available and have redundancy to ensure reliability. '-57- Table 3-4 ASSUMED QUANTITY DISCOUNTS Unit Costs Technology Component 1,000 10,000 100,000 Ripple Receiver/Switches Unsophisticated $90 $85 $80- Sophisticated $180 $160 $140 Power Line Receiver/Switches Carrier Unsophisticated $110 $100 $90 Sophisticated $160 $140 $120 Transceiver/Switches Unsophisticated $210 $190 $170 Sophisticated $340 $320 $300 Radio Receiver/Switches Unsophisticated $55 $50 $45 Sophisticated $90 $80 $70 Transceiver/Switches $95 $93 $90 Hybrid Receiver/Switches (Radio/PLC). Unsophisticated $55 $53 $50 Sophisticated $95 $88 $80 Telephone Transponder $260 $255 $250 Coaxial Tap to Point of Control $260 $255 $250 Cable Addressable Tap Unsophisticated $100 $95 $90 Sophisticated $110 $105 $100 -58- One concern is the preparedness of the electric power system. Geographic and technical considerations which could increase costs were dealt with in the dispersed and adverse condition cases Aside from the factors addressed in those cases the only consideration which could affect costs identified was the availability of a telecommunica- tions network between the central control facility and any signal transmitters located at sites remote from the central control facility (often at substations). In some cases, utilities may already have existing communication links to the transmitter site. These may be dedicated telephone lines, microwave networks or coaxial cable systems.' If such a system is needed and not already in place, there will be additional cost for its installation. The magnitude of this cost will depend on the telecommunications system chosen, geography and the availability of a nearby telecommunications infrastructure (e.g., telephone lines). Results Table 3-5 presents the results for each case. Only the equipment cost component of capital costs is estimated. Installation costs depend primarily on the amount and cost of labor, which may vary significantly by country. Similarly, operating and maintenance (O&M) costs are primarily a function of labor cost. For this reason, neither installation nor 0&M costs are estimated. Hence, when cost estimates for different load management systems or different configu- rations of the same system are compared, only a partial -picture is obtained. Base Case Comparison For the 1,000, 10,000 and 100,000 point-of-control situations, there are great differences in equipment costs among the systems. Radio and hybrid (which relies on radio for -long-distance communica- tion) are the lowest cost systems. Radio involves the least-cost communications equipment and uses a relatively inexpensive receiver at the point of control. The cost shown for hybrid is slightly higher than that for radio. The prices for hybrid-system for PLC receivers quoted by the only vendor are higher than those for similarly-capable radio equipment. This negated the presumed cost advantage of hybrid systems-having 10-12 control points share a single radio receiver. Coaxial cable systems are by far the most expensive systems reviewed. These systems have very high-cost communications and point-of-control components. Telephone systems were the next most expensive system in the base case because of the high cost of the sophisticated point-of-control equipment required. Ripple and power line carrier costs were in between, with power carrier being slightly more expensive, mostly because of somewhat more expensive receivers at the point of control. -59- Table 3-5 EQUIPMENT COSTS OF REMOTE CONTROL SYSTEMS (Million 1982 U.S. $) 1,000 CONTROL POINTS Case Adverse System Base Dis- Condi- Bidirec- Sophis- High Case persed tions -tional ticated Cost Ripple , 0.13 0.11 0.13 4.05 0.27 4.08 Power Line Carrier 0.13 0.13 0.13 0.28 0.27 0.51 Radio 0.07 0.07 0.10 0.12 0.23 - 0.25 Hybrid (Radio/PLC) 0.08 - 0.09 - 0.25 0.26 Telephone 0.32 0.32 0.A2 0.32 0.74 0.74 Coaxial Cable 0.81 6.07 0.83 0.84 1.01 6.31 10,000 CONTROL POINTS Case Adverse - System Base Dis- Condi- Bidirec- Sophis- High Case persed tions tional ticated Cost Ripple 0.92 0.91 0.93 40.09 1.76 40.18 Power Line Carrier 1.03 1.04 1.07 1.98 1.57 3.95 Radio 0.52 0.52 0.54- 0.96 1.66 1.94 Hybrid (Radio/PLC). 0.67 - 0.71 - 1.94 1.97 Telephone 2.90 2.90 2.90 2.90 6.47 6.47 Coaxial Cable 7.83 60.30 7.85 7.97 9.09 61.95 100,000 CONTROL POINTS Case Adverse System Base Dis- Condi- Bidirec- Sophis- High Case persed tions tional ticated Cost Ripple 8.5 8.4 8.6 400.6 14.7 400.9 Power Line Carrier 9.1 9.2 9.5 17.6 13.0 36.8 Radio 4.5 4.5 4.5 9.0 14.1 18.1 Hybrid (Radio/PLC) 6.2 - 6.2 - 17.3 17.3 Telephone 28.1 28.1 28.1 28.1 62.2 62.2 Coaxial Cable 77.6 602.1 77.6 78.9 89.0 617.5 -60- Equipment cost comparisons between radio and hybrid and between ripple and powerline carrier are particularly close. The relative cost advantages of one over the other would depend in large part on the specifics of the application and the exact price quotes from vendors. Economies of Scale Each of the technologies except telephone and coaxial cable exhibits substantial economies of scale between 1,000 and 10,000 points of control. Still lower costs per point result in the 100,000 point case. (See Table 3-6.) A substantial portion of these economies result from discounts for very large scale purchases of point-of-control equipment, as quoted by vendors. Should these discounts not be realized, economies of scale would be reduced, in some cases, substantially. Most of the economies of scale, are realized over a range of system sizes below 10,000 points of control. Table 3-6 BASE CASE COST PER POINT FOR DIFFERENT SIZE SYSTEiS (1982 U.S. $) Number of Points 1,006 10,000 100,000 Cost Per Cost Per % Change Cost Per % Change System Point Point from 1000 Point from 1000 Ripple 126.0- 91.9 -27.1 85.3 -32.3 Power Line Carrier 130.0 103.0 -20.8 91.3 -29.8 Radio 73.0 52.1 -28.7 45.2 -38.0 Hybrid (Radio/PLC) 81.0 67.1 -17.2 62.2 -23.2 Telephone 316.0 289.6 -8.4 280.6 -11.2 Coaxial Cable 809.9 782.6 -3.4 775.7 -4.2 Dispersed Case Ripple, power line carrier, radio and telephone do not exhibit major cost increases in moving to a less dense setting. The reasons vary by technology. For ripple, more communication equipment is required but, because of the much lower KVA load per point assumed for rural areas, this equipment need not be as powerful. The two effects balance each other -61- out. The costs of moving from an urban to a rural setting in an actual situation will depend on the relative loads per ripple pulse generator. The costs of a power line carrier system do not increase dramatically in a dispersed setting because the cost of the communica- tion component is relatively small compared to total costs. Even though more signal injection equipment is used, the additional cost does not make much difference. Radio costs do not increase dramatically for the same reason as those for PLC do not; most of the cost is in the receivers, not the communication equipment. A dispersed case was not estimated for hybrid systems because these are specifically designed only for dense urban applications. Costs of telephone systems do not increase-for rural areas because the telephone system infrastructure was assumed to -be already in place in both urban and rural areas. If this were not the case in the rural area, costs would go up (but the communities would obviously reap benefits far beyond load management). The equipment cost of coaxial cable systems, already high, jumps by an order of magnitude in the dispersed case because of the vastly increased amount of communications equipment required. Adverse Conditions This case is designed to examine the effects of technical or geographic conditions (other than density) which could increase costs. The "adverse" condition varies by technology. For telephone, none was identified.. The effects of the others are discussed below: * Ripple. The assumed adverse condition was a 100 percent increase in load per point. This doubled the cost of the pulse generator. Since.the pulse generator is such a small portion of total costs, however, total costs do not rise significantly. * Power Line Carrier. Depending upon the design and loading of the power system and the design of the power line carrier, there may be significant difficulties in communicating over even a few kilometers. To remedy this situation, repeaters are installed every few kilometers. This increases total system costs by five to ten percent. * Radio. Hilly terrain, which blocks transmission, is a significant problem for radio systems, solved by adding repeaters. The addition of a few added repeaters does not significantly add to cost, but even more difficult terrain, coupled with dispersed points of control, would probably substantially increase costs above those shown in Table 3-6. -62- * Hybrid Systems (Radio/PLC). The particular hybrid system analyzed used radio for long distance signal transmission and thus has the same difficulties as do radio systems. * Coaxial Cable. Long distances of transmission require additional hub signaling units. Rough geographic terrain could be one cause. Adding an additional hub increased costs, but not significantly in the scale indicated. The cost was only a small portion of the total system. The relative lack of effect of these adverse conditions occurs because they all act on the communications component of cost. At the large program sizes (10,000 and 100,000 points), this component represents only a small fraction of total costs. At the smaller program size of 1,000 points, the effect is more noticeable. Bidirectional The costs of all systems except for telephone and coaxial cable (which are inherently bidirectional) are greatly increased by the addition of two-way communications. This is most dramatic for ripple, where the signal injection unit costs $4,000 per point because of the high power requirements. Indeed, coupled with the ripple control's slow data transmission rate (around 0.5 bits per se'cond), this virtually rules out 2-way ripple commudications for anything other than a limited number of high-value applications. Two way communication by radio and power line carrier, while- much cheaper than ripple, is still not a well-developed commercial technology. Indeed, one possible 2-way system would be to use ripple to the points of control and power line carrier for the return communications. Sophistication The sophistication of .each of the systems was increased by adding redundancy to the central control facilities to increase reliability, by using the most advanced communication system available, and by increasing the number of functions and complexity of the point of control equipment. For each technology, this raises costs substantially. It is difficult-to compare technologies for this case because the capabilities of the most sophisticated units varies greatly. High Cost Case For most remote control systems, there is a considerable range between the lowest-cost application and the highest cost, "gold-plated", difficult situation. This illustrates that the economic feasibility of a given load management system is very application-specific. Characteristics of the technology, the utility and the load all affect equipment costs. -63- Chapter 4 UTILITY AND CUSTOMER CONSIDERATIONS This chapter discusses the utility concerns and the characteris- tics of loads which make load management desirable and possible in developing nations. The chapter also presents contract issues which must be resolved between utility and customer. LOAD MANAGEMENT POTENTIAL IN DEVELOPING NATIONS Several considerations affect the future of load management for an electric utility. These are supply adequacy, load factor, generation mix, management capabilities and customer acceptance. The signifi- cance of each of these considerations to a specific developing country depends upon the unique situation in that country. Supply Adequacy The adequacy of electrical supply for a utility -system is most broadly measured by its current and projected reserve margins. The relationship between supply and demand requirements is a major consi- deration in assessing the potential for load management. If load growth is greater than planned generating capacity additions, then reserve margins and reliability fall,, and the utility has difficulties in meeting load. A utility which is, or expects to be, capacity- constrained generally must plan additions to -its generating plant. In the period between utility identification of a need for additional generating resources and the time that the utility becomes substan- tially committed to the construction of. new facilities, load manage- ment may be a particularly attractive option. System Load Considerations The attractiveness of load management systems depends upon the "peakine " of the system as expressed by annual and daily load factors... These load factors depict the relative needs of the system for baseload and peaking capacity. 1/ Load factor during a period is determined by the following equation: Load Factor = Consumption (MWh) / (Peak load x Hours) "Annual" and "Daily" load factors are calculated over periods of 8760 and 24 hours respectively. if the annual load factor is low (less than 55%), the utility is likely to have a significantly "peaked" annual load profile. In that case, the implementation of load management programs serves to produce greater utilization of existing capacity resources and, possibly, greater generating efficiency. This can yield both immediate energy. cost savings and deferrals or cancellations of future capacity addi- tions in generation, transmission, and distribution. The season during which the annual peak occurs is influenced by both climate and status of development. Load management techniques were initially developed in Europe where utilities were winter-peaking because of high space heating loads. Utilities in other countries may have loads which peak in different seasons. Some developing nations, for instance, experience summer peaks because of fan use or irrigation pumping. These different loads and seasonality patterns imply different load management strategies. The load which is controlled may - or may not -- be the load which causes the peak. It is more important that the reduction in load be coincident with peak demand, whatever the nature of the controlled load. The potential feasibility of electric load management programs is further influenced by daily load factors. Low daily load factors reflect significant fluctuations in daily load levels. -Large varia- tiona in daily load re-juirements typically result in inefficient utilization of generating resources and increased energy costs. Where low daily load factors lead to the regular use of fossil fuel burning generators to satisfy customer demands, load management programs may offer substantial energy cost savings. Generation Mix A utility's mix of existing and planned generating resources determines the relationship of generation costs to electric demand and to the shape of the load curve. A utility which meets peak demands with oil-fired combustion turbines, for example, would usually have a large incentive to reduce the operation of these peaking units. Conversely, hydroelectric facilities with reservoir storage provide a flexible means of meeting daily peaks without consuming expensive fuels. Thus, load management to reduce daily peaks is generally less-attractive where hydroelectric generators with reservoir storage constitute a significant portion of generating capacity. The cost and performance characteristics of a generating unit make it most economical for a certain duty cycle. For example, high capital costs and low operating costs, typical of large coal or nuclear units, make these type of units most economical for baseload operations. Optimal dispatch of an electric system occurs when there is a match between the intended duty cycles of available generation capacity and patterns of electrical load. Utilities with sharply "peaked" demands would be best served, for instance, by a generation mix with a high proportion of peaking units. Utilities with "flat" demands would be best served by a high proportion of baseload units. -65- Load management may be used, in part, to bring the loads profile more closely into an economic match with the generating mix. The matching process, however, is dynamic. Over time, both the load profiles and generating mix change. This may sometime cause uninten- ded effects. Load management in Germany, for instance, achieved very high daily load factors in the early 1970's. These high load factors, however,. had the unintended effect of increasing the utilization of expensive-to-operate peaking units. Such units were all that was available to meet demand in the filled-in demand valleys. Management Capabilities The planning and implementation of load management programs are often viewed by utility personnel as complex and difficult. under- takings. This perception stems from three factors. First, the equip- ment and terminology of load management are relatively unfamiliar. Second, these programs typically require interaction among individuals and departments within the utility who have had only limited prior contact and who may understand little of each other's activities. Utilities with demonstrated technical and management capabilities may be better prepared to integrate load management technologies into their system. Utilities with less technical and management capabili- ties may find it difficult to implement a load management program effectively. Third, load management programs require significant interaction with electricity consumers beyond traditional activities (e.g., hookups, provision of service, metering and billing). New types of cooperation must be established. Customer Acceptance The customer, too, *must have both a means and an incentive to redude the load. Disruption of production activities will be an issue for some types of industrial loads, for example, where processes must be operated continuously. If turning off or reducing the use of electrical equipment or shifting its use to another time is overly burdensome,- little coincident load reduction will occur. Obviously, not all loads can be controlled without disruption. Domestic electric- lighting in the early evening may be one example. Many utilities in developing nations have early evening peaks made up largely of this end use, and its reduction may be neither practical nor desirable. A utility which intends to implement a load management program must secure and maintain customer participation. These customers use electricity because it serves their personal needs or allows them to carry on with their own business activities. Since load management may represent a reduction in the quality of service for the customer and may require altered business practices or lifestyles and addi- tional customer effort, incentives may be needed. Usually, these incentives take the form of rate discounts or credits for the purchase of electricity contingent upon participation. Incentives, however, may also be provided by increased rates for nonparticipants. x-66- Customer acceptance may also be influenced by the frequency and duration of service interruptions and by the availability of usable energy substitutes. Developing countries vary considerably in types and characteris- tics of electric loads. The extent to which controllable loads are present in developing countries may be affected by: * mix of traditional and modern economic activities; * split between urban and rural electricity consumers; * penetration of electricity by sector and end-use; * overall energy use patterns and costs; * agricultural and other productive practices; and * climate. The modern sector generally consumes more electricity and is more amenable to load management, but some traditional activities may also be controllable. Typically, electricity replaces animal power and sometimes a diesel engine. Traditional service, commercial and manufacturing facilities may become electrified or replace diesel generator -sets with grid-supplied electricity in which case they may be subject to load management. Rural electrification programs require major capital investments. The initial levels of demand in newly-electrified areas may be low, as may be the load factor, because, at least in part, rates are set too high. Load management may help to control these rates by ensuring an efficient utilization of power resources from the start.