THE RELATIONSHIP BETWEEN ONSITE GENERATION AND

Chapter V

THE RELATIONSHIP BETWEEN

ONSITE GENERATION ANDCONVENTIONAL PUBLIC UTILITIES

Chapter V.–THE RELATIONSHIP BETWEEN ONSITEGENERATION AND CONVENTIONAL PUBLIC UTILITIES

Page

Background . . . . . ........................139

Electric Transmission and Distribution .. ....141

Thermal Transmission and Distribution . . . . . .143

Sale off PowerGenerated Onsite . ...........144Metering ....., . . . . . . . . . . . . . . . ... ... ...144Safety . . . . . . . . . . ......................144Quality of Power Fed Back to Utilities .. ...145Local Distribution Capacity .. .. ... ... ....145Economic Dispatch. . . . . ................145

Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....146Time-of-Day Pricing. . . . .................146Storage Strategy . ......................147Load Management. . ....................147Offpeak Electricity . ....................148

Ownership . . . . . . . .......................148Utility Attitudes . . . . . . . . ................150

The Cost of Providing Backup Power From anElectric Util ity . . . . . . . . . . . . . . . . . . . . . . . . . . .151

Backup PowerWithEnergy SourcesOtherThanElectricity . . . . . . . . ...................159

AppendixV-A.– Analytical Methods. . . .. ....161

LIST OF TABLES

Table No. Page

l. A Comparison of the Cost of Transmittingand Distributing Energy in Electrical,Chemical andThermal Form . . . . . . . . . . . .

2. Construction Expenses of ElectricCompanies . . . . . . . . . . . . . . . . . . . . . . . . . . .

3. District Heating Systems . . . . . . . . . . . . . . .4. The Capital lntensityof Major U.S.

Industries . . . . . . . . . . . . . . . . . . . . . . . . . . . .5. The Fractional Difference Between the

Utility Costs[¢/kWh] Required to ProvideBackup Power to the Systems Shown andthe Costs to ProvidePower to a ResidenceEquippedWith an Electric Heat Pump . . . .

6. Levelized MonthlyCosts for a Well-lnsu-lated Single Family HouseShowing theEffect of Marginal Costing for BackupPower . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

140

142143

149

153

154

Table No. Page

7. Ratio of Price Utilities Can Pay forSolar Energy Generated Onsite to thePrice Charged by Utilities for Electricity. ..155

8. The Fractional Difference Between the UtilityCosts Required To Provide Backup Power tothe Systems Shown and the Costs ofProviding Power to a High Rise ApartmentEquipped With Central Electric Air-Condi-tioning, Electric Hot Water, and BaseboardResistance Heating. . ..................156

9. The lmpact of Offpeak Storage on Utilitycosts . . . . . . . . . . . .....................157

IO. Levelized MonthlyCosts for a Single FamilyHouse Showing the Effect of MarginalCosting and BuyingofOffpeak Power ....158

11. Leveiized MonthlyCosts of Several Kinds ofEnergy Equipment in a Single Famiiy De-tached Residence in Albuquerque, N. Mex.160

A-1. Average Characteristics of PrivatelyOwnedUti l i t ies in 1974 . . . . . . . . . . . . . . . . . . . . . . .163

A-2. Numbers of Buildings in “Typical’’UtilityModeled WhichUseHeating and Coolingand HotWaterEquipment . .............163

A-3. industrial Load Profile Used..... . ......164A-4. Characteristics of Equipment Used in the

Ut i l i ty Mode l . . . . . . . . . . . . . . . . . . . . . . . . . . 170A-5. Assumptions About the Nonconventional

Systems Used in the Utility lmpactAnalysis . . . . . . . . .....................170

LIST OF FIGURES

Figure No. Page

l.improvement in the Pattern of ElectricityConsumption in England’s SouthwesternElectricity Board Resulting From a Shiftto Time-of-Day Electricity Rates . .........146

A-1. A Typical Load-Duration Curve for anE lec t r ic Ut i l i t y . . . . . . . . . . . . . . . . . . . . . . . . . 162

A-2, Ratio of PeakDemands to AverageDemands as a Functionofthe StandardDeviation of ’’Wake-Up’’Times . ..........165

A-3. Hourly ’’Miscellaneous” Electric LoadProfile for Single Family House WithDifferentTypes of Diversity . .............166

A-4. inverted Load-Duration Curve of aTypical Electric Load. . . . . . . . . . . . . .......168

Chapter VThe Relationship Between Onsite Generation

and Conventional Public Utilities

BACKGROUND

The value of onsite energy systems cannot be addressed by examiningtheir performance as isolated systems. While it is possible to construct on-site solar devices capable of operating with no connection to other powersources, it is seldom economically attractive to do so when other sources ofpower are available.