- In all cases the customer must be able to reduce or reschedule the use of electricity-consuming equipment, and the utility must be insti- tutionally capable of enforcing this control. Four broad classes of modern-sector customers may be served by electric utilities: industrial, commercial/institutional, residential and agricultural. This ability to control loads varies within each class. Yet, some general conclusions may be drawn about each one of them. INDUSTRIAL CUSTOMRS The industrial bases of different nations vary dratatically. The industrial customer class encompasses mining and manufacturing activi- ties. Some of these are more electricity intensive than others. Not only do industries differ, but often there are different production -67- processes and practices within individual industries. Frequently, the load of a single industrial customer is large enough to justify a significant effort to control just that one load. Some developing countries have a large and diverse industrial sector; the industrial sector of others consists of only a few facilities of the same type. Each country faces unique industrial load management problems. General Principles Industrial-sector load management must be carefully tailored to the specific requirements of the facilities served. Load management may be implemented when an industrial facility stores output or energy, reschedules electricity-intensive activities, or provides supplemental energy coincident with the utility system peak, usually through cogeneration. All three of these strategies may potentially require a substantial investment by the facility owner. Storage of output (or energy) at an intermediate or final stage of production allows the stored output to act as a buffer of supply for subsequent stages. When the equipment which produces an intermediate output is shut down or operated at reduced capacity, there are still inputs to subsequent stages. Similarly, shipments of finished pro- ducts may be from inventory. The ability to reschedule electricity- consuming activities is the key to enabling these activities to be performed at times off the utility system's peak. -CogenVration may allow the industrial electricity consumer to produce electricity during utility system peak periods. Storage of output or energy may require space and materials- handling equipment for solids or storage tanks for liquids. The amount of storage capacity required depends upon the length of the period over which production is to be shifted. Shifting production on A daily cycle requires less -storage, for instance, than on a seasonal cycle. Rescheduling requires adequate productive capacity to meet produc- tion targets during off-peak periods. For continuously-operated facilities, more equipment may be required. For plants operated in shifts, more shifts or manpower may be required. Cogeneration almost always involves a large capital cost for prime mover, generator and energy distribution equipment. Once an invest- ment in cogeneration is made, however, the cogeneration plant may not be operated just to meet the utility peak. In some cases, for instance, it may be more economical to operate the cogeneration plant in baseload or thermal load-following modes. These other operational modes may have advantages or disadvantages for the developing country or its electric utility. The operating mode will largely be a func- tion of the daily and annual tradeoffs between self-generated and purchased electricity. The need to meet thermal loads may also constrain cogeneration plant operations. -68- The capability to store output or energy, reschedule operation or cogenerate varies with the industrial process and specific facility involved. In many cases, the investments may be so large as to preclude economical load management. In other cases, they may be cost effective. The remainder of this section on industrial loads is divided into three parts. The first part discusses the application of the general principles to some of the more important industrial electricity end uses. The second provides three examples of different industries which use some of these processes. The last part describes several institutional barriers to industrial load management. Industrial Electricity End Uses Five end uses are responsible for most industrial electricity consumption: * electric motors, * process heating, * HVAC and lighting, a machine tools and other equipment, and * electrolytic devices. Different industrias and facilities use these (and other) electricity end uses in various combinations. These combinations determine the' technical potential for load management. In addition, computer control systems- which are rapidly evolving in major industries, may provide load management opportunities. Electric Motors. The major uses of electric motors in industry are to drive four types of equipment: pumps, fans, compressors and conveyors. Pumps are used in many industries to move liquids or gases. Refineries, chemical plants and paper mills, for example, typically have many pumps. The utilization of pumps can generally be shifted in time when there is adequate intermediate storage between pumping stages. In some types of facilities, such as oil refineries, only certain production processes out of the main process flow can be rescheduled. Production is continuous and scheduling considerations are already complex. Chemical plants sometimes find it difficult or expensive to stop and start production; frequently, they must maintain specific flow levels to maintain chemical reactions. Most industrial fans are used for waste gas dispersal. Some are used for pollution control equipment, light material transport or air -b9- cooling. The flexibility of fan operation for waste gas dispersal generally depends upon the criticality of moving the waste gas. Similarly, the flexibility of fans used for pollution control equip- ment depends upon the importance of control. Fans used for material transport can sometimes be rescheduled if there are intermediate stockpiles. Compressors are primarily used to produce compressed air. In some cases, they may be used for air or gas liquifaction. The production of compressed air generally involves the storage of output in a storage tank. Typical compressed air applications include pneumatic equipment in machine shops, assembly lines and glass bottle production. Usually, the demand for compressed air follows *the production schedule. Depending upon the capacity of the storage tank, there may be a buffer that allows off-peak production of compressed air. Air liquifaction, on the other hand, is a continuous process which is difficult to stop and restart. Hence, this is a difficult application for load manage- ment. Conveyors frequently offer the largest potential demand reductions for electric .motors where there is intermediate storage of materials. Steel mills, production lines, food processing plants, and wood products manufacturing frequently-have adequate amounts of storage for short-term rescheduling. Process .Heating. Electric furnaces are generally used for specialized .purposes where there are stringent temperature and time requirements. The primary examples are the manufacture of electronic - components and aerospace materials. Typic&lly, these are batch processes. Load management depends upon the ability to- schedule batches and the length of a production run. If the production run is too long, it may be difficult to avoid the utility's peak period. HVAC and Lighting. HVAC and lighting usually are a component of industrial electricity demand, but (depending upon the nature of the industry) usually not a large component. There are certain production processes that are highly sensitive to environmental conditions (e.g., temperature, humidity, dust and lighting); for example, the manufac- ture of electronic components and textiles. Even in environmentally- sensitive operations, some portions of HVAC load may be controlled, usually through short interruptions. Lighting is more difficult to reduce because of its frequently direct relationship to worker produc- tivity and safety. Machine Tools and Other Equipment. Manufacturing facilities use electricity for the operation of a wide variety of machine tools and other equipment used for fabrication or testing. The types of equip- ment which are used are as diverse as are manufacturing processes. For each type of equipment, there are usually also variations in size and operating characteristics. . -70- In some circumstances, the operation of this equipment -can be incorporated into a load management program. Incorporation generally requires that this equipment be scheduled for operation at times off of the utility's peak period. Typically, this is possible only where the electricity-using equipment is only one stage of a more involved process and where the materials that constitute the input for the electricity-using process can be stored for off-peak use. Scheduling of off-peak operation may, in some cases, imply new secondor third- shift operations, which may increase costs. Where a facility is capacity-constrained and already operating multiple shifts,, such rescheduling of output may be. made more difficult, but is not necessarily infeasible. Electrolytic Devices. Electrolytic devices are used for the reduction of various metal ores and for a variety of other processes such as electroplating and electronic component fabrication (e.g., circuit electricity). Metal reduction is particularly electricity- intensive. Where such facilities exist they may constitute a signifi- cant portion of a utility's load. The operation of electrolytic dev'ices is frequently continuous. In such circumstances they constitute a flat, constant load. Load may be difficult to control on a time-of-day basis under such circums- tances. Moreover, facilities for the more electricity-intensive process (e.g., the Hall process for aluminum) tend to be sited where electricity is cheap. Increases in rates for these electric customers may sometimes conflict with a nation's economic development objectives. Computer Control Systems. Industrial computer control systems are becoming increasingly common in many manufacturing industries, notably industries with complex processes such as steelmaking, petroleum refinery chemical production and automobile manufacturing. The type of computer control may vary from programmable and numeric controllers for specific processes to mainframe computer control systems. Increased use of computer control systems provides an opportunity for load management - but one that may be difficult to realize. Computer control systems allow industries to exercise greater control than ever before over manufacturing processes. Potentially, electric load management could be incorporated into the operational plans of these systems. The problem, however, is that enterprises generally install computer control systems because they need control to ensure efficient and economic production, not to control electrical loads. Incorporating load management into the computer control system requires careful evaluation of the operations of the specific facility. In some cases, operational constraints may preclude load management or make it difficult to implement. Pipelines, for instance, often have SCADA systems of their own to operate pumping stations. A pipeline must maintain flow and pressure throughout the system to meet deliverability requirements. Yet, where there is -71- adequate excess capacity, some' hourly or daily scheduling may be achievable and implemented through the pipeline SCADA system. Load Management in Specific Industries Three very different industrial processes are discussed below to provide a few examples of applying the general principles to specific industries and end uses. Petroleum Refineries. Petroleum. refineries are heat-intensive, continuously-operated facilities consisting of many different inter- dependent processes. Refineries vary widely in size and complexity. For all refineries, the basic process step is distillation of crude oil. Many refineries, however, have unique combinations of other processes. These processes either prepare the crude for distillation (e.g., sweetening operations) or modify the products which emerge from distillation (e.g., catalytic cracking, alkylation). Refinery operation must be carefully balanced in two ways. First, the slate of output products must be matched to the market demands for those products. This requires precisely controlling flows among the various processes. There are complex relationships among process outputs and inputs. Second, the thermal balance of the plant must be maintained. Some processes require the constant addition of uniform temperature heat; others may release thermal energy. Activities which affect either of these balances may disrupt the operation of the refinery. Electricity demand is primarily derived from the operation of motor-driven pumps. - Electricity is generally only a small fraction. (5-16 percent) of total refinery energy usage, most of which is thermal energy required to sustain various processing steps. Typically, the electric load profile of a petroleum refinery is flat, reflecting the refinery's continuous operation. Both storage of output or energy and rescheduling of operations are difficult because of the interdependence of the processing steps. While storage of intermediate or final outputs may be technically feasible, high costs associated with the construction of storage capacity may preclude economic implementation. Cogeneration is frequently integrated into refinery operations and is probably the most attractive load management option for refine- ries. A number of refineries in several European countries (France and England, for example) operate cogeneration plants at times when utility demand charges are high. The result is that the utility sees a shift from a flat load profile to one with inverted peaks (dips) at the times of utility system peaks. '-72- One oil refinery in France, for example, generates its own electricity during the winter peak price periods of the time-of-use tariff under which it purchases electricity.- These peak periods occur in both morning and late afternoon. During these periods, nearly all electricity is self-generated. and there are preciptious drops in demand for electricity from the utility. By contrast, on Sunday, when the same rate applies at all times during the day, the load profile is nearly flat. Cement Production. Unlike petroleum refineries, cement plants are batch operations. Cement plants are also much smaller than petroleum refineries. These are three basic sequential steps to cement produc- tion. * Raw material preparation: Raw material for cement consists of various minerals containing lime, silica and alumina in appropriate quantities. Preparation activities include blending, crushing and grinding. Grinding is generally performed in ball or roller mills. If the "wet" process is used, water is added to produce a slurry. If the "dry" process is used, the material goes to the next step without added water. * Clinker production: "Clinker" is the - fused material that results from heating the mixture prepared in the previous step. Heating is done in a kiln .- a rotating, cylindrical furnace lined with ceramic materials. * Grinding and mixing. The clinker generally emerges from its production as a mixture of hard lumps of various sizes. The final step is to mix the clinker with a small amount of gypsum (calcium sulfate) and to grind this mixture to a fine powder. . Fossil fuels are generally used to fire the kiln. Electrical energy is used for motive power. Most of this electricity is used for crushing and grinding the raw material and clinker; some is also used for rotating the kiln. There may be daily, weekly and seasonal components to the electric load profile. Daily variations occur as batches of materials are processed through each of the steps. Weekly variations occur as production batches are scheduled; equipment, such as grinders, is down for maintenance; and (sometimes) production is halted for weekends or holidays. Seasonal variations in load occur when demand for cement is seasonal. The batch nature of the manufacturing process and variations in load provide opportunities for scheduling of electricity-consuming activities. Production steps may be performed at times off the utility peak on a daily, weekly or even a seasonal basis. The steps which can be most readily scheduled are raw material and clinker grinding - those which are most electricity intensive. -73- Off-peak scheduling of grinding operations requires adequate grinding capacity to make all the necessary clinker and end product during off-peak hours. This capacity may not always be available, especially during seasons of peak cement demand. Storage for interme- diate or end product is also necessary. Longer storage cycles (e.g., annual) require greater storage capacity than do shorter cycles (weekly or daily). Additionally, the amount of moisture in both the intermediate and final products must be controlled. The overall economic feasibility of load management by rescheduling operations and storing ouput for specific cement plant will depend upon (1) the rate incentives from the utility and (2) the cost of any additional grinding or storage capacity. Usually, daily or weekly rescheduling will be less expensive than longer-term rescheduling. Such reschedu- ling has been readily achieved in many cement plants. One dry-process cement plant in the northern United States produces about 500,000 tons of cement per year. Demand for concrete from this plant is greatest during the summer, during which time the electric demand may reach 10 MW. The local utility, however, peaks in the winter when the cement plant has an excess of both grinding and storage capacity. This excess capacity allows winter grinding opera- tions to be rescheduled. Demand averages about 5 MW during off-peak winter periods. During the peak period of the electric tarriff, (10:00 a.m. to 6:00 p.m. weekdays) this demand is reduced to about 2 MW by shifting grinding of both clinker and raw materials to off-peak periods of the day or week. Assembly Operations. Many industrial facilities perform an assembly function. That is, they combine components manufactured elsewhere into a final product. Assembly plants make everything from trucks to transister radios, and range in size from establishments employing thousands to onily a few workers. Assembly operations vary tremendously, depending upon the product. Typical electric loads include space conditioning, motors for materials handling and machining, HVAC and lighting. Special- purpose equipment may also be used. Various assembly operations may be performed in sequential or parallel order. Manufacture of some components may sometimes be integrated with assembly. Assembly facilities frequently operate on a shift basis. Output is varied by adding or subtracting shifts, either for the plant as a whole or for production-constraining operations. Load-management opportunities arise from scheduling electricity-intensive operations during off-peak shifts or at more opportune times during the peak shift. Such rescheduling involves storage of intermediate (or finished) parts and, in some cases, the hiring of additional labor for new shifts. Where intermediate storage space is limited, the cost of adding new shifts is significant, or the balance of the production would be disturbed, rescheduling costs would have to be offset by adequate rate incentives to make load management cost-effective. -74- A major manufacturer of automobile engines in the midwestern United States saves $250,000 (U.S.) per year on a time-of-use rate for an engine assembly plant. Prior to implementation of the time-of-use rate, the plant had peak load of 16 MW, which was coincident with the serving utility's peak. This peak was trimmed to 12 MW by changing the hours of a shift on a machining line from 6 a.m. to 2:30 p.m. to the hours of 5 p.m. to 1:30 a.m. Institutional Considerations Technical considerations in industrial load management are important, -but institutional considerations may be crucial, especially for developing nations. The need to maintain production, lack of access to hardware and high capital costs may all provide barriers to industrial load management programs. The primary objective of any manufacturing facility is to fulfill production .goals within budget - regardless of whether production goals and budget are sec by market conditions or by central planning. Any activity whose purpose is different will be evaluated by plant management in terms of its effect on the achievement of these goals. Plant management will want to maintain control at all times over the facility. An effective load management program allows the plant management to maintain this control and. provides both the means and the incentives for load management. In recent years, rising energy costs have spurred energy conserva- tion by industries throughout the world. Generally, the firms which conserved energy in developing nations were large and energy intensive and were frequently multi-national organizations. These firms had: * incentives to save money through conservation, * knowledge of how to conserve, * ready access to conservation hardware, and * available capital for investment. Smaller, local firms in developing nations frequently lacked one or more of these capabilities. Hence, they found it difficult - or were not interested - in conserving. Energy costs may not have been a significant proportion of total costs; engineering staff may have fully preoccupied just maintaining production (or may have been unaware of conservation opportunities). Similarily, production staff may not have been trained to conserve energy. The firm's engineers may also not have had easy access to vendors of conservation equip- ment. Finally, smaller firms may not have had adequate access to sources of capital for conservation improvements. -75- These problems are even more severe for load management. The economic benefits accrue primarily to the utility -- unless shared with the participating firm. While conservation often has direct benefits in increased efficiency or decreased costs, load management may involve interruption or reorganization of production. Moreover, conservation sometimes involves only relatively simple-to-implement "housekeeping" activities and relatively-accessible and inexpensive equipment. Load management may be considerably more complex and expensive. The investment in storage, production or cogeneration capacity may be larger than the investment in conservation -- and the benefits less direct; The lack of economic benefits, may be a parti- cularly severe problem for small, capital-constrained firms. Industrial load management programs in industrialized countries tend to focus on the use of pricing signals to industrial customers while leaving the mechanisms for responding to these signals to plant management. Sometimes, these systems also involve direct- communication with a limited number of large industrial customers. These tariffs take a variety of forms. Each is a unique adaption to the electricity supply and industrial conditioti of the country involved. Similarly, there is no simple way to categorize the response of industrial firms to these tariffs. Each industrial response is uniquely determined by the nature of its production process and the market for its products. OTHER CUSTOMERS Other customers include: commercial/institutional, residential and agricultural customers. Commercial/Institutional The commercial/institutional customer class consists of a diverse set of electricity users who may be in either the private or public sector, sometimes depending on a country's economic system. Generally, they are far fewer in number, but considerably larger, than the residential customers. Typically, these users include retail and wholesale facilities of all kinds, and office buildings, hospitals, hotels, and street lighting as well as public service, schools and military facilities. Electricity may be used for a variety of purposes by commercial and institutional facilities. There are four major categories of loads which contribute to commercial/institutional electricity demand: * Motor loads for fans, elevators, pumps, etc. may be signifi- cant part of commercial/institutional demand and may, in some cases, be controlled. -76- * Lighting of office, retail and other occupied space, while noticeable, is frequently not a large portion of demand. However, it may be controllable through improved building design, turning lighting off at appropriate times, relamping and rheostat control. * Heating, ventilation and air conditioning (HVAC) loads depend on the climate of the country and the stock of HVAC equip- - ment. Large building systems are frequently designed with excess HVAC capacity and frequently offer a large potential for Load management. However, because of the complexity of large building temperature regulation and thermal characte- ristics, RVAC systems are not good candidates for remote control by the utility. Typically, the penetration of electrical HVAC equipment is not great in developing nations, except in Large buildings. * Other electrical equipment is used for water heating; office functions (typewriters, copiers); cooking; computers; dish- washing; and other uses. Frequently this use may be physi- cally reschedulable, but such rescheduling may require adjustments such as changes in working hours or practices. The telative magnitude and daily use profile of each of these electricity uses varies with the type of facility. Typically, aggre- gate, commercial/institutional demands are greater -during working hours, but the profiles of individual facilities may vary- widely. The primary determinants of a nation's overall commercial/institutional load profile are the existing stock of electric equipment, work practices, and climate. The ability to shift loads to non-peak hours also varies with the type of establishment. Retail facilities and schools, for instance, are frequently constrained to operate during daytime hours - which frequently overlap with the hours of utility system peaks. Hospitals and hotels may operate throughout the day, but may have peaks of activity during business or evening hours. Electricity use may be for a diverse set of purposes ranging from lighting to the operation of special medical instruments. Some uses of electricity (e.g., room lighting) may be determined more by patients, staff, or guests than by general management. Yet, some activities subject to central control may be rescheduled. Water may be heated, for example, off peak hours and stored for use during peak hours. This may, however, require a significant investment in thermal storage, which may or may not be economically attractive. Office buildings may have similar opportunities, although their energy use tends to be greatest during the country's business hours. Residential Typically, residential customer classes consist of large numbers of very small-volume -electricity customers. Within this broad -77- customer class, there are several uses of electricity. These include lighting, cooking, water heating, space heating, and electrical appliance cooling. The penetration of these uses for a specific country generally depends upon the size of the population, climate, size and type of dwelling units, individual income levels, and the general availability of electric appliances for consumer use. Air conditioning, for example, may be desirable in countries with warm climates but low levels of personal income may indicate low penetra- tion in the domestic sector. Individual dwelling units served by the utility may range from the most basic shelters with only lighting and small appliances (e.g., radio) to large homes with many electrical uses. The daily and seasonal electricity use patterns depend on the nature of the service being provided and domestic practice. Lighting and cooking, for example, are uses whch peak consistently at predict- able times of the day, and in all seasons, while water heating may be either periodic or steady. Space heating and cooling are associated with seasonal use. Still other use patterns, such as those for electrical appliances, vary with the types of appliances in widespread use, and are often correlated with consumer income levels. The primary concern in residential-sector load management is to find end uses which represent significant controllable load at the time of system peak. Generally, electric space heating, water heating and air conditioning are most controllable; their operation can be curtailed or postponed for short periods. Yet,, in many countries their use may not -be widespread. -Pumping groundwater for domestic purposes may be scheduled off peak if there is adequate storage capacity. In some countries, the daily electric system peak occurs in the early evening and consists mostly of domestic lighting, cooking, and' radios. Without affecting people's lifestyles, this is very difficult to control. Additionally, these may be exactly the uses that one wishes to encourage to improve the standard -of living. Agriculture Electricity use in agriculture depends upon the crops, livestock herds, and agricultural practices of the country under consideration. In many developing countries, the use of electricity for groundwater pumping and irrigation is a major benefit of rural electrification programs. In these countries, electricity often replaces the animal or human power previously used for pumping. Electricity may also be used for grain drying, livestock husbandry (e.g., cow milking, egg incubating), refrigeration and the preliminary processing of food and fiber materials. Load management is, as usual, most applicable where there are large controllable loads at the time of system peak. The major concern in the agricultural sector is to avoid damage to crops or livestock. This concern is usually most relevant with regard to irrigation pumping, where the delay of water for just a few hours can sometimes reduce crop yields. In some areas, the accessibility of water and the cimes o: J availability may be limited. In such areas irrigation pu.pinZ may be the major rural use of electricity. Different types of irrigatirn ar- used under different conditions. Each places different constraints -,n load management. Some crops may require irrigation on a 24-hcur per day basis. Some vegetable crops, on the other hand, may flourish wit. daily, but not necessarily constant, irrigation. Othtr cr:p -- including some fruit orchards -- reqaire water only every severaL days. Load management strategies depend upon the crop: * Full or partial interruption at peak daily periocs. This .%iy not be possible for heat-sensitive or 24-hour irrigated crops. * Daily rotation of irrigation among ocrtionso of the crl. This strategy is most applicable to croos which do nor oe 4 . water on a daily basis, such as fruit orchards. * Higher off-peak and lower on-peak irrigation. Th:s stcnat, may require increased pumping capability and not ops for some heat-sensitive crops. * Off-peak pumping into reservoirs and o.-Desk . crops. This strategy -ay req!tire- su*-:.nti ii-- pumping capability and restrvoir ster-ta. Clearly, any load management program fcr irriatj%n U.'It ; tailored to the specific-needs-of the crops served. Additirnal, an agiicultural load manaigement programs n:ust be coordinated -irh r . farmers. To avoid disruption to gr.owing practices, unchcided i. ruptions of irrigation, for instance, shotild not be undertakeiz. MATCHING LOAD MANAGEMENT SYSTEMS TO TYPES OF LOADS Load management systems perform useful functions for an el utility when they reduce the consumption of electri:: ey -cd which are significantly coincident with the utility 'yste on . load management technologies may be rore readiiused ~ certain end uses than other technologies. Th;e possi.e o mismatches) between the end uses prevalent at :he time of system pc.- and load management technologies are important considerations v system selection. The capability of different load management systems to cnntrol specific types of loads varies. Some are applicable only to spe- iHi: types of loads; others may be used for a wide and flexible ran-e ol applications. Some systems can manage only one t.pe of lo.id acco:C"t to a fixed schedule; others can control multisle oa.:: wit. degrees of flexibility. Up to three functions are performed by load management sytem: * Timing - The determination of when load managecnt are to be taken. -79- * Switching - Moving the end-use load to another operational state (e.g., "off" or to an alternate energy source). * Alternate service - In some systems, provision of an alternate source of power when the utility service is curtailed. The ability, cost and flexibility of load management systems to perform each of these functions determine its applicability to a specific type of load. Table 4-1 summarizes the types of loads to which load management systems are most applicable. Each technology for which a load management system is applicable is marked with an "A" or a "C". An "A" means that the end use is readily controlled by the technology. A "C" indicates that, while control is feasible, it may be costly or difficult. A discussion of how each of the load manage- ment system characteristics affect the applicability of each class of technology follows. Local Controllers Local controllers basically involve timing, switching, and/or temperature light sensing components. The characteristics which set them apart from other load management systems is that functions are determined on site. Frequently, this is solely at the disdretion of users, so that the use of local controllers may be matched to their specific needs. This is particularly important for complex commer- cial, industrial and agricultural applications which may have unique and changing requirements, although much simpler applications may be served as well by local controllers. The integration of the timing, switching, and sensing functions often involves relatively straight- forward and simple technologies. Larger commercial and industrial facilities, however, generally require more sophisticated systems (e.g., microprocessor-based controllers). Except for load management -thermostats, there are no specific technical limitations on the end uses which may be served by local controllers. Remote Control Systems Remote control systems differ from local controllers in that they shift the timing function to a central utility control point. This gives the utility greater control and makes the electrical system more reliable, but may make more difficult the control of complex or sensi- tive loads at the customer's service location such as might be encoun- tered in some industrial or commercial applications. Economies of scale usually dictate that there be large numbers of customers. Hence, while other types of loads may be served, application of remote control systems to residential loads is particularly appropriate. Another potential limitation of a remote control syst-em is the number of different types of loads which can be controlled. Greater flexibility and the ability to control more types of loads require sophisticated and expensive equipment. Voice telephone systems are . -80- effective only where there are a few very large controllable indus- trial facilities. However, the time and effort required for each call generally limits application of this technology to a small number of very large consumers. Communications and Information (C&I) Systems There are two types of C&I systems: SCADA and Multi-building systems. SCADA systems allow the utility's own internal functions (i.e., generation, transmission and distribution) to be controlled simultaneously with the performance of load management functi-ons. "Multi-building" systems are generally applied to large commercial and industrial electricity. consumers. They are particularly useful for this purpose. Such.systems allow each user to coordinate its energy management activities with those of other participants and with the needs of the utility. Moreover, the reason why an energy user would want to participate in a multi-user program is that his own energy use patterns are so complex and vital that he would prefer not to- commit individually to the required load reductions of a load management program. Cooperating with others allows the benefits, responsibili- ties and risks to be shared among the participants in* a mutually- beneficial way. Therial Energy. -torage S-sti=s Systems which store thermal energy perform by switching the timing of energy consumption rather than the "on" or "off" state of the load. Thermal energy storage systems, by their nature, are intended to control space and water heating and cooling loads. Process heating applications are limited by the temperature at which the thermal energy is stored. Generally, high temperature process heating requirements may be served by thermal energy storage systems only when they are used for pre-heating. Supplemental Energy Storage Systems Such systems switch to alternate service at times of electric ucilizy peaks. Active and passive solar energy systems may be used for space and water heating applications in all sectors. Cogeneration plants may serve both thermal and electrical energy needs of the sites they serve. Economies of scale and temperature requirements generally dictate that cogeneration systems be used for large industrial or district heating applications. Large, dense residential apartment or commercial developments may also be served by cogeneration systems. Cogeneration systems generally involve very substantial investments. Economical utilization of cogeneration systems may impose timing requirements on system operation, with -fuLl power output during utility peak periods. Coordination between utility and cogenerator is thus required for load management objectives to be achieved. Dual heating systems may be used for space, water or process heating applications in any sector, provided that the alternate fuel is available. Table 4-1 APPLICABILITY OF LOAD MANAGEMENT TECHNOIDGIES TO CUSTOMER LOADS LOCAL CONTROLLERS REMOTE CONTROL SYSTEMS C&T SYSTEMS Load Mgmt. Power Tele- Co- Multi- Priority Time Demand Thermoo- Radio Ripple Line phone axal building Relays Controllers Limiters State Control Control Carrier (Signal) Hybrd Cable Systems II. Applicability to Customer Load Types * Domestic - Space Heating A A A A A A A A A A - Water Heating A A A A A A A A A A - Air Conditioning A A A A A A A A A A - Refrigeration A - Cooking A A - Lighting A A - Total Duelling A A A A A A A a Commercial - lVAC Systems A A A A - Lig,ting A A A - Elevaturs/Escalators A A - Kitchen Facilities A A A - Computers A -Fans A A A A * Agricultural A - Pumping A A A A A C C C C - Drying A A A A C C C C e Industrial - Electric Motors 1) Compressors A A A 2) Fans A A A A 3) Pumps A A A A 4) Conveyors A A A - Process Heating A A A - Hachine Tools A A A - Lighting A A A - Steam Generation A A - IIVAC A A A Key: A - Readily applicable C - Typically, a costly or difficult application Table 4-1 (Contihued) APPLICAIILTY OP F AD NAUAGDJENT 'EIHObLELiU TO CUMTUMR UAMAM THLMIA. ENKRuv s'ustacK tiSTENs Ceramic låeatr Storagt Prdauri,sd Ico tinCround Combined Ha A iolfagte Room Contral Wster Coo Het Alnud& Cycie haity Uy.le m ype Unit Iinitu BIeat :torate tdima- Mtord&* EinI ½Stelaa knerKY Sstemia - SAae etigA A . - Witer Heating A A 4 - Air Conlitioning A A A - Refrigerationl - (Mdaing -u i.iitni - Total I'silinag -- Il..isystna i A A A A A . - A.,:Jat inc - .oval ..rålhbt.alators 6 - K atJ.eu Facalitiem - Lomå, * AgaiÉ.uiItiie , - li 1is l &ad'a't iat 1> ..as.qr.dase.rB .f l'.B. ). i...aÉv.y.as . - Nig ..4. Hr-. . g - M . m11.. q . - Iu'.4 . 