It is tautological that designing an optimum approach for providing energyin a given region requires that all equipment for installing and consumingenergy be considered as components of a single integrated energy systemdesigned to meet a fixed set of energy demands: maintaining building in-teriors at comfortable temperatures, providing lighting, supplying heat for in-dustrial processes, etc. Any attempt to simplify the problem by consideringthe capabilities of components in isolation must result in a less efficient out-come. Moreover, it is likely that without taking this synthetic perspective,some critical aspect of the overalI system will be neglected.

Performing this kind of analysis is difficult because of the complex andhighly interdependent energy systems which have emerged over the past fewdecades, the variety of equipment which is currently in use, and the bewilder-ing variety of devices now u rider test and development. This chapter providesa perspective on some of the major issues confronted in integrating onsitesystems into larger systems for supplying energy and provides the basis formaking realistic estimates of the cost of operating onsite equipment.

The design of an optimum energy networkwhich includes onsite solar facilities re-quires choices in the following areas.

. How much of the backup energy s h o u I dbe supplied from energy storage equip-ment, and how much backup e n e r g yshould be supplied from convention Ienergy sources? (This usually transIatesinto determining the opt i m u m size foronsite storage equipment )

●

●

●

140 ● Solar Technology to Today’s Energy Needs

If it is possible to transmit energy inexpen-sively, overall energy costs can usually bereduced by connecting together as manyenergy consumers as possible. If onsite gen-erating systems are not connected, eachgenerating unit (solar plus backup) would

have to be large enough to meet the peakdemand of the building or industrial processwhich it is designed to serve. The load fac-tors of individual buildings are usually veryunattractive (see table V-1 ). Onsite load fac-

Table V-1 .—A Comparison of the Cost of Transmitting andDistributing Energy in Electrical, Chemical, and Thermal Form

Capitalcost

$ 9 2 / k W (4)--

NOTES: I1A ~apltal ,ecoveV factor Of O . I 5 is used to calculate annual caPital

charges.2Assumes a capacity of 2,500 MW Ior one 765 kv line.

3Utility construction expenditures of $1,7 billion in 1975 for 3,762 ad.ditional miles or S461,000 per mile average. (Statistlca/ Yearbook offfre Electrlc Utl//ty Industry for 1975, Edison Electric Institute, NewYork, N.Y , Oct. 1975.)

4 The 1970 Natlona/ Power Survey, Federal power Commiss ion ,

Washington, D. C., p. 1.13.8, Dec. 1971.51nvestor.owned electric utilities spent approximately $851J million on

transmission O & M costs in 1975 for 1.5x10’Z kWh. (S fahstlca/ Year.book of the E/ectrlc Utjllty /ndustry for 1975).

6Assume$ an end.use efficiency of 100 percent7Nat,ona/ Gas Survey, u,S Federal Power COmmisS.iOn, VOI. 1, P. 34.

1975.8AII natural gas companies spent $53 I million in 1976 for 1,845 miles

of new transmission pipeline or $287,000 per mile average. (7976 GasFacts, American Gas Association, Arlington, Va., 1977).

9 Four percent of total natural gas consumed was used for PiPeilnefuel In 1976 This is equally allocated to transmission and distribu.tion. (AGA Gas Facts).

IONatlona/ Gas Survey, U.S. Federal Power Commission, VOI. Ill, P. 129,1973.

11 All natural gas Utllilles spent $1 1 billton in 1976 on o & M fortransmission for 148 trillion cubic feet (TCF).

12ASSumeS end.use efficiency Of 65 Percent

13tLResidential Energy Use to the Year 2(toO: C o n s e r v a t i o n a n d

Economics,” Oak Ridge National Lab, Report ORNIJCON.13, OakRidge, Term., Sept. 1977,

141nyestor.owned electric utilities spent S2.8 bdlion on COn StrlJCtiOfl Of

d i s t r i b u t i o n f a c i l i t i e s f o r 2 1 , 7 0 0 kW of new capac i ty in 1975.(Statlshcal Yearbook of the E/ectr/caf Ut///fy /rrdustry for 7975.)

151nvestor.owned utilities spent approximately $1.59 billion in 1975 for

distribution O & M costs for 1.5x10’2 kWh. (Statistlca/ Yearbook ofthe E/ectr/ca/ Utl/lty Industry Iof 7975.)

16 Calculated from the average cost of S400 per customer (Privatecommunication—American Gas Association) with an average hotwater and space heating requirement of 36.4x10] kWh (121 MCF) peryear and an assumed load factor of 0.4.

17AII natural gas utilities spent S1.1 billion on distribution 0 & M In1976 for 14.8 TCF (AGA Gas Facfs),

18 See volume Il.19 f1Evaluation of the Feasibility for Widespread Introduction Of COal in.

to the Resldenttal and Commercial Sectors. ” Exxon Research andEngineering Co., Linden, N. J., Vol. 11, p. 6.11, Apr. 1977. These twostudies give a construction cost of about S14 milllon for the sizesystem in question, which requires a peak capacity of 70,000 kW, asshown by reference in note 20.

20 Annua\ o & M costs are calculated by assuming they are 3 Percent

of capital costs, which is the average of the percentages for gas andelectric systems.

21 The cost of energy lost in transmission was estimated usln9

0.04c/kWh for electricity y and 1.5c/kWh for thermal energy,

Ch. V The Relationship Between Onsite Generation and Conventional Public Utilities ● 141

tors can be as low as 15 percent, while typi-cal utiIity load factors are 50 to 60 percent.Moreover, each onsite facility would haveto provide enough redundant equipment toachieve acceptable levels of reliability. If aninterconnection is avaiIable, however, it isnecessary only that the combined output ofall generating units in the system be able tomeet the aggregate peak demand of the re-gion. The aggregate peak will be lower thanthe sum of the individual peak demandssince individual peaks will occur at differenttimes (this is usualIy called “diversity” in thedemand). The advantage of the connectionis magnified by the fact that most gener-ating devices operate less efficiently whenoperated to meet an uneven demand. Inter-connections also permit greater freedom inselecting generating and storage equipment(onsite and centralized facilities can beselected as they are appropriate), and it iseasier to optimize the efficiency of the totalsystem throughout the year by controllingthe performance of each system in the net-work in response to the total load.

The problem of uneven loads is a par-ticularly difficult one in the case of electricutiIities since generating and storage equip-ment tends to be extremely expensive, al-though chemical and thermal transport sys-tems can also benefit from balanced loads.It may prove feasible, for example, to pipehot water generated in collectors located ona number of separate buildings to a centralthermal storage facility, If this storage facili-

ty is large enough, collectors need only havean annual output large enough to providefor heating and hot water requirements andstorage losses. Such systems may requireless collector area per building unit thanconventional solar heating and hot watersystems using relatively small amounts ofstorage.

While connecting energy generating andconsuming devices into a single energy net-work can lead to significant savings, thetransmission and distribution systems re-quired can be extremely expensive. Thecosts of several types of energy transportsystems are summarized in table V-1. Com-parisons of this type can be somewhat mis-leading because costs will vary greatly fromsite to site, but the table at least allows acrude ranking of alternatives. It indicates,for example, that transporting energy inchemical form is by far the least expensiveapproach. It is interesting to notice, how-ever, that distributing energy in the form ofhot water over distances of 1 to 2 miles is on-ly about 30 percent more expensive thantransmitting electrical energy over typicaldistances from generating facilities to con-sumers, I n this comparison, no attempt wasmade to share the cost of the trench dug fort h e h o t w a t e r p i p e w i t h p o t a b l e w a t e r ,sewer, telephone, or other Iines which couldbe placed in the excavation, The electricdistribution costs would have been signifi-cantly higher if buried cables were used.

ELECTRIC TRANSMISSION AND DISTRIBUTION

Electric transmission and distribution isan expensive undertaking. Over half of thecapital invested by electric utilities in theUnited States is invested in a massive net-work of transmission and distribution equip-ment. In recent years, the ratio betweencapital investment in electric generating,

IStat/st/cs of Pr/\ate/y Owned E/ectr/c Ut///eses in theUn/ted States, Federal Power Commlsslon, 1974, p 40

transmission, and distribution equipmenthas fallen because of the rapid increase inthe cost of generating plants [see table V-2),Each dollar now invested in generatingequipment is accompanied by a 16-cent in-vestment in transmission equipment and a23-cent investment in distribution systems,The high cost of the electric transmissionand distribution system is due in part to thefact that the lines have relatively low load


Table V-2.—Construction Expenses of Electric Companies

1965 1967 1969 1971 1973 1975 1977

Total Investment (billions ofdollars) . . . . . . . . . . . . . . . . . . . . . 4.03 6.12 8.29 11.89 14.91 15.09 19.50

Production equipment (%). . . . . 32.3 41.7 46.1 56.3 58.9 65,1 68.3

Transmission equipment . . . . . . . 23.3 21.5 18.7 15.2 13.7 11.5 10.7Distribution equipment. . . . . . . . . 39.4 32.3 29.2 23.3 22.6 18.7 15.7

Other . . . . . . . . . . . . . . . . . . . . . . . . 5.0 4.4 4.0 5.1 4.8 4.7 5.03

SOURCE. EBASCO 1977 Business and Economic Charts (Ebasco Services, Inc., New York, NY. )

factors. Hot water and steam distributionsystems used for heating buildings typicallyhave load factors of 33 percent (see tablev-1 ).