1. - A 16~~ k 9 Nt Table 4-1 (Continued) APPLICABILITY OF LOAD ANAGEHENT TECHNOLOGIES TO CUSTOMER IOADS SUPPLEHENTAL ENERGY SYSTEMS Solar Energy Systema Cogeneration* Dual Heating Active Passive Systems Systems II. Applicability to Customer Load Types a Domestic - Space Heating A A A A - Water Heating A A * A - Air Conditioning A A - Refrigeration - Cooking - Lighting - Total Dwelling * Commercial - JIVAC Systems A A A A - Lightinig - El.vator/Escalators - Kitchen Facilities - C,)mputers - Fdns * Agriculture - Pumping - Drying A A o Industrial - Electric Motors 1) Compresaore 2) Fans 3) Pumps C 4) Conveyors - Process Heating A A - Machine Tools - Lighting - Steam Generation A A - RVAC A A A A Key: * Thermal cogeneration application only. A - Readily applicable C - Typically, a costly or difficult application - Uncommon or coat application CONTMACTUAL ARRANGEMENTS A load management program involves the utility and its customers to an extent considerably beyond the usual sale and purchase of electricity. It is important that all parties clearly understand what is required of them in a load management program. A written agreement may be the means to ensure this understanding. The specific terms of any load management contract depend upon the needs of each of the parties involved and on the legal practices of the relevant country. Where,there are large numbers of participants, standard contracts for each type of load management service may be necessary. In other cases, it may be more advantageous to negotiate unique contractual provisions tailored to a large end-user. A number of considerations are common to any agreement between a utility and program participants. 4 contract that does not explicitly address these basic concerns could potentially lead parties to work at cross purposes or could leave unresolved the very issues which may later cause problems. These common considerations are summarized in Tacle 4-2: Table 4-2' CONSIDERATbONS FOR ACREEMENTS ON LOAD MANAGE-MENT SEViCES Consideration Discussion A.. Parties The individuals or organizations who are the participants in the program a) Utility and who may be held responsible for b) Customer certain actions must be specified c) Third Party along with those individuals or organizations who may be signifi- cantly affected by actions taken under the contract. This may include, for example, separate specification of the owner of a participating facility and the customer of the utility when they are not the same. 2. Location of Customer Since the customer may obtain service Facilities at more than one location, this is an important descriptor of the participating facilities. 3. Equipment to be Controlled It may not be desirable for all loads at a service location to a) Type(s) be subject to control. For this b) Capacity reason it may be important to specify which equipment or loads -85- Consideration Discussion are to be affected. In some instances the amount of load or capacities of equipment subject to control may be differentiated by seasons, months, days of the week, or other parameters. 4. Character of Service Any .limits on the character of the service which the customer is to a) Voltage and Phase receive or limits on the portion b) Capacity Limits of the customer's service which are to be subject to control must be specified. For example, it may be agreed that only service obtained at a specific voltage level will be controlled. In addition, the amount of capacity which the utility will provide may be differentiated by time-of-use. 5. Permitted Load Control This is a statement of who is Activities allowed (or required) to do'what,. when and under what conditions. a) Type of Utility Customers often prefer some prior Actions warning of service interruptions or b) Timing, Frequency, curtailments as well as limits on the and Duration frequency and duration of load con- of Service trol activities. Interruptions or Curtailments c) Prior Notice of.Control Activities d) Required Customer Actions e)a Limits to Activities 6. Metering Special metering may be either a necessary or optional feature a) Type of Meter of a load management program. b) Frequency of Reading Where such metering is a considera- c) Cost Responsibility tion, the type of metering to be employed should be specified along with the division of cost responsi- bilities for the metering equipment and its installation. 7. Rates Rates for load manage ent partici- pants are generally designed to a) Rate Schedule provide cost-justified incentives b) Payment Conditions for load reduction activities. -86- Consideration Discussion c) Method of Billing Rates may be either detailed within the load management contract or referenced to a standard schedule. A key concern is often the method to be employed for measuring the kW or kWh of load reduced by the customer during periods of control. Load management incentives are generally treated as credits or discounts to the customer's regular electric service bill. Rarely are cash incentives paid directly to customers, except perhaps in experi- mental programs. 8. Installation of Load The parties need to agree on the dis- Management Equipment tribution responsibilities and costs related to the installation of load a) Who Pays management equipment. It may also be b) Who Owns What necessary to specify where and how c) Where and How such equipment will be installed. Equipment is Unlike most electric service equip- to be Installed ment, load management equipment tends to be installed on the customer's side of the meter. Thus, standard utility-customer relations may require some modification to reflect this change. 9. Modifications to The electrical equipment and Agreement requirements of many customers may change over time. Such a) Addition/Deletion or changes are likely to influence Replacement of either the amount of load subject Equipment to control or the reliability of b) Utility Approval load reduction capabilities for c) Notice the utility. To the degree that specific types of changes can be anticipated, it is important to provide the means for implementing timely contract modifications such that adverse impacts to the parties are minimized. 10. Non-performance by the The most critical form of non-perfor- Customer mance by a customer under a load a) How Determined management program is failure to b) Penalties provide load reductions when required. Since such non-perfor- mance can cause the utility to -87- Consideration Discussion incur significant additional costs and potentially create hardships for other customers,.penalties for non-performance may,be a necessary aspect of the load management contract. This is particularly applicable where the utility does not have direct control over power flows to the specifit loads within a customer's facilities which are' subject to control. If penalty provisions are to be included, they must specify how - non-performance will be measured or identified and how penalities will b-e computed. Generally, penalties are tied to a measure of demand during the period(s) of requested load reduc- tion. Penalties which are unduly harsh may, however, significantly discourage customer participation in load management programs. 11. Maintenance of Equipment The utility must be'granted reason- by Utility able access to equipment installed on the customer's premises to insure a) Inspection its proper operation and to perform b) Maintenance maintenance activities. Not all c) Responsibilities maintenance responsibilities and costs for load management equipment are necessarily borne by the utility. Depending on the type of equipment employed and the capabili- ties of the customer's staff, it may be more cost-effective or desirable for the customer to bear certain maintenance or inspection responsi- bilities. 12. Time Period This is a simple statement of the a) Term period over which the akreement b) Termination holds. If the participant leaves c) Notice the program early, certain utility d) Payment of Un- investments may not have been paid .amortized Expenses back and need to be accounted for. -88- Cotnsideration Discussion 13. Liabilities a) Damage to Utility Both utility and customer may have Equipment physical access to each other's b) Damage to Customer property in the course of the Equipment -operation of the load management systems. CONCLUSIONS Load management may be effective for loads of all classes of customers - industrial, commercial/institutional, residential and agricultural. The characteristics of the load management system must be matched to the attributes of the loads. The industrial sector consists of a diverse set of electricity consumers, although the industrial base of some countries may be quite limited, and. single industrial facilities may represent significant portions of a country's electric load. But in most cases, the primary concern of plant management will be maintaining production, not conducting -load management. Utility industrial load management practice throughout the world has been to design tariff incentives. Sometimes, direct communications with a small number of large indus- trial customers is part of the program. -In most cases, this tariff approach has been effective,. Residential loads, by contrast, are frequently controlled by specific local control, remote control and energy storage devices or by combinations of these devices. Many residential customers in developing nations, however, have few controllable loads. Commercial loads vary and sometimes have 'characteristics of either industrial or residential customers. Some types of commercial insti- tutional customers have fixed hours of operation and thus find it difficult to shift loads. The primary agricultural load which may be subject to load manage- ment in many countries is irrigation pumping. Management of this load must be carefully controlled in order not to disrupt crop growing cycles. Load management implies a new type of relationship between the electric utility and the customer. This relationship should be carefully designed so that it will be mutually beneficial over a long period of time. The East Africa Power and Light Company (EAP&L), the electric utility which serves Kenya, has been using a mix of ripple control and time switches for load management since the late 1940's. The -89- objective of this system has been to reduce peak demand, to control load during unexpected generation outages and to minimize generation costs. Over the years, target goals for each objective have been determined by the generation mix and the balance of load demand and available firm power. The ripple control system is also used for switching purposes on EAP&L's transmission and distribution system. Currently, nearly 30,000 customers are served by these load management systems. Over 90 percent of. these are controlled by the ripple system and the rest are controlled by time switches. Ripple control is preferred because it allows greater flexibility than time switches. Types of loads controlled include water heating, water pumping and street lighting. Customers who opt for interruptible service are provided an incentive in the form of a -lower tariff of about 55 percent of the cost of noninterruptible service. Enhancement of the system is continuing, with new ripple receivers being added in a phased program. Eventually, the entire territory served by EAP&L will be covered by the ripple control system, except for the Mt. Kenya area and isolatedf areas served by diesel generators. -90- Appendix A VENDORS OF LOAD MANAGEMENT EQUIPMENT This Appendix identifies vendors of load management equipment and categorizes them according to type of system. The equipment of a sample of vendors is also described in detail. The purpose of this description is to provide examples of key product features. The samples chosen for detailed description are very small subsets of the available equipment. Selection was based primarily on availability of information and does not constitute an endorsement of any products. Although a few load management technologies have been widely used for many years, several technologies have been only recently developed aind have only limited experience in the commercial marketplace. Documentation of equipment and applications thus varies in availabi- lity and quality. Standardization of equipment among vendors is, at best, minimal. Equipment and systems offered are typically not compatible with a different vendor's equipment. Interchanging componelts among different systems in the same generic category (such as replacing or adding comhponents in a- radio system) is typically an unknown capability. Throughout the World, there are hundreds of manufacturers of equipment which may be used for load management. These vendors and their equipment are quite diverse. Vendors of any type of equipment vary according to a number of characteristics, each which may affect their ability to serve utility customers in developing nations:. * geographic locations of facilities; * types of products offered; * function performed (e.g., original equipment manufacturer, -system designer, sales agent); * - markets in which products are sold; * ongoing support available to customers; and * size, manpower and financial capability. Even within a generic equipment category, products vary according to: * type and number of functions; -91- * flexibility; * performance characteristics; * price for each performance level; * availability to specific customers; * reliability; and * useable documentation. Some equipment is sold specifically for utility load management. Other equipment is sold for other purposes (e.g., timing, industrial process control), but can be readily adopted for load management purposes. Tables A-1 through A-4 (at the end of this section) list potential vendors of load management equipment. These vendors were identified by contacts with the companies and by reviewing literature and directories. New vendors and products are enter the load management and related fields at-a rapid rate. There are undoubtedly many vendors, particu- larly of local controllers and solar equipment, who are omitted from the lists in this Appendix. This Appendix should serve only as a starting point in an effort to identify manufacturers of any specific type of equipment. Potential purchasers of load management equipment should obtain current information about the potential vendors before. making any selection. LOCAL CONTROL SYSTEMS Table A-1 lists vendors of local control and communication systems. The following discussion provides a brief description of some representative equipment. The selected vendors offer, equipment in which the sophistication is generally limited to simple mechanical or microprocessor based technology. However, certain vendors of the energy management systems offer large, complex computer-based systems. Priority Relay Jameson Electric. This priority relay system interlocks two electric appliances and then interrupts the power to one appliance when the other is in use. The controller, typically, interlocks the electric water heater with the range, dryer, air conditioner, ,or a heating circuit. The controller manages non-inductive loads of 35 amps or less. When the interlocked appliance or circuit is activated, the controller senses the current and trips a relay shutting off the water heater. -92- Time Controllers Entertec/Schlumberger. A 24-hour time switch is offered. The system can operate in either synchronous or asynchronous mode based on the electric power supply frequency or can be based on a quartz clock. Depending upon the model, the production device controls up to four circuits with several different switch configurations and allows up to 16 openings and 16 closures in a 24-hour period. Xencon Inc.. A multi-circuit, digital display, programmable, seven day clock that switches circuits on and off. It is available in four models, differentiated by the maximum number of circuits to be controlled (4, 8, 12, or 16). Scheduling capability allows for 10-40 separate on/off operations to be assigned to one or any combination of the independent circuits being controlled. The cycle length is adjustable from 0 to 24 hours in 1 hour increments. Manual control override is possible. The Xencon time controller has an omitting device which can omit any operation(s) on any day(s) of the week. It can also schedule holiday operations one week in advance. Simple on-site installation with no major wiring involved. Battery back up is provided in case of a power -failure. Omron Electronics Inc. The Omron product line consists of microprocessor-based sequence and programmable controllers capable of controlling multiple circuits. The capacity of the Omron products are a function of both the number of control circuits and the number of functions programmed by the operator. The sum of number of circuits and functions determine the units capacity requirements--input/output capacity (1/0). Omron products range in capacity from 12 1WO to 96 1/0. Robertshaw. A multi-circuit, digital display, programmable seven day time clock is offered by Robertshaw which is similar in function to Xencon and Tork's microprocessor time controllers. In comparison to the other two time controllers, this