The electric transmission and distributionlines now in place are typically 90-percentefficient, but it is hoped that improvedtechnology will lead to overall efficienciesof 92.5 percent by the year 2000.2 The im-proved efficiency, however, will most likelyadd to the capital cost of the systems. Gaspipelines are typically 90-percent efficient,the losses being due primarily to the factthat gas is withdrawn to run pumps in thepipeline. 3 No comparable information isavailable on the efficiency of hot water andsteam transport systems. Losses are a strongfunction of the ground and fluid tempera-tures, the distances traveled, and the aver-age size of the pipelines. Some recent steamdistribution systems, however, have experi-enced losses on the order of 15 to 20 percentand have created serious difficulties for thesystems relying on them.4

‘ERDA-48, Volume i, pp. B-10 and B-n, 1975.‘American Gas Association.‘American Public Power Association, private com-

munication, November 1977

The cost of maintaining transmissionequipment can also be substantial. In 1974,the cost of operating and maintaining thenetwork of electric transmission and distri-bution lines owned by privately owned util-ities in the United States exceeded the costof operating and maintaining the generatingfacilities (fuel costs excepted). ’ Annualmaintenance costs for a smalI hot water dis-tribution system can amount to about 3 per-cent of the initial capital cost of the equip-ment.6 Maintaining steam systems mayprove to be significantly more expensive.

In addition to direct costs, transmissionIines can have serious environmental conse-quences. It is estimated that over 3 millionacres wilI be required for new transmissionlines by 1990. ’ Much of this constructionwill occur in scenic areas where oppositionis likely. In addition, it is possible that thelarge electric fields produced by high volt-age transmission lines may be harmful. Thequestion is being investigated and the re-sults are inconclusive at this time.

‘Statistics of Privately Owned Electric Utilities in theUnited States, op. cit , pp. 35, 39,40, 42,43.

‘W. R. Mixon, et al., Technology Assessment ofModular Integrated Ut;//ty Systems, Oak Ridge Na-tional Laboratory, ORNL/HUD/MIUS-25, pp 3-189 and3-190.

7Thermo-E Iectron Corporation.

Ch. V The Relationship Between Onsite Generation and Convention/ Pub/ic Uti/ities ● 143

Table V-3.—District Heating Systems

District Heating Systems in the United States

44 city systems

Total steam sold (millions of pounds). ., . . . . . . . . . . . . . . . . 84,246

Total steam delivered to system (millions of pounds) . . . . . . . . . 96,672

Annual system load factor . . . ., . . ., ., . . . . . . . . . . 33%

Number of customers (in thousands) . . . . . . . . . . . . . . . . . . . . . . . . . 14,903

Length of distribution system (in miles) . . . . . . . . . . . . . . . . . . . . . . . . 573

Installed air-conditioning (tons) . . . . . . . . . . . . . . . . . . . . ... , . . . . . . . 666,051

New York

32,702

38,469

37‘A

2,514

100

569,945

Note New York City has the largest American system, selling nearly 10 billion kWh per year The load Is 30 percent residences. 45 percentoffice bulldings, 11 percent industrles and 13 percent institutions.

SOURCE Offlcial Proceeding, 80th Annual Meetlng of the International District Heating Asscciation, June 1969, pp. 22-30, quoted on p 24of ORNL-HUD-19, Ibid

District Heating Abroad

Sweden Denmark W. Germany U.S.S.R.(1973) (1973) (1973) (1971 )

Energy sold for district heat(millions of kWh) ... . . . . . 12 14 38 1,100,000

Units connected . . . . . . . . . . . . . . . . . . 600,000 30% of dwellings 83,000 75%

SOURCES I G C Dryden, The Eff/c/enf Use of Energy, IPC Science and Technology Press, 1975, p 359

Quoted In Teploengetika, Vol 18, No 12, 1971, pp 2-5

W Hausz, and C. F Meyer, Energy Conservation Is the Heat Storage Well the Key?” Puh/Ic Uf,/,tfes Fortn,ghfl} APrIl 24 1975

THERMAL TRANSMISSION AND DISTRIBUTION

Most hot water and process steam is con-sumed close to where it is generated, but anumber of cities have systems for distribut-ing steam to residences and industries. I nthe United States, many older systems havebeen abandoned, and few new systems arebeing built. The Modular Integrated UtilitiesSystem (MI US) in Jersey City, N. J., is one ofthe few recent exceptions to this trend. Theabandonment of steam distribution is duemostly to the overalI decline of onsite gener-at ion. Th is tendency is re in forced by thefact that many older steam-distribution sys-tems were not designed to return water tothe generating plant, as the turbines oper-ated on untreated water. This procedure be-came impractical with the addition of large,

high-pressure, steam-generating facilities re-quiring expensive water purification sys-tems. Table V-3 estimates the capacities ofdistrict heating in the United States andabroad. On the other hand, district heatinghas been used much more extensively inEurope. For example, 30 percent of the resi-dences in Denmark are connected to districtheating systems. Sweden estimates that 70percent of its multifamily units and 20 per-cent of its single family homes will be con-nected to district heating systems by 1980.West Germany plans to provide district heatto 25 to 30 percent of its dwellings by 1980,8

81 G C Dryden (ed ), The Eff/c;ent Use of Energy, IPCScience and Technology Press, 1975, p 358.

144 ● So/ar Technology to Today’s Energy Needs

The feasibility of using district heatingsystems depends much on the density ofdwelling units. A study conducted in con-nection with the MIUS project est imatedthat a garden apar tment complex in theP h i l a d e l p h i a a r e a ( 6 0 b u i l d i n g s w i t h 1 2apar tment un i ts per bu i ld ing) would costabout $410 per unit for heated water dis-t r i b u t i o n a n d $ 3 3 0 per un i t fo r ch i l l edw a t e r , 9 A MIUS system was actual ly in-stalled in Jersey City, NJ., for approximatelythis amount.10 A preliminary estimate of thecost of a large district heating system capa-

9G Samuels, et al , M/US Systems Ana/ysis, /n/t/a/Cornpar/sons of Modular Sized /rrtegratecf Ufi/;ty Sys-tems and Corrverrt;ona/ Systems, Oak Ridge NationalLaboratory ORNL/HUD/MIUS-J une 6, 1976, p 144

IOPaul R Achenbach and John B Coble, Site Ana/-ys;s for the App/;cat;on of Total Energy Systems toHous/ng Developments.

ble of serving a community of 30,000 peoplein a mixture of apartments and single familyunits indicates that the cost per unit for thisdispersed system would be nearly threetimes as great, The ability of a system toamortize these costs depends on the cost ofthe energy supplied and the yearly energydemand of each building. A rule-of-thumbapplied until recently in West Germany wasthat a “break-even” housing density was onewhich required 44 MW t/km 2 for existing ur-ban areas and 28 MW t/ k m2 for new devel-opments. Recent increases in fuel prices,however, have led them to consider areaswith demands as low as 14 MWt/ k m2. Thegarden apartments in the MIUS study had ademand of approximately 30 MWt/ km2.

1‘1 G.C Dryden, op clt , p 251

SALE OF POWER GENERATED ONSITE

METERING

No major technical barriers need to beovercome in designing meters for onsiteenergy equipment, but new types of metersmay have to be developed for this purpose.If a utility owns onsite equipment with ther-mal output, some technique must be foundfor billing customers for the energy pro-duced by the onsite device. One utility hassuggested that the simplest technique wouldbe to bill a customer on the basis of actualcapital and maintenance costs, althoughmeters capable of measuring the energygenerated by solar thermal systems ofvarious sizes are available. 12 13 An electricutiIity in Florida is using such Btu meters onsolar hot water heaters installed under itsauspices. 4

1‘Phi Iadelphla E Iectric Company In an interview re-ported In the General E Iectrlc study Conceptual Design and Systems Analysis of Photovolta;c Power Sys-tems.

1‘For example, the E Iectron Advancement Corpora-tion sells meters measuring 1400 to 1500 F water atflow rates of 30 to 100 gpm at a price of $400 to $500(price list Feb 9, 1977)

“So/ar Energy Digest, January 1977, p 9.

The metering problems of onsite electricgenerating systems depend on the nature ofthe customer’s relationship with the localutility, If the utility is wilIing to purchaseenergy at the same price as it sells energy toonsite users (or if it owns the generatingequipment itself), it may not be necessary tochange metering systems — because conven-tional meters can subtract from the netenergy account when energy is being soldand add to the account when energy is pur-chased. This practice is currently permittedin several New England States on an ex-perimental basis. ” In cases where energy issold at a different price from that pur-chased, dual meters will be required —withone ratchet to read sales to the customerand one ratchet to read purchases from thecustomer,

SAFETY

There could be risks associated with theinstallation of onsite electric-generating

“Ben Wolf, Gemini Company, private communlca-tlon, Apr 27, 1977.

Ch. V The Re/atlonship Between Onsite Generation and Convention/ Public Utilities ● 145

equipment which would not automaticallydisconnect from utility lines when repairsare being made or when a utility Iine breaksin a storm or accident. As one utility put it,this “poses no problem for trained utilitycrewmen who treat al I Iines as hot unlesslocally grounded The major problem canarise through exposure of laymen and chil-dren to potentially hot lines before repaircrews arrive on the scene “ 16 This problemcan be eliminated if the onsite generatingequipment is automatically taken off theline whenever the power fails. This is astandard feature in at least one design ofsmalI power-conditioning equipment nowon the market. 17 The utilities Interviewed inthe General Electric study indicated thatany type of onsite equipment which did notincorporate such a feature could be fittedwith a device al lowing the utlIity to severthe connection with the utility distributionnetwork through telemetry. Such units wereestimated to cost about $100.18 (The sameunits could also be used for interruptibleservice and for load management ) It wouldbe necessary for the utility to approve theonsite equipment connected to its grid inorder to ensure that adequate safety fea-tures of this type were installed