time controller is not as versatile. A maximum of only four electrical loads can be individually programmed and scheduling capability allows for only six events or cycles per load. The time controller can omit days and also be programmed for forty-eight scheduled holidays per year. A battery back-up and manual override control are standard features also available. Staefa. The EMS Series controlers are single- and multiple- building energy management control systems. Utilizing a centralized computer-control complex and direct-wire remote control relays; the EMS product are capable of managing up to 5000 and point controls and sensors. -93- The EMS systems are capable of thermostats control, optimal start of heating and cooling systems, time scheduling, demand limiting, and duty cycling. Staefa produces a wide range of "smart" sensor devices, as well as, a variety of end point control devices. The versatility obtained from the variety of sensor and end point controls provide a flexibility which allows the Staefa control system to monitor and manage numerous types of loads. Demand Limiters (Energy Management Systems) Robertshaw. Several types of energy management systems are avail- able that are differentiated by the functions desired and size of the facility. The 2616 Energy Controller is capable of controlling up to 16 groups of electrical loads. It is a programmable microprocessor unit with manual override and power reserve operation -capabilities. The system performs time of day scheduling, duty cycling, optimum start/stop time, and electrical demand limiting. Duty cycle intervals are either 15, 30, or 60 minutes. For electrical demand limiting up to 3 demand limits may be entered for time of day pricing/regulation. This system uses the facility's existing power lines as a control medium. The controller is capable of interfacing with intelligent terminals for remote programming and information gathering. The communication medium is via a dedicated telephone line. Optional remote printers are available, also. This system is designed for small to medium size buildings. The DMS 2400-3 is a more complex system developed for large facilities. Such a system is used in applications where a hundred or more points are being controlled and monitored. The system is capable of duty cycling, optimum stop/start time, demand limiting, and outdoor air optimization (enthalpy control). The system consists of an operator console, a central processor, and field control panels. The communication medium is via two-conducted pair cable. The central processor is a disk drive computer with user-generated software capabilities. Used in conjunction with the operator console (including video display unit) they form the control command center. The command center controls and monitors the field control panels. The field control panel is a microprocessor unit which performs the control and scanning functions and will communicate with central control when a change of state occurs. In' actuality, several buildings or facilities are capable of being controlled and monitored by the central command control. The system can also interface with other separate fire/security systems. -94- Another system is available that will control the same number of points using similar control functions as does the DMS :2400-3. However, the system uses a different communication medium and was originally designed more for building management purposes with load control as one of the available features. The System 55 uses existing power lines as communication medium. It will monitor equipment, rooms, and open areas for fire, security, and status reasons. Its load management software capabilities are limited in comparison to the DMS 2400-3. Johnson Controls. The Power/Perfect 5000 controls facilities in the intermediate range of 920 and 7,400 square meters. It is similar in function to Robertshaw's 2616 Energy Controller. The Power/Perfect 5000 is a microprocessor based unit containing an operator panel for programming and a 40 character alphanumeric display for programming aide and statistical reporting. The unit controls a maximum of 32 electrical loads. It is capable of duty cycling, demand limiting, time of day scheduling, and optimum start/stop time. The communication medium is via existing power lines. Johnson's energy management system is capable of remote control and monitoring through the use of an optional CRT unit and printer. In addition to-a power reserve unit, this system also has manual, temperature, and humidity control overrides. Honeywell. Several energy management systems available ranging from simple functioning units to complex computer controlled systems. Honeywell's basic load controller is flexible because it is available in 14 different models. Each model capable of one, all, or a combination- of load control functions. However, only three load control functions are offered: demand limiting, time of day s-chedu- ling, and duty cycling. (Other similar load controllers described include an optimum start/stop function.) These load controllers control either 10 or 20 electrical loads (or groups of electrical loads). System status and programming errors are indicated by a light emitting diode (LED) display. (A more limited means of display as compared to an alphanumeric display.) Remote programming and monitor- ing can be accomplished by an optional CRT unit and printer. The communication medium is via a twisted pair cable. The system includes a power reserve unit and manual and temperature overrides. A much larger and more complex system is offered by Honeywell which is capable of controlling several buildings or approximately 40,000 points. The Delta series (1000 or 5000) is an integrated facility management system utilizing distributed processing techniques. -95- The central element of the system is a minicomputer with an extensive software library. The minicomputer operates in conjunction with several peripherals devices (remote operator terminals) located throughout the facility or facilities. The next level of control is performed by subsystem modules which communicate with the minicomputer via twisted pair wire. Each sub- system contains .its. own software and processor and shares the data processing load with the central computer. Subsystems interpret signals from an even lower level module. This lower level module is wired into individual sensors and the equipment being controlled. - The system is capable of numerous load control functions such as duty cycling, chiller optimization, demand limiting, optimum start/stop, and enthalopy control. In addi!tion to its energy manage- ment control, Delta is capable of fire, security, and equipment management control. Barber Coleman. Energy management systems available which are similar in function and complexity to Honeywell's systems. The Micro/8000 - is a microprocessor, based unit capable of controlling a maximum of 16 groups of loads. This load controller offers the same basic control functions as other energy management systems similar in size. The functions include duty cycling, demand limiting, time of day-scheduling, and optimum start time (optimum stop time not included). System status and programniing errors are indicated by numeric and LED displays. Optional CRT units and printers-provide the system with remote programming, controlling, and monitoring capabilities. Several load controllers can be controlled by one CRT unit. This gives the system flexibility because it has the potential to expand into several facilities. The system is equipped with a power reserve unit and manual and temper-ature control overrides. The communication medium is via four conductor shielded cable. Barber Coleman offers a much more complex facilities management system which is similar in function to Honeywell's Delta series, although not as sophisticated. The ECON VI performs typical energy management control functions with status reporting and equipment management (used for maintenance purposes) capabilities also. The control functions include demand limiting, duty cycling, optimum start time (stop time not included), enthalpy control, and chiller optimiza- tion. The communication medium is via four conductor shielded cable. The ECON VI is capable of controlling well over 40,000 points. A microcomputer, with user-generated software capabilities provides the central control for the system. The central control computer (used with a CRT unit and printer) receives and stores data -96- and also sends appropriate command signals to remote subsystem modules. These modules are wired to both sensing and controlling devices, thus, providing load management controlling and monitoring capabilities. Optional remote control terminals can be added to increase system control flexibility. Load Management Thermostats MicroControl Systems. A load management thermostat is offered capable of controlling up to 16 electrical loads. The thermostat sets back temperatures when the facility is unoccupied and turns off other uneeded loads. It can accomodate up to eight temperature sensors, each assignable to any number of pieces of equipment. The thermostat has an -intelligent display with alphanumeric prompting and information. In addition to temperature control and optimum start time capabilities, the thermostat also performs duty cycling and load scheduling. The thermostat is equipped with a power reserve unit. Solidyne Corp. Two types of load management thermostats offered which are differentiated by the number of air conditioners and heat stag-s duty cycled (either eight air conditioner stages and one heat sta:e or four air conditioner stages and one heat stage). Conti- nuously varying duty cycling is performed based upon outside tempera- tures. Temperatures -inside vary within a predetermined comfort zone. REMOTE CONTROL SYSTEMS Table A-2 lists vendors of remote control systems. Most of the components found in these remote systems are vendor specific and can not operate within another vendor's system. A sample of equipment is discussed below. - Radio Control System Motorola Inc. A complete radio control system consisting of a microcomputer based load controller, a base station, and binary digital radio switches. The load controller is capable of automatic and manual control. In automatic operation, electric loads can be shed or restored in a preprogrammed manner according to time-of-day rate periods. Two types of load controllers are offered, the ESC-400 and the ESC-200. The ESC-200 has up to 60 control codes and controls groups of radio switches through the application of several matrix strategies. The ESC-400 has up to 120 control codes and is capable of load control through either a matrix strategy, gradual strategy, or interval strategy. This gives the system more flexibility by providing a -97- choice of control strategies suitable for numerous control applica- tions. The ESC-400 can also be operator configured and if control requirements change or expand the operator can revise the operating parameters in accordance to the needs. Both systems include a CRT unit, optional printer, and a time clock (battery operated in case of power failure). The ESC-200 has a logic unit and the ESC-400 has a disk system and software package. The base stations are equipped with CSK (Carrier Shift Keying) signalling capability. This modification allows the base station to signal to binary digital radio switches as well as tone and tone digital (AFSK) radio switches. The communication medium with the controller is via telephone lines or -microwave circuits. The base station consists of a transmitter and an optional repeater located at the station or in the surrounding area. The repeater allows the system's coverage area to be expanded beyond the base station's trans- mitting range. A repeater positioned at the base station extends the coverage of adjacent systems. Base stations are capable of responding to simulcast (binary digital sytems only) or sequential tone signal- ling from the central controller. Motorola offers single function, dual function, and set/reset binary digital radio switches. In a binary digital coding structure, the receiver will not be falsely activated by tone or tone (AFSK) digital systems sharing the same channel. The set/reset switch is a latching type relay switch in which the relay remains in the same position until the correct command to open or close the relay is received. In a dual function application, the radio switch can be used to control independent loads separately or together. Single and dual function receivers automatically return to original position after a seven or fifteen minute period. An optional hand-held, microprocessor controlled, low powet FM transmitter will test the overall operation of the radio switches at the installation sites. The radio control system has an emergency shed function capable of deferring a large part of the system's load at once in an attempt to relieve blackout possibilities or other emergency conditions. Regency Electronics, Inc. A complete radio control system is also offered by Regency Electronics. The system includes a computer control unit, digital tone generator, base station, and receiver switches. The computerized controller can be preprogrammed for the load requirements of a certain time period. When discretionary load is to be deferred, the computer instructs the tone-generator to transmit the necessary message. The controller can be interfaced directly with a utility's supervisory control computer. Base stations consist of FM transmitters equipped with FSK signaling capability. -98- The radio switch is a crystal controlled FM receiver with microprocessor controlled circuitry, available in dual and single-function. Although more flexible than single or dual tone receivers, this receiver is available only in tone digital and not binary. Restoration period is either seven or fifteen minutes. The system is capable of emergency shedding. An optional hand-held FM transmitter will test radio switch operation at the installations. McGraw-Edison. A very unique, atypical, radio control system using AM radio as a communication medium. The system includes a computerized control center, an AM transmitter with a computer modulator, and radio receiver switches. The central control computer is similar in function to Motorola's ESC-400. The unit consists of a CRT terminal, a printer, a disk memory unit, and a software package for operator configuration. The base station can broadcast signals over 161 kilometers. For greater coverage and the elimination of obstruction problems, repeaters can be used. Messages sent from the control center via a telephone or microwave link, are processed by the computei modulator unit and then broadcasted by an AM transmitter. Two types of digital tone receivers are offered. One is a basic receiver which uses. various command codes directly and can. e addressed only as one of a group. The other receiver is microprocessor-based, handles a greater variety of commands, and can be individually addressed. The microprocessor cap-ability makes ic possible to alter local control switching strategy based on demand level, temperature, or other parameters. In addition to extending the receiving range by using an AM band, the system experiences freedom from jamming or interference caused bY other systems on the same channel. Plectron Corporation. Standard FM receiver/switch and latching- type receiver available. The standard receiver operates on rvo different frequencies, has a seven minute restoration period, and receives only single-tone commands. The latching-type receiver operates on three different frequencies and receives dual-tone commands. Ripple Control Systems Brown Boveri. The Brown Boveri ripple control system consist of a master control unit, substation injection units and load controL receivers. The master control unit consists of a microcomputer system capable of demand control and monitoring and SCADA communication for over 500 load groups and 98 substation remotes. It has a modular system design which supports expansion beyond standard load management capabilities.. The substation injection converts telemetered data from the Master Control Unit and injects the low-frequency ripple message -99- into the transmissions or distribution network. A microprocessor- based local control unit with the substation injection unit communi- cates with the master control unit, performs diagnostics and allows for manual operation at the substation. Signals may be injected from 4.2 kV to 161 kV over three-phase parallel or single-phase neutral (ground coupling circuits). LCR 500 load control receivers are available with 1, 2, 3 and 6 circuit control capabilities. They allow more than 20,000 individual addresses. Entertec (Schlumberger). The Pulsadis control system incorporates a centralized control unit, injection equipment, and control receiver relays. The Pulsadis central control console is a self-contained microprocessor-based programmable unit. Peripheral equipment is available to provide added communication capabilities either to a private or a central -computer complex. The basic Pulsadis system utilizes a coding system which allows up to 300 individual commands. More sophisticated software is available to expand the system capabi- lities to more than 1,000 commands. The Pulsadis injection equipment can operate on either 50- Hz or 60 Hz distribution systems. The control frequency ranges from 168 Hz to 340 Hz depending upon the characteristics of any utility distribu- tion system. The injecti-on equipment- possesses a power range of 15 kVa to 250 kVa. - The Pulsadis offers a variety of receiver models ranging from the electromechanical AIT 100, which is designed for mass applications, to the AIT 6000, a microprocessor-based unit with up to 3 independent control contacts. Landis and Gyr. This ripple control system (called "Control 80") uses three basic functional elements: a central programming unit, injection equipment, and control receivers. The central programming unit is a programmable microprocessor-based central controller. Time-dependent control commands can be automatically initiated. The unit can be connected directly to the system's injection equipment or interfaced via telephone or microwave links. The software package used contains the routines and procedures necessary to generate individual, group, global (SCRAM), and SCADA commands. The unit consists of a microprocessor unit, operator console, optional printer, and a battery back-up for data protection. Parallel, neutral/ground, or series injection equipment convert central control generated commands into audio frequency pulses and couple the signal into the transmission network. Parallel injection equipment offers wider area coverage for large integrated transmission systems whereas neutral/ground equipment is more economically suited for distribution networks and pilot systems. -100- The parallel injection system operates over an injection voltage range of 12 kV to 115 kV and an injection power range of 10 kVA to 1200 kVA. Neutral/ground injection is for distribution voltage applications (up to 40 MVA load). For both parallel and neutral ground injection applications, the equipment consists of a control unit, a frequency converter, switchgear equipment and a coupling filter. Input power is converted at 60 Hz into selected audio