QUALITYBACK

There shouId be

OF POWER FEDTO UTILITIES

no difficulty in construct-ing equipment capable of providing powerto utilities which meets utility standards ofvoltage regulation and frequency control.19

One manufacturer of small inverter systemshas sold 65 units which are integrated into

“Paclflc Gas and t Iectrlc Company Interview pub-lished In General E Iectrlc’s photovoltalc study citedpreviously

‘ ‘Alan W Wilkerson (President of Gem[nl C o m -pany), S y n c h r o n o u s /n~ers/on Techrr/ques for Utiliza-tion of Waste Energy, 1976

‘‘Conceptual Des/gn and S ysterns Ana/ys/s of Photo-\o/ta/c S ysterns, General E Iectrlc Corporat ion, Sche-

nectady, N Y , Mar 19, 1977, pp 5-9“General Electrlc Company, op clt , pp 5-4

utiIity systems, and quality of power has notbeen an issue 20

LOCAL DISTRIBUTION CAPACITY

One potential technical difficulty, iden-tified in the General Electric study, is thepossibility that onsite units which feed elec-tricity back into the power distribution grldwould exceed the capacity of the Iines andtransformers serving the area It is unlikelythat onsite units would produce excesspower for sale at a rate high enough to ex-ceed the peak purchases of the onsite cus-tomer. There could be some problems inolder communities, where distribution sys-tems were installed without considering thepossibility that a substantial number of res-idences might be equipped with al l-electricsystems. A typical distribution system, how-ever, should be able to accommodate theoutput of residential photovoltaic systemswithout major changes in transformers orIines.21

ECONOMIC DISPATCH

Electric utilities now control the schedul-ing of their generators via computer. Thisoptimizes the efficiency of their entire sys-tem on a minute-to-minute basis throughoutthe day As long as a relatively small numberof a utiIity’s customers are using solar equip-m e n t w h i c h c a n g e n e r a t e e l e c t r i c i t y , n o

l a r g e - s c a l e s h i f t i s n e c e s s a r y i n c u r r e n teconomic dispatch practices. 22 The net loadto the utility would fIuctuate throughout theday, but current equipment and manage-ment schemes are adequate to handle therelatively large fluctuations that already oc-cur with daily cycles, local weather varia-tions, and industrial energy consumption.The utilities interviewed by the G e n e r a lE lec t r ic Company ind icated that spec ia ldispatch strategies would not be r e q u i r e d ,even if onsite generating devices were in-stalled by 10 to 30 percent of their cus-tomers.

‘OGemlnl Company, op clt“General Electrlc conceptual design study, pp 5-3“General E Iectrlc conceptual design study


The Dow Chemical Company’s examina-tion of industrial cogeneration was “unableto identify any problems or potential prob-lems of transient stability attributable todispersed industrial generation. The effectof dispersed generation close to load cen-ters is to improve system integration andstability problems.”23 However, if dispatch-ing or system stability became a problem,the difficulty could be resolved by using theinterruptible service devices discussed earli-er. When onsite users were producing too

much energy for a utility’s needs, the onsitedevices could simply be disconnected fromthe load. (The cost of this could be includedin the $100 per unit cited for interruptibleservices meters. ) It is also possible thateconomic dispatch of electric utiIities couldimprove if a number of onsite generatingfacilities were equipped with sufficient on-site storage of a fossil backup system. Dis-patch is not a problem for systems relyingon chemical fuels for backup.

CONTROL

TIME-OF-DAY PRICING

Some of the advantages of connecting en-ergy generating and consuming devices witha common energy transport system cannotbe realized if control over the equipment isnot exercised by a central authority capableof optimizing the performance of the inte-grated system. This control can be exerciseddirectly by a utility if it owns all of the stor-age and generating equipment in a system,but it can be exercised indirectly by such ap-proaches as time-of-day pricing. For exam-ple, with time-of-day pricing, the costs ofnonoptimum performance of equipment notowned by the utiIity wouId be communi-cated to the owners of th is equipmentthrough higher prices for energy consumedfor backup power and lower rates for anyenergy sold to the utiIity. The electric ratesnow in effect in most parts of the UnitedStates, however, do not have the effect ofenforcing optimum al locations between on-site and centralized generating equipment.I n fact, many of the current rates tend to dis-courage onsite generation in spite of poten-tial cost savings. (This problem is treated inchapter VI Legal and Regulatory Issues. )

Figure V-1 illustrates the dramatic changein electricity consumption in Great Britainwhich occurred after a time-of-day electric

‘ ‘The Dow Chernlcal (lompdny, et al , Energy /n-dustr/a/ Center Study. NSF Grant #OE P 7+2042, June

1975 , p 70

rate was imposed. Before the rate was in-troduced, very little electricity was con-sumed during the night and an enormous

Figure V-1 .— Improvement in the Pattern ofElectricity Consumption in England’s SouthWestern Electricity Board Resulting From a

Shift to Time-of-Day Electricity Rates

in-

--- 1972/73 CONSTANT DAILY LOADLINE

O 3 6 9 12 15 18 21 24

HOURS

Winter-average Weekday Load Curve

SOURCEAsh bury, J G and A. Kouvalls, E/ectrlc Storage Heaflng The Ex-perience in Eng/and and Wa/es and Jn fhe Federa/ Repub/lc ofGemrdr?y Arqonne Nallonal L a b o r a t o r y , E n e r g y a n d E n\ fro n men ta( Systpms DIVIS ron A N LJES 50 1976 P 14

Ch. V The Relationship Between Onsite Generation and Conventional Public Utilities . 147

crease occurred in the morning when elec-tr ic heaters and other equipment wereturned on. After the variable rate was intro-duced, many consumers purchased onsitethermal storage devices which could becharged during the night and used to heatbuildings during the day. The result was amuch more uniform pattern of energy con-sumption.

STORAGE STRATEGY

Onsite solar electric devices integrated in-to electric utility grids provide a good exam-ple of some of the d i f f icu l t ies which canarise from local control over storage equip-ment, A solar electric system which is notconnected to a utility grid would charge itsbatteries during the day and discharge itsbatteries during the night. This is preciselythe wrong st rategy of operat ion f rom theperspective of the electric utility since stor-age devices owned by the utility would becharged during the night, when demands arelow, and discharged during the day when de-mands are greatest.

There will be some overlap between thetwo operating strategies since both types ofstorage wouId be discharging near sunsetand during cloudy days, but it is clear thatthe storage equipment would be used tobest effect if it were controlled by the util-ity. The advantage of using the solar electricdevices to meet utility electric demands di-rectly during the day (instead of sending it tobe stored) is arnplified by the fact that stor-age devices are typically only about 75-per-cent efficient This logic would apply even ifa very large fraction of the utility’s energywere derived from solar sources, although inthis case the strategy of operating individualstorage systems would closely parallel theoperation of utiIity storage. (A quantitativeevaIuation of this issue appears i n the finaIsection of this chapter, )

LOAD MANAGEMENT

The performance of isolated and inter-connected energy systems can be improved

if control is exercised over devices whichconsume energy as well as over energy stor-age and generating equipment. Clearly, en-ergy consumers will want to exercise asmuch discretion as possible over the amountof energy they use and when they use theenergy, but they also may be willing tochange their consuming habits to some ex-tent if they are required to pay large pre-miums for energy consumed during periodswhen energy is relatively expensive to pro-duce. Consumers may be willing to post-pone or defer the use of appliances such asdishwashers, disposals, clothes washers anddryers, and other equipment when electrici-ty is expensive.

The utility can exercise control throughthe use of “interruptible service” equipmentwhen onsite equipment includes onsite stor-age, Such devices wouId permit the utiIity toturn off water heaters and other applianceswith storage capabilities during periods ofpeak demand. Equipment of this type hasbeen installed for relatively large-scale test-ing by the Detroit Edison Co. If this equip-ment could ensure that onsite generatingequipment (whether thermal or electric) pur-chases backup power only during off peakperiods, the cost of backup energy to the on-site customer might be reduced — perhaps tothe point where energy could be bought andsold at the same rate. An experiment was re-cently conducted in Vermont in which theseappliances were automatically turned offwhen utility rates were high. The customerswere able to switch them back on again, butin most cases were wilIing to wait untiI ratesfe l l .24 Wel l - insu la ted wa te r hea te rs andfreezers are able to operate effectively evenif their supplies of electricity are automat-ically shut off during the day when electric-ity prices are high. Several cities in Germany

have utilities which are able to exerciseelaborate control over energy-consumingequipment The central load-management

“J G Ashbury and A Kouvalls, E/ectr/c S t o r a g eHeating The Experience in Eng/and and \l’a/es and Inthe Federal ~epub/Jc of Cermany, A r g o n n e Ndtlona iLaboratory, Energy and Environmental Systems DIv I-$Ion, ANL F S-50, 1976, p 20


computer can control both the generatingequipment and e lec t r ic i ty -consuming d e -vices. The computer sends a signal down theelectric wires which automatically shuts offindustrial equipment, refrigerators, waterheaters, and other equipment where energyuse can be deferred during peak periods. 25

As was the case with the control overstorage, however, the strategy of deferringdemand for energy will depend strongly onhow the solar devices are connected withother energy equipment. If the solar deviceoperates in isolation, an attempt should bemade to sh i f t a l l demands for energy toperiods when the sun is shining, If an elec-tric utility is used for backup power, how-ever, it will usually be preferable to shift thedemands which would require backup pow-er to the late evening.