frequency and coded messages. The injection level is 1.0 to 2.0% line voltage. The control receivers are programmable micro processor-based units capable of operating two relays at three different signal frequencies: 220, 260, and 340 Hz. The Control 80 system also offers receiver meters that provide centralized control of time-of-use metering and load control for demand deferral. The power source for all Control 80 receivers is the 60 Hz line voltage. The systems coding structure permits definition of load types, load groups, and areas in a message signal. The same message permits iirdevendeat curn-on and. turn-off commands in any combination. The coding forma: is based on a 50-bit message. The system is capable of aot ocy control and time-of-use metering, but also power factor or o snd vottage regulation. RTE Zel1wezer. The Decabit is similar in function to Landis and Gyr's ripple system, and is also comprised of the same basic compo- nents: a control unic.injection equipment and control receivers. The central control unit consists of a microprocessor unit (this includes a four hour battery reserve and a printer) and a power supply unit. The Decabit's initial programming and program changes for fixed time operation, as well as channel selection for manual operation, are accomplished by means of decade switches located on the microprocessor unit. This differs from Landis and Gyr's system where programming and cther operator intervention is performed on a separate operator console (CRT display unit). Records of manually and automatically ,xecuted operations are provided only by a printer as opposed to Landis and Gyr's ability to display operations on a CRT screen as well as on a printer. The injection equipment consists of an audio-frequency converter- type generator and necessary coupling equipment. The audio frequency generator maintains an injection level of 1.0% line voltage at an injection power range of only 10 to 200 kVA. Coupling arrangements are available only in parallel or series injection. -101- Two types of solid state Decabit ripple control receivers are available. One receiver handles up to two load relays and the other up to four, neither have time-of-use metering capabilities. The power source is the 60 Hz line voltage. For multiple channel receivers, each load control switch or relay can be programmed independently of the others within the same frequency. All signal transmissions are coded using a patented coding scheme with error detection. Individual commands can be combined into groups of 5, 10, or 15 and transmitted by one single master command within 5.5 seconds. A single command consists of 10 bits. . Zellweger also offers a unique, atypical, complementary system .that expands the functions of the Degabit system to permit transmittal of information from the low voltage supply network to the substation. This bi-directional operation can be added to the current system when needed. Power Line Carrier Control Systems American Science and Engineering Inc. The ASEP carrier system is capable of uni and bi-directional communication. Typical of load management systems using existing power lines for communication and control, this system consists of three basic units: a central control computer,-substation control units, and remote units. The ASEP carrier system uses a frequency of approximately 6 kHz in a bi or uni-directional mode over the distribution network. The system is operated by a data dispatch computer with a redundant backup. The computer contains disc drives in which one is used for software development. CRT consoles and line printers are also part of the computer system. The communication medium between the central control computer and the distribution substation unit is via a telephone or microwave link. The substation unit, which retransmits signals from a central control onto the distribution line, consists of a control unit and a coupling transformer. Actual transformation of 120 V, 10 amp signals onto the distribution bus is performed by coupling transformers. One substation control unit serves up to five distribution buses (15 separate phases). There is one distribution transformer, 10 KVA minimum, per distribution bus phase. The substation control unit selects a bus, phase, and polarity combination to route signals. Signal frequencies are at 5790 Hz. For uni-directional control, two receivers are available. Both receivers can respond to 56 separate block address command codes. One receiver plugs into existing four-terminal meter socket without further wiring. It controls single loads, only. The other receiver is a surface mounted unit that is received directly into the circuit -102- feeding the electrical load. This receiver, when used with an auxiliary switching unit, is capable of controlling two loads. Both receivers have 15 minute restoration periods that can be adjusted. In bi-directional applications, load control is performed by transponders. These transponders receive commands, control loads, read and store metering data, and transmit metering and status data back to the data dispatch computer. Transponders respond to primary and secondary block address load control commands from the data dispatch computer. When used with separately mounted auxillary switching units, transponders are capable of controlling up to seven deferrable loads. Transponders, like receivers, are either mounted on a convienent surface or plugged into an existing power terminal meter socket. Several types of tranponders are offered, adding flexibility to the carrier system. Transponders vary according to the function desired, such as load control, three-part time-of-day metering, total energy meter reading, load survey, and rate experiment applications. Optional equipment including temperature monitors, pulse initiator adaptor, isolators, and distribution automation units car. 4-i added in an attempt to optimize the system's cost effectiveness ane strengthen capabilities. These devices will perform such functions as capacitor bank switching, voltage regulator tap changing, and distribution automation. General Electric. A complete power line carrier system available in uni and bi-directional form. Similar in function to the ASEP carrier system, the AMRAC carrier system is compriaed of a central control computer, a substation control unit, and a remote unit. The computerized central control center is capable of manual and automatic operation. In the automatic mode, commands are executed based on a time schedule. Manual operation allows the operator tQ specify either a single command or a file containing a list of commands to be executed immediately. The control computer include3 all the hardware and software necessary to perform load controi, automatic meter reading, and distribution automation functions. The control unit includes a minicomputer with disc and mag-type unit, a CRT terminal, and a line printer. Similar to the ASEP carrier system, the communication medium between the central control center and the substation control unit is a telephone line or microwave circuit. The substation control unit requires 1120-volt 60 Hz power and couples to the distribution feeti primary through a coupling capacitor-line tuner combination. -103- Communication between the utility's distribution substations and remote units is a flexible design utilitzing combined frequency- multiplexing and harmonic avoidance techniques. The ASEP system utilizes only one carrier frequency whereas the ARMAC system utilizes a group of eight different signal carrier frequencies. Receivers used in uni-directional control are either mounted on a convenient surface or plugged into an existing meter socket. Receivers are capable of controlling up to three electrical loads. For bi-directional applications, transponders are available only in socket mount configuration. They also control three loads and, in addition, are capable of remote meter reading (for time-differentiated rates or load survey), transmitting the metering data, and transmitting the status of six external contacts. With transponders located -at the meter, pilot wire must be installed between the transponder's control initiating output and appliance control relay. However, if that distance is great, receivers could be located at the appliance for control with trans- ponders remaining at the meter for communication. In this instance, with the transponders control power disengaged, both de,vices are assigned identical command addresses in or*der to function as single unit. The restoration period for both receivers and transponders can range from three to seventeen minutes. A distribution automation unit is also available that will perform such functions as feeder switch monitoring and control, capacitor bank switching, voltage regulator setting, and feeder redeployment. Telephone Control System Metretek, Inc. A complete bi-directional load control system using the existing telephone network as a communication medium. This system consists of a computerized central control, existing telephone lines, remote point module (which is built into the customer's meter), and a switching device. Central control contains a computer with diskette and disk drives, a CRT, and a printer. The computer also has a receiver subsystem with 2 to 8 receivers and telephone interface sets. In this system, a larger portion of intelligence resides at the meter. Communication is initiated by the remote module rather than interrogation from a central computer. Information is also stored in the remote module. The remote module, which is attached to the customer's existing telephone lines, calls the central location and transmits the required information. -104- Once the data is stored in the central computer, the receiver system will transmit new instructions to the meter. These instruc- tions include synchronizing the real-time clock, setting the new callback time, and any functional changes. Functional changes include converting from time-of-day meter to demand meter and changing load control -arameters. The remote module, wired to a switching device located at the appliance being controlled, is capable of controlling two loads, progressively. A variable duty-cycle control is available, which allows certain appliances, such as air conditioners, to be shut of for a preset percentage of time instead of being deleted. This duty-cycle is preset and not programmable from the central computer. This system, used mainly in residential applications, is capable of not only involuntary load control, but also voluntary load control such as time-of-day and demand metering. R+ybrii Sy'stem ciet"ifi: Atlanta. A uniq,te load management codcrol system ... ..- :th radiu n cvtr line catrrier technologies. The hybrid vs: a cnsisr o 2 t.-*;6a;z- genarato.r unit, en F.4 radis frequency transmnter, radio receiver/carrier current transmitter (RCX unit), and a power line carrier receiver switch. M:e message-eera:or unit (MGU) is th- central control- center of -he system originating a series.of load shed commands in accordance with ;reprogrammed instructions. The MGU is actually a microprocessor controller with an operator control panel and a CRT unit. The concroller's software program includes a maximum of twenty separate control groups (groups of water heaters, air conditioners, etc.). Each group is capable of being independently scheduled as a function f percent load shed, time-of-day, day of week, holiday, external sensing, or manual control. Thc 4W-11 cart implement a graduated load shed and rescoration progran as %ell as a SCRAM function. It can also be interfaced with a utility's SCADA system. The MGU coordinates communication to various sites by sequentially keying up to seven transmitters. The communica- tions medium is via a dedicated communications link, such as a leased telephone line. The commands received by the transmitter are in a digital message format. The transmitter will broadcast the messages to RCX units. The code structure of the message permits the MGU to separately address up to 32 RCX groups with a maximum of seven different command functions within each group. -105- The RCX unit converts only those radio frequency messages with proper address identification into a power line carrier format and then injects them onto secondary distribution lines. The power line carrier receiver switch, located at the point of control, is capable of controlling only one load. Its restoration period is approximately 7.5 minutes. Scientific Atlanta offers digital encoded radio receiver switches for use in smaller applications where the RCX unit would not be cost effective. The radio switch, located at the point of control, is available in either single or dual function. Its restoration period is also 7.5 minutes. By eliminating the use of the RCX unit the system becomes totally radio controlled. Scientific Atlanta also offers two other types of load control systems that can be used alone or in a hybrid fashion with the power line carrier medium. AM radio control and CATV (Cable television) control systems are of recent development with limited experience. Sine Wave Alteration Systems Emerson Electric, Inc. A complete bi-directional- load control system utilizing the power lines of electric utility's distribution network in an unconventional manner. This prototype system (TWACS), introduced in 1980, -operates . within the frequency range of the power frequency, for both signal transmission toward the consumer (outbound) and also toward the sub- station from the consumer (inbound). Use of the .existing power frequency range (usually 60 Hz) for communication implies excellent signal propagation without having to bypass calacitor banks, use-high power, or use repeaters as needed in ripple and power line carrier systems. System configuration includes a central control unit, a sub- station control unit, and remote transponder units. Each unit performs some degree of control and data processing. The computerized central control unit provides the control and data base management for the system. Communications with the substation control unit is via a dedicated voice grade channel. The microprocessor substation control unit is capable of: automatic initiation of schedule inbound and outbound tasks, polling and storage of remote transponder unit responses, error detection and correction, load command processing, through operator interface. The substation control unit controls an outbound modulation unit and an inbound data acquisition unit. All of which and including substation and modulation transformers, are housed in the utility's distribution substation. -106- Commands, generated by the central control, are injected and transmitted by the outbound modulation unit (under the timing control of the substation control unit) to remote transponder units. The inbound acquisition unit, upon instruction from the substation control unit, detects and processes information sent by remote transponders. The data is retained at the substation until requested by the central control unit. Microprocessor based remote transponders are mounted onto standard meter sockets. Transponders receive and perform requested operations and then transmit data back to the substation. Trans- ponders are capable of controlling up to six loads. TWACS is not only capable of load control but also load research, distribution automation, automatic meter reading, and time-of-day meter reading. GEC Measurements. The Cyclocontrol system is a complete uni-directional load management system utilizing the- utility's distri- bution network. It was originally developed and tested in England. The -Cyclocontrol system differs from typical carrier sPste"S because its signaling system relies on altering the shape of -he supply voltage wave form in the region of zero voltage. Once the signals have been impressed on a high voltage network at a requisite level, they -will propagate over considerable distances without signi- ficant attenuation. As compared to other conventional syste:as, Cyclocontrol has high signal reliability. The system's basic configu- ration, similar to a typical carrier system, consists of a micro- processor based control unit, a substation transmitter, and remote -receivers. The control unit utilizes a microcomputer with optional data printers, visual display units, and digital tape cartridge aad cassette recorders. The microcomputer is comprised of a central processor, memory, switch memory, and a programmer. The software capabilities of the microcomputer enahle it to be programed to respond to external inputs such as time of day, ambient temperature, and system demand. The control system monitors the power consumption of the distribution network and calculates predicted maximum demand. If load shedding is required then the substation transmitter is instructed to disconnect an appropriate number of electrical loads. The compunication medium between the control unit and substacion transmitter is via a multidrop telephone data link. The control system is adaptable and flexible and can be arranged to control existing load control installations of different philosophies such as ripple or power line carrier systems. -107- The transmitter, located at the substation, injects control commands through a standard distribution transformer onto .11 kV busbars. The transmitter is connected to the LV side of the trans- former by means of four cables and a standard fused isolator. A single transmitter is capable of signaling an entire 11 kV/415 volt or equivalent distribution system with fault levels of up to 300 MVA and above. The transmitter is capable of manual control in which a signal is initiated by pressing a button located on the control panel of the transmitter. However, this action is only adequate for smaller applications such as remote control of security lighting, fans, process and industrial plant. Single, double, and triple channel receiver relays are avail- able. Receivers incorporate mechanically latched contactors and an internal feedback circuit from the contactors contacts, so that interruptions to the power line do not affect the receive relay or cause the contacts to get out of step with the control signals. The signal coding system uses several marked zero crossing pulses in a 34 cycle period. This provides the system with high security from a false operation by noise. If any pulse is received in a wrong position, the receiver logic rejects the si-gnal and resets, ready for reception of repeat signals. COMMUNICATION AND INFORMATION SYSTEMS Communication and information systems utilize computer systems with communication, data acquisition and analyzing functions.requiring extensive software configuratibn. The following vendors of C & I systems are capable of providing the software programming required for each specific application. Multi-Building Systems TERA. The TERA Corporation designs the system and provides the software capabilities for a multi-building system. The information center is a minicomputer equipped with a databank consisting of the established electrical demand patterns for each facility in the multi-building system. The databank is developed by ESCO personnel, who inventories the electrical loads of each facility to identify their magnitude, timing, and priority in the event a load reduction was required. The computer also monitors existing energy demands of the multi- buildings, weather conditions, and utility power output and, in addition, will provide instructions to each remote terminal depending upon the circumstances. The software technology required for these types of computer functions is provided by TERA. The communication -108- medium between the computer and remote terminals is via a network of dedicated phone lines. Others. Other vendors are available who offer systems which can be designed to resemble a multi-building system. However, such systems are limited and incomplete because they lack the appropriate data acquistion capabilities and software technology needed for the proper implementation of a multi-building system. Vendors such as Robinton Products, Honeywell, and Barber Coleman offer systems which are similar in design to a multi-building system yet were developed for other load management purposes. Robiaton Products offer ALICE (Automated Load Information by Communication Exchange) originally designed as a remote revenue and load research metering system with limited load control capabilities. Honeywell and Barber Coleman offer the Delta series and the Econ VI, each a large energy management system capable of controlling several buildings through a complex computer controlled operation (both systems are discussed in greater detail in the energy management system section). If equipped with a proper database and software -configuration, these systems could be utilized in a multi-building SCADA (Supervisory Controls and Data Acquisition) Hitachi,-LTD. The SCADA system offered by Hitachi is based on a- three level hierarcy structure. The first level is th& central dispatching center which executes controls based on broad -decision making processes involving the total network and total operations of the electric utility. Controls are performed. in coordination with lower level control centers/stations (large scale), power stations, and high voltage transforming stations in an attempt to operate the power distributing network system in an efficient manner. The second level is the secondary dispatching center which controls its centralized control stations, (small scale) power -stations, and substations in accordance to the central dispatching center to improve the secondary network's operation. The third level is the centralized control station which controls the operations of its (small scale) power stations and substations and executes their resources management. The centralized control station operates in accordance to the secondary dispatching center. Large scaled Hitachi on-line process computers with CRT with printers, microcomputers, and operator consoles are used at each level. Typical functions of the SCADA system include automatic- frequency control, economic load dispatching, system monitoring, power flow calculations, system stabilizing control, and automatic network operations. -109- Harris Controls. This system, like Hitachi, uses a hierarchy system structure. By utilizing the concept of distributed tasks and multiple computers, system functions such as data acquisition, display handling, bulk data processing, or applications programs can be assigned to different computer levels. The system's structure is not designed into three distinct levels as Hitachi's but rather it is designed according to specific installa- tion requirements. In this system, the on-line computer performing the process functions is the controlling central processor unit. This CPU is responsible for configuring the system peripherals, CPU assign- ments, system status monitoring, failover, and restart and initializa- tion functions. At the secondary level universal, microprocessor-based control- lers are utilized. The functions of these controllers are determined by software downloaded to it by the host computers. It communicates, monitors the status, and converts the format of remote terminal units. Remote terminals perform the necessary information gathering and control functions. Overall, the system is capable of several load management func- tions such as economic dispatch, automatic generation control, load shed control, reserve monitoring, interchange scheduling and evalua- tion, and regulating margin dispatch. Siemens A. G. The INAUT 8-FW telecontrol system is a modular designed SCADA system. Controlled by the utility's control computer the 8-FEW interrogates r.emote receivers for load -information and commands remote relays to perform control functions. The 8-FW has Jhe capability to utilize all standard forms of information transmission, including telephone lines, power lines, and radio links. The 8-FW's modular design allows for easy expansion of both functions and end point control capabilities. THERMAL ENERGY STORAGE SYSTEMS Table A-3 contains a partial list of vendors of thermal energy storage systems. The following vendors of thermal energy storage systems generally offer equipment for space conditioning applications, especially space heating. However, certain systems offered (pressu- rized water storage and ACES) are capable of domestic water heating. These vendors also provide some type of manual control or thermostatic/time switch control with their equipment. However, thermal energy storage systems are capable of being controlled by a previously discussed local controller or remote control system. This would give the utility a more direct opportunity to control system peak demands. -110- Ceramic Heat Storage - Room Units AEG-Telefunken. Several high temperature- ceramic room heaters available, differentiated by their size, weight, and storage capacity. The smallest unit is 63H x 63L x 25D cm, weighs 108 kg, and has a maximum storage capacity of 16 kWh. The largest unic is 63H x 129L x 32D cm, weighs 309 kg,.and has a maxium storage of 49 kWh. A variety of control devices are also offered' for the operation of the room heaters. Control panels can be adapted to time of day rates, special storage heat rates, or demand limited rates. Up to 200 heaters can be controlled in a multi-zone configuration. Stiebel Eltron. Two types of high temperature ceramic room heaters available. One type is 63H x 24D cm and is available in varying length and weight (58 to 132 cm and 89 to 240 kg). The other type is a more compact size (48H x 30D cm) with models also varying according to length and weight (114 to 152 cm and 175 to 240 kg). Central control of the room heaters is available only in one type of device. This device can operate forty individual storage units. The flexibility of the system is found in the room thermostats. Three types are available with capabilities ranging from standard regulating features to temperature set back of the room. The temperature set-back clock thermostat can also control the set-back of'standard thermostats in other rooms. TPI Corporation. Six models of high temperature ceramic room heaters with varying dimensions. The smallest unit weighs 157 kg, has a storage capacity of 22.2 kWh, and measures 66H x 102L x 37D cm. The largest unit weighs 306 kg, has a storage capacity of 47.8 kWh, and measures 79H x 124L x 37D cm. The control panel is available in four different sizes depending upon the number of room heaters under control (15 heaters maximum control capability). A standard wall mounted room thermostat is used to control room temperature. An adjustable control thermostat, mounted on the heater, limits the amount of heat stored when a prede- termined temperature is reached. Ceramic Heat Storage - Central Units TPI Corporation. -Three types of high temperature ceramic central heaters available for residential and commercial/industrial applica- tions. One type, used for small residential applications, has a storage capacity of 112 kWh and weighs 771 kg. The heater is comprised of three units: a storage section, damper section, and blower. A larger type, used in residential applications is available in six different models. The storage capacity ranges from 140 to 200 kwh and they all weigh approximately 1360 kg. These models are divided into seven sections: a storage section, damper section, support plenum, two intermediate plenum sections, a night heat section, and an optional cooling cabinet (used for air conditioning purposes). The third type of central heater is used for commercial/ industrial applications and is available in four models. Storage capacity ranges from 140 to 200 kWh and they all weigh approximately 2041 kg. This unit is divided into three sections: storage section, damper section, and support plenum. A solid state charge controller is found in all heaters which automatically regulates the amount of stored heat in the unit as required for the next heating cycle depending on outside temperature and core temperature. Initiation of the charging period can be signaled by a utilities meter, a controller, or an energy management system. For large applications, several heaters can be installed with a central control -for charging. Heat flow is transferred through the facilities existing ductwork system. Pressurized Water Heat Storage Megatherm. A pressurized steel tank charged with treated water to a predetermined level capable of either domestic hot water heating or space heating. Water in the tank, which is permanent and uncirculated, is heated by immersion electric elements from a maximum of 1380C to a minimum required for heat exchange. Heat exchangers are inserted below the water level io heat the space heating or domestic water. Control of tank temperature is achieved by pressure controllers. A high limit temperature safety controller is used to shut off all power if maximum temperature is exceeded. Because the tank utilizes a closed system build up of mineral deposits on the heat exchanger and electrical elements is eliminated. The unit is capable of being operated by a time controller or energy management system. The tank can be located indoors or buried outdoors. Ice Cool Storage CALMAC. A cooling storage system utilizing solid ice for a facility's .cooling purposes. High density polyethylene, insulated, ice tanks make and store ice. The ice tanks use a standard package chiller with a conventional shell-and-tube evaporator and an automo- tive antifreeze brine to freeze ice solid in the ice tanks. Inside the tanks are equally spaced plastic tubing (rolled in a cylindrical shape) in which the brine is circulated through to freeze the water. -112- The modular ice tanks are available in two sizes. The smaller size weighs 1.8 metric tons (filled), 1.32 m in height, 1,2 m in diameter, and has a 32 metric con-hour capacity. The larger size weighs 2.1 metric tons (filled), 1.93 m in height, 1.2 m in diameter, and has a 48 metric ton-hour capacity. Tanks may be placed in or on the ground or on roofs, or inside in basements. In comparison to chilled water storage, the ice storage is smaller and can be factory made. Tfie equivalent cooling load in chilled water storage would require almost ten times the amount of space needed for ice tanks. In applications requiring large storage capabilities, multiple tanks can be utilized. Girton. Ice cool storage system utilizing a tank constructed of welded, hot rolled steel, with polyurethane insulation. . Tanks are equipped with forty serpentine pipe coils. Each coil has an indivi- dual thermal expansion valve and drier for each coil. Refrigerant is circulated throughout the coils to freeze ice. Each coil is mounted on a steel sheet to provide secondary refrigeration surface and to promote accelerated ice growth between the coil pipes during the storage phase. During the charging periods, the coils chill the water to 00 or 1.1oC, ice will then accumulate on the sides of the coils at a preset thickness of 2.5 to 7.6 cm. For cooling, chilled water is circulated .through the secondary equipment as a coolant, or is used to cool air. Tanks are available in twenty different sizes and can be used in multiple numbers for a single application. The smallest unit has a 45.5 metric ton-hour capacity, weighs 5.22 metric tonq_(filled) and has dimensions of 1.52 x 3.86L x 1.52W m. The largest unit has a 489.8 metric ton-hour capacity, wights 51.2 metric cons (filled), and has dimensions of 2.74H x 10.57L x 2.13 W m. Because of the similar ability to use multiple tanks in a single application and the additional ability to select from a larger number of tank sizes, Girton (in comparison to Calmac) provides a facility with a greater degree of flexibility in designing an efficient ice cool storage system. In addition to HVAC applications, Girton ice tanks can be used in the industrial and process industries (e.g., used for product cooling in the dairy, creamery, or food product industries). In-Ground Heat Storage Smith-Gates. Performed mats of resistance heating cable specifi- cally designed for embedment deep beneath the floors of-a single story commercial or industrial facility. -113- The cable is placed 30 cm beneath the bottom of a 15 cm concrete slab (45 cm beneath the surface of the floor.). The cable consists of a glass fiber core wrapped with nichrome wire, covered with 1059C DVC inner insulation and a nylon outer jacket, which is in turn wrapped with braided copper ground wire. Mats are available in ten different lengths, from 1.68L x .91W m to 19.5L x .91W m. Mats are also available in several voltage frequencies. A thermostat sensing bulb in conduit for controlling slab tempe- rature and a high temperature safety cutout taped to the center cable of a mat are both part of this system. The actual control is performed a time controller and room thermostats or a Smith-Gate's Selective Load Center. The Selective Load Center automatically allows the mats to be charged when enough power is available without exceeding a predetermined maximum amperage load. To insure maximum performance of the heating mats, the facility must be well insulated including insulating the inside of foundation walls-. Also, sand, used as a base for the mats, must be 30 cm in depth, well compacted, and free of stones land organic material. A plastic. vapor barrier is optional. Moving, water such as an under- ground spring or tidal flucation will hurt the heating system.'s performance. Any overhead door will require a 15 to 20 kW supplemen- tary heater. Combined Heat and Cool Storage Systems Combined heat and cool storage systems represent recently developed systems. However, equipment offered by vendors for these systems is generally a previous thermal storage system retrofitted for this particular application. The heat and cool system might even require a combination of retrofitted thermal storage equipment in which, usually, no single vendor offers. ACES. Annual cycle energy system is an applied system utilizing custom designed equipment and, as such, does not represent a commercially available system. Applications of ACES's are site-specific with no single manufacturer providing the complete packaged system. For example, an ACES project might utilize a specially designed Carrier water-to-air heat pump with an appropriately sized Girton or Calmac storage tank. DCES; Calmac. A daily cycle energy system has been developed by Calmac, by utilizing the heat produced from their existing ice cool storage tanks. When freezing water into ice, in. the ice tank, a certain amount of heat is given off into the chiller. As heat goes through the compressor on its way to the condenser it picks up additional heat, and, thus, heating is provided to the facility. -114- The DCES uses the same cooling technique and equipment as the ice cool storage system. The ice tank's dimensions and storage capabili- ties remain the same, the only difference being that it has been retrofitted to include heating capabilities. Girton. This daily cycle energy system also utilizes heat from existing ice tanks but in a much different manner than Calmac. Girton ice tanks are equipped with a heating mode for generating heat during the cool seasons. In the heating mode, heating elements are turned on automatically or manually. Water is recirculated over these elements continuously by a small low-head, high volume circulator to gradually increase the water temperature within the tank. During this opera- tion, the tank provides hot water through the main circulating pump, over to the air-handling unit coil and back. Girton utilizes the same tank for DCES purposes as it does for ice cool storage with the addition of a heating mode. The cooling technique is similar in both systems. Thus, all the various tank sizes and storage capacities available for the ice cool storage system are also available for the daily cycle energy system. SUPPLEMENTAL ENERGY SYSTEMS Supplemental energy systems potentially involve sizeable initial investments. With this type of outlay, extensivk feasibility studies should be performed to justify the initial and future costs (mostly applicable to the commercial and industrial -sector). Solar Energy Systems There are numerous manufactures of solar energy systems of all types throughout the World. Although a few vendors offer products to an international clientele, many are small firms which sell primarily to a domestic market. The solar industry has also been characterized by a. rapid turnover of firms. The products which are offered are diverse in type, design and quality. Prospective buyers should first determine the availability of systems and serve in their country. Cogeneration Systems Selection of an appropriate prime mover for cogeneration should be based on a number of characteristics: available heat recovery rate, electrical capacity, size range, and quality of thermal products. There are two types of cogeneration systems: bottoming and topping cycle systems. In bottoming cycles, waste heat is recovered from an industrial process and run through a waste heat boiler- generator set to produce electricity. In topping cycles, a prime- mover engine is coupled with a generator. Waste heat from the prime mover is recovered to satisfy thermal loads. -115- Bottoming cycle cogeneration systems and many topping cycle systems are custom designed and constructed to meet the design confi- gurations of the industrial process to which they are adapted. There are frequently multiple vendors for different components of the system. Topping cycles systems are far more common that bottoming cycles. A variety of engines including diesel, gas engines and gas turbines may be used as the prime mover. Cogeneration -- or self generation without waste heat recovery -- requires that the prime mover be coupled with an electric generator. Table A-4 presents manufactures of prime-mover generator sets. There are many other prime mover manufactures who do not sell prime mover-generators sets, but whose equipment could readily be adapted to cogeneration by coupting it with a generator. Dual Heating System Obtaining the necessary components required for a dual heating system depends on the type of heating system currently installed at the facility. A completely packaged system is not commonly offered by one specific vendor. The vendors discussed below both offer dual heating converting units that retrofit an existing fossil-fired system to a dual heating system. There are vendors, such as the Square D Company, that are currently evaluating the market potential of dual heating systems with considerations of offering a completely packaged-system. P.S.C. Controls. A dual heating unit is offered by P.S.C. Controls that is installed into the plenum section of the furnace. This unit will convert an existing oil-fired forced air heating system into a dual heating system. The system can be controlled manually or by an outdoor thermostat. Two models are available differentiated by the amount of power needed for the electrical heating mode (either 15 kW- model, which would adequately heat a 1,400 square meter building, or a 9 kW Model for smaller applications). - Electro Industries. A complete line dual heating conversion units is offered by Electro Industries. These units retrofit an existing fossil-fired heating system into a dual heating system. The Electro Boiler converts an existing water heating gas or oil boiler into a dual water heating system. This type of converting unit also applies to hydronics space heating. The unit is mounted adjacent to the existing boiler and is available in four different models with electrical power ranging from 13.5 kW to 27 kW. The Electro Mate converts an existing oil or gas-fired forced air heating system into a dual heating system. The system can be control- led manually, by thermostat, or by a remote control system. The converting unit fits into the plenum section of the furnace and is available six different models with the electrical rating ranging from 10 to 25 kW. -116- Table A-1 VENDORS OF LOCAL CONTROLLERS ComVanv TPC* ICU* AB Regin x x Kallered, Sweden ACEC X X Brussels, Belgium AEG-Telefunken - x X Frankfurt, Germany AMF Paragon Electric Co. X Two Rivers, WI, USA Advan-Oerlikon Ltd. X Bombay, India Allen-'radley Ltc. X Miton Keynes, UK Allen-Mortin Slectronics Ltd. x X Wolverhampton, UK Anritsu Electric Co., Ltd. X X Tokyo, Japan Automatisme-Contr8le Paris, France Avant Electronics Ltd. X Telford, UK B.M. Controls A/S I I Redensted, Denmark BBC Brown, Boveri & Cie AG X x Baden, Switzerland Sallmoos AG X X Horgen, Switzerland * TC: Time/Programmable Controllers. IC/U: Industrial Controls/Unclassified. -117- Table A-1 (Continued) Company TPC* ICU* Barber Coleman Co. x Rockford, IL, USA Bestobell Australia Ltd. X Crows Nest, N.S.W., Australia Bino Elprodukter AB X Farsta, Sweden Blichfedt Electronic APS X Ryesgade, Denmark Brillie (SCOP) x Montbrison, France Buh !Automatic X X Birkerod, Denmark CGE Distribution X X Paris, France C.I.T.;E.C. X Paris, France Chuo Electronics Co., Ltd. X Tokyo, Japan -Comsip Entreprise X Rueil, France Contraves AG X X Regensdorf, Switzerland Crouzet (S.A.) x X Valence, France DYNA X Paris, France Danfoss X X Nordborg, Denmark Dansk Klockner-Moeller A/S X X Kastrup, Denmark Datalogic France X X Rungis, France Deif A/S X X Kobenhavn N., Denmark . -118- Table A-1 (Continued) Company TPC* ICU* Diehl GmbH & Co. x Nuremberg, Germany Dieter Grasslin x St. Georgen, Germany Durr Ltd. I x Warwick, UK E.F.S. X Vernaison, France ELTAKO-Apparatebau -x Fellbach, Germany EUR-Control I X x Saffle, Sweden E,lan I x Neuss, Germany Elbau A/S x Bagsvaerd, Denmark Electrugrume automation I X La Verpilliere, France Elesta AG Electronik I Bad Ragaz, Switzerland Eltronic-Antriebstechnik X Gumlingen, Switzerland Erni & Co. Electro-Industrie x X Bruttisellen, Switzerland Eurotherm Automation I x Charbonnibres les Bains, France Faure(ets) X Floirac, France Fuji Electric Co., Ltd. X X Tokyo, Japan Fujitsu., Ltd. X Kanagawa, Japan GEC Automation & Control x Milperra, N.S.W., Australia -119- Table A-1 (Continued) Company TPC* ICU* Gama-Electronic X St. Etienne, France Girsberger Elecktronik X Eglisau, Switzerland Gorgy Timing X Paris, France Grossenbacher Electronik AG X St. Gallen, Switzerland Hitachi, Ltd. X X Tokyo, Japan Honeywell GmbH X x Offenbach, Germany Honeywell, Inc. X X Minnetonka, MN, USA IBO - Electronic K Meilen, Switzerland ISS Clorius International A/S X X Skovlunde, Denmark Indigel AG x Andelfingen, Switzerland Industrie Service X Calvire, France Ingeniorfiemaet Redan A/S X X Risskov, Denmark Instrumentation Ltd. X Kota, India Inter Control Herman Kohler Elektrik x Nuremberg, Germany Izumi Denki Corp. x Osaka, Japan Jameson Electric Priority Relay Findlay, OH, USA Japan Radio Co., Ltd. K x Tokyo, Japan -120- Table A-1 (Continued) Comoany TPC* ICU* Jasmin Electronics Ltd. x x Leicester, UK Johnson Controls International Inc. I x Milwaukee, WI, USA Kamstrup-Metro A/S I x Abyhoj, Denmark Klockner-Moeller Elektrizitats GabH X x Bonn, Germany Knudsen, Nordisk, Electricities Selskab I Kobenhavn Denmark Kontron Microcomputer GmbH x x Echning, Germany - Kosan Brunato A/S x X Copenhagen, Denmark L&J Industrie-Electronik GmbH X Alsbach, Germany LOG-ELEC x x Beure, France Logik A/S x Malov, Denmark MEC - Fabrica de aparelhagen Ind. LDA x X Sao Iria de Azola, Portugal MS Mecantronic System AG X Biel-Bienne, Switzerland IMerk Electronic x Bad Salzuflen, Germany Mi conic X X Uidigenswil Micro Matic A/S X X Odense, Denmark MicoControl Systems Inc. . Milwaukee, WI, USA -121- Table A-1 (Continued) Company TPC* ICU* Micropross X Lille, France Mitsubishi Electric Corp. X X Tokyo, Japan Nippon Electric Co., Ltd. X X Tokyo, Japan Oki Electric Industry Co., Ltd. X Tokyo, Japan Omron Tatsisi Electronics Co. X X kyoto, Japan P.M.I.C. X Lisieux, France PILZ GmbH &-Co. X Esslingen, Germany PSC Control Ltd. X Laval, Canada Pacific Technology I Kirkland, WA, USA Reliance Electric Ltd. X X Telford, UK Renaud K Reneins, France Robertshaw Controls Co. - X Richmond, VA, USA Roth-Electric GmbH X Gauting, Germany S.e.t.e.c.a. X Jarville, France Sandlom & Stohne AB K Stockholm, Sweden Satari Electric Co., Ltd. K Tokyo, Japan Satur Systems Ltd. I X Barnsley, UK -122- Table A-i (Continued) Company TPC* ICU* Selectron Lyss AG X Lyss, Switzerland Sepa Societa di elettronica per l'automazione I Turino, Italy Shibaura Electronics Co.. I Saitama, Japan Siemens S.A. I x Saint-Denis, France Sodeco - Saia AG X Murten, Switzerland Solidyne Corp. X x National City, CA, USA Spring - Electronik Zurich, Switzerland Staefa Control System x X San Diego, CA, USA Svenska Telemekanik A3 X Flen, Sweden Tabai Mfg. Co., Ltd. X Osaka, Japan Thiim A/S I x Herlev, Denmark Thorn Automation x X Rugeley, UK Tobu Electric Co., Ltd. X Tokyo, Japan Tork, Inc. I X Mount Vernon, NY, USA Toshiba Corp. X X Kanagawa, Japan Tour & Andersson AB I x Johanneshov, Sweden Trimax Controls x X Sunnyvale, CA, USA -123- Table A-1 (Continued) Company TPC* ICU* Weinlich Steverungen x Reilingen, Germany Wuest Electronik X X Weinfelden, Switzerland Xencon Inc. x San Farfael, CA, USA Yokogawa Electric Works, Ltd. x Tokyo, Japan Zander & Ingertrom AB X Stockholm, Sweden Ziegler x Langer, Germany -124- Altran Electronics Carson, CA, USA Radio American Science and Engineering, Inc. Cambridge, Mass, USA PLC ASEA Vasteras, Sweden PLC Brown Boveri and CIE Mannheim, Germany Ripple Emerson Electric Co. St. Louis, USA Sine Wave Alteration Enertec (Schlumberger) Chasseneil, France Ripple GEC Measurements Stone, UK Sine Wave Alteration General Electric Somersworth, NH, USA PLC Landis and Gyr San Jose, CA, USA Ripple Metretek, Incorporated Melbourne, FL, USA Telephone Motorola, Inc. Communications Group Schaumburg, IL, USA Radio Plectron Corporation Overton, Neb, USA Radio Regency Electronics, Inc. Indianapolis, Ind, USA Radio -125- RFL Electronics Wiltshire, UK Ripple Zellweger-Uster Ltd. Uster, Switzerland Ripple Scientific Atlanta Atlanta, GA, USA Hybrid Scientific Atlanta Horton, UK Siemens Aktiengesellschaft Munchen, Germany Ripple -126- Storage Space Heaters AEG - Telefunken Frankfurt, Germany AF-Energikonsult AB Stockholm, Sweden Airelec Radial Brunner Aubervilliers, France Bertin Mecanique de Prdcision Beziers,.France Buderus Aktiengesellschaft Herborn, Germany Creda France Paris, France Calmac Manufacturing Corp. Englewood, NJ, USA Emil Low KG Baden-Baden, Germany ERO(STE) Sorgues, France G. Bauknecht GmbH Stuttgart, Germany Giradot, s.a. Chaumont, France Kohl Calais, France Kuppersbusch Aktiengesellschaft Gelsenkirchen, Germany Malag-Werke Bretten Bretten, Germany Noirot-mt Paris, France -127- Storage Space Heaters (Continued) SensoTherm AB Sandviken, Sweden Megatherm East Providence, RI, USA Smith Gates Corporation Farmington, CT, USA Steibel Eltron Villacher, Austria Studsvik Energi teknik AB Nkyoping, Sweden Rochling - Metallwaren KG Nuremberg, Germany TeknoTern Energiteknik AB Bastad, Sweden TPI Corporation Johnson City, TN, USA Storage Water Heaters AEG - Telefunken Frankfurt, Germany A.T.E.L.E.C. Paris, France Deleage Saint Malo, France ELWA Elektro-Warme Munchen, Germany Franz Kaldewei GmbH & Co Ahlen, Germany Joh. Vallant GmbH u. Co Remscheid, Germany -128- Storage Water Heaters (Continued) Petitemange (ets) La3resse 88, France Robert Bosch GmbH Wernau, Germany Si.emens-Electrogerate GmbH Munchen, Germany Sciebel Eltron GmbH & Co KG Holaminden, Germany Termalec Paris, France Theodor Hettler GmbH Jungingen, Germany Floor Heaters I.G. Bauerhin GmbH Rothenbergen, Germany PSC Controls, Ltd. - Quebec, Canada Colorway Heiz Starnberg, Germany Kauffer & Co. GmbH Mainz, Germany -129- Appendix B APPROACH TO EQUIPMENT COST ASSUMPTIONS FOR REMOTE-CONTROL TECHNOLOGIES Equipment costs for Tables 3-5 and 3-6 are estimated as the sum of the costs of the various types of equipment making up the total load management system. Each cost component is calculated as the product of the number of items of specific types of equipment times the price of the equipment. This price is adjusted for applicable quantity discounts (see Table 3-4). Cases differ in cost because different numbers of equipment or equipment with different capabilities (and prices) are required. Prices (in U.S. dollars) are typical of those prevailing for currently-available equipment, but do not represent the prices of specific vendors. Prices do not include delivery costs, installation costs or financing charges, all of which will vary with the country. Numbers of equipment types are typical of utility practice for the speci- fied case and number of control points. Prices and numbers of equipment in an actual load management program may differ from those indicated, reflecting vendor price quotes and7utility-specific considerations. Example Calculations Tables B-1 through B-3 provide an ekample of the calculations which were performed for Power Line Carrier (PLC) systems of 100,000 control points. Table B-1 specifies the utility system assumptions for the various case. (Other assumptions were specified in Table 3-3.) Table B-2 presents the cost buildup for all six cases. Table B-3 shows percent changes from the base case results. In this example, the following typical assumptions were made: * Central Control. In the base case, central control consists of a computer and associated peripherals and communications equip- ment. In the sophisticated and high-cost cases, two more capable computers are used to handle increased numbers of func- tions and to increase reliability through redundancy. * Communications. One substation control unit is sited at each substation. A tranceiver at the substation is required for 2-way PLC communications between the substation and the point of control. Under adverse conditions, a repeater is. required for every ten miles of distribution line. * Point of Control. The base case consists of a receiver/switch with one control function relay located at each point of control. In the sophisticated case, four functions are control- led. Hence, more expensive receiver/switches with four relays are required. In the bidirectional case, the tranceiver/switch at each point of control must communicate back to its substation. -130- Table B-1 UTILITY SYSTEM SPECIFICATIONS BASE CASE - URBAN Control Points; 100,000 Area Covered; 100 Points: 1,000 MVA Load: 300 MVA Load/Point: .003 Substations. 20 Points/Substation: 5,000 MVA Load/Substacion 15 Points/Line 63 Transmission Line: 100 Distribution Line: 1,500 -131- Table B-2 EQJIFeM1 CDSTS OF PLC IDAD KAGENT SYS'IS 100,000 CONcM Fols (1982 US$) Case Adverse Bidirec- Sohis- Cost Elenent Base Dispersed Conditions tional ticated High Cost CDNORCL URCTOWS 1 1 1 1 4 4 CNAL CaTIROL Coputer Unit Cost 30,000 30,000 30,000 30,000 40,000 40,000 No. of Units 1 1 1 1 2 2 Cost of Units 30,000 30,000 30,000 20,000 80,000 80,000 btal 30,000 30,000 30,000 30,000 80,000 80,000 Ca4NICATMDS Substation Unit Cost 5,000 5,000 5,000 15,000 5,000 15,000 Control Unit 1b. of Units 20 35 20 20 20 35 Cost of Units 100,000 175,000 100,0001 300,000 100,000 525,000 Distribution Unit Cost 0 0 0 0 5,000 5,000 Automation No. of thits 0 0 0 0 20 35 Units Cost of Units 0 0 0 0 100,000 175,OOQ Repeaters unit Costs 0 0 2,500 5,000 0 2,500 No. of Units 0 0 150 20 0 2,100 Cost of Units 0 0 37,500 100,000 0 525,0000 Total 100,000 175,000 475,000 300,000 200,000 595,000 POn OF CNMROL Receiver/ Unit Cost 90 90 90 0 120 0 Switches No. of Units 100,000 100,000 100,000 0 100,000 0 Cost of Units 9,000,000 9,000,000 9,000,000 0 1,200,000 0 Transceivers/ Unit Cost 0 0 0 170 0 300 Switches No. of Units 0 0 0 100,000 0 100,000 Cost of units 0 0 0 17,000,000 0 30,000,000 Total 9,000,000 9,000,000 9,000,000 17,000,000 1,200,000 30,000,000 TOTAL SYSTE CDSTS 9,130,000 9,205,000 9,505,000 17,600,000 13,000,000 36,750,000 C)STS PER POINT 91 92 95 176 130 367 -132- Table 3-3 PERCENT CHANGE FROM BASE CASE COST: RIPPLE Points of Control: 100,000 Percent Change Base Case Dis- Adverse Bidirec- Sophis- High Cost Element Cost persed Condition tional tickted Cost CENTRAL CONTROL 310,000 0 0 900 2,567 2,567 COMMUNICATIONS 100,000 75 375 200 100 585 POINT OF CONTROL 9,000,000 0 0 89 33 230 TOTAL SYSTEM COST 8,537,000 0.8 4.1 92.8 42.4 302 ENERGY DEPARTMENT PAPER SERIES EGY PAPER No. 1 Energy Pricing in Developing Countries: A Review of the Literature by DeAnne Julius (World Bank) and Meta Systems (Consultants). September 1981. 121 pages, includes classified bibliography. Reviews literature on the theory of exhaustible resour- ces and on sectoral, national and international models for energy demand. Emphasis on project selection cri- teria and:on pricing policy as a tool of -energy demand management. EGY PAPER No. 2 Proceedings of the South-East Asian Workshop on Energy Policy and Management edited by Michael Radnor and Atul Wad (Northwestern University). September 1981. 252 -pages. Contains the edited version of the lectures and discus- sions presented at the South-East Asi#n Workshop- on Energy Policy and Management held in Daedeok, South Korea, October 27-November 1, 1980. Topics that are addressed inaclude: the overall problem of energy policy and its relationship-to economic de- velopment; the-management of energy demand and related data; the role and value of models in energy planning, and the use of energy balances. Transport and rural sectors are also discussed in terms of their relation- ship to energy planning. EGY PAPER No. 3 Energy Pricing in Developing Countries: Lessons from the Egypt Study by DeAnne Julius (World Bank). December 1981. 14 pages. Study on the effects of energy price change in a devel- oping country. Provides insight into the mechanisms through which energy prices affect other prices in the economy and, therefore, the incomes of rich and poor consumers, profitability of key industries, the balance of payments, and the government budget. EGY PAPER No. 4 Alternative Fuels for Use in Internal Combustion Engines by G.D.C., Inc. (Consultant). November 1981. 179 pages, includes appendices. Presents several alternative fuels used as replacement for conventional (gasoline and diesel) fuels in inter- nal combustion engines. These alternatives, including LPG, natural gas, alcohol and producer gas, are deriv- able from natural resources that exist in so many de- veloping countries. Also provides up-to-date informa- tion on the newest alternative fuel option currently available and those that are being developed and tes- ted. EGY PAPER No. 5 Bangladesh: Rural and Renewable Energy Issues and Prospects by Fernando R. Manibog (World Bank). April 1982. 64 pages, includes bibliography. Analyzes subsector issues and recommends courses of action for energy project possibilities; identifies renewable energy projects which could create a positive impact in the short to medium term. EGY PAPER No. 6 Energy Efficiency: Optimization of Electric Power Distribution System Losses by Mohan Munasinghe (World Bank) and Walter Scott (Consultant). July 1982. 145 pages, includes appendices. Discusses the reasons for high existing levels of power distribution losses in developing countries. Identi- fies areas within a power system where loss optimiza- Cion would be most effective. Shows that reducing losses is often more cost effective than building more generation capacity. EGY.PAPER No. 7- Guidelines for the Presentation of Energy Data in Bank Report October 1982 - 13 pages (ncl. 4 Annexes). Masood Ahmed (World Bank). The growing importance of energy issues in national economic management has led to increased coverage of the energy sector in many types of reports. However, there is still no clear, consistent and standardized format for presenting energy sector information. This paper reviews the problem and proposes guidelines for policymakers and operational staff who deal with energy issues. The paper is divided into three parts: part one sets out the basic framework for presenting aggreg- ated energy data - "the national energy balance"; part two deals with the use of appropriate units and conver- sion factors to construct such a balance from raw de- mand and supply data for the various fuels; and part three briefly discusses special problems posed by: (i) differences in end use efficiency of various fuels; (ii) the inclusion of wood and other noncommer- cial energy sources; and (iii) the conversion of pri- mary electricity into its fossil fuel equivalent. EGY PAPER No. S External Financing for Energy in the Developing Coun- tries by Althea Duersten (World Bank). June 1983. 66 pages, includes appendices. Provides an overview of energy financing in the devel- oping countries. Identifies energy.investment require- ments and past financing patterns. Discusses the his- toric roles of multilateral and bilateral assistance programs in helping to mobilize financing, particularly for low income oil importers and in providing economic and sector advice. Examines the role of official ex- port credit, and discusses lending by private financial institutions which has been the predominant source of financing for energy projects in the middle and higher income developing countries. EGY PAPER No. 9- Guideline for Diesel Generating Plant Specification and Bid Evaluation by C.I. Power Services Inc. (Consultant). December 1982. 210 pages, includes appendices. Explains the characteristics and comparative advantages and disadvantages of large low speed two-stroke diesel engines intended for electric generating plant service, and develops a bid evaluation procedure to permit com- paring of bids for both types. EGY PAPER No. 10 Marginal Cost of Natural Gas in Developing Countries: -Concepts and Application by Afsaneh Mashayekhi -(World Bank) July 1983. 21 pages, includes -appendices. Defines the concept of marginal cost and average incre- mental cost. Uses the detailed supply, demand and investment data to apply this concept to estimate the average incremental cost of natural gas supply to major markets in ten developing countries. Demonstrates that the cost of natural gas delivery to the city-gate in many developing countries is far below the cost of competing fuels.