OFFPEAK ELECTRICITY

It is sometimes argued that electric util-ities will be able to sell “off peak” electricityat a rate which covers only the cost of oper-ating a large “baseload” plant and the rel-atively inexpensive fuels which can be usedin these plants. Such rates are possible, butthey must be considered promotional since,

in effect, they subsidize the price of elec-tricity during the night by charging daytimecustomers for all other utility costs. Thesecosts are capital charges on generatingplants and transmission and distribution sys-tems, the costs of maintaining the transmis-sion and distribution lines, and all otheroverhead costs — including the added costof maintaining dual meters for daytime andnighttime rates.

It is also important to recognize that thereis not an unlimited supply of “off peak”power available in a given utility. There aremany possible uses for the power availableat night, in addition to storing heat forbuildings. The power can be used to chargebatteries for electric vehicles and in other in-dustrial procedures which can be deferredto use night rates. The utilities may find thatthey require the “off peak” nighttime energythemselves to charge their own storage de-vices, if utiIity storage must be used as a re-placement for the oil- and gas-fired genera-tors now used to meet utility peaks. (TheNational Energy Plan places major emphasison eliminating utility use of oil and gas. )And utilities must also make some use ofoff peak periods to maintain their equip-ment.

OWNERSHIP

The complex rates required to encouragedesign of onsite equipment best suited forthe energy network of which it is a part,would, of course, be obviated if utilitiesowned the onsite systems outright. Whilethere clearly are disadvantages associatedwith expanding the monopoly position ofutilities, there are a number of reasons forbelieving that utility ownership of onsitesolar equipment may be attractive in manycircumstances:

● UtiIities are uniquely able to optimizethe mix of generat ing and storageequipment in their service area and todevelop control strategies for minimiz-

“ J [; Ashbury, op [-It , p 20

ing overall utility costs. (This could, ofcourse, also be done by a companyowning only transmission and distribu-tion equipment, )

Utilities compare the cost of energyderived from new solar equipment tothe cost of generating energy from newelectric-generating equipment or themarginal cost of gas from new sources,while all other solar owners must com-pare solar costs to the lower imbeddedcosts of energy which determine com-mercial rates.

Utilities are probably better able toraise large amounts of capital for long-

Ch. V The Relationship Between Onsite Generation and Convention/ Public Uti/ities ● 149

term energy investments than any othertype of organization. As the statistics intable V-4 indicate, electric uti l i t ies re-qu i re severa l t imes more cap i ta l perdollar of sales than typical industrialfirms (although the capital intensity hasdec l ined rap id ly in recent years be-cause of the rapid increase in fuelcosts) At the end of 1974, electric util-ities owned approximately $150 bill ionin plants and equipment— nearly 20percent of all business plant equipmentin the United States.26

Table V-4. —The Capital Intensity of Major U.S.Industries

[Dollars of plant to secure $1.00 of revenue]

$Investor-owned electric companies:(1965) . ., . . ., . ., ... ., .. .4.51(1970) ., ., ., . . ., ., ., ... ., .. .,4.39(1975) ., ., ., ., ., ., ... . . . . .3.20(1977) (est. ) . , ., . . . 2.96

Bell Telephone System (1971 ) . ., . . . .. ..2.9510 ma jo r r a i l r oads (1971 ) . , . . , . . . . . . . . . , 2 . 48Gas transmission companies (1971 ) . . . . . .2.43Gas distribution companies (1971 ) . . . . . . .. .1.6210 major integrated oil companies (1971) ., .. ..,1.25500 diversified industrial companies (1971) .. ...0.8750 major retailers (1971) . . . . . . . . . .. ..0.43

SOURCES Naf~ona/ Gas Survey U S Federal Power CommjsslonEbasco Services Incorporated (New York, N Y ), 1977Bus{ness and Economic Charts p 28

An ability to raise capital for long-term investments is particularly criticalfor solar energy devices since solar en-ergy is very capital-intensive and typi-calIy requires a number of years to re-turn the initial investment. Utilities,therefore, may be uniquely able to pro-vide initial capital for solar devices insituations where individuals (particular-ly individuals in lower income groups)and organizations may find the capitalrequirements prohibitive. Reverses inthe stock market, uncertainties about

“Federal Power Comm[ss[on, National Power Sur-vey, Flnanc/ai Outlook for the Electric Power Industry.Report and Recommendations of the Technical Ad-visory Comm/ttee on Finance, December 1974, P 50

the future of the energy industry, anddifficulties in obtaining rate changesfrom utility commissions have, how-ever, made it progressively more dif-ficult for utilities to raise capital in re-cent years,

Utility capital tends to be less expen-sive than capital required by more spec-ulative industries. A typical utility canraise 50 percent of its capital fromdebt–while most manufacturers relyon debt for only 10 to 15 percent of newinvestment capital, the remainder ofthe capital being purchased at higherrates from investors, 27 Utility capitalcosts may, however, be higher thanthose experienced by homeowners andcan be higher than the cost of capitalavailable for financing typical residen-tial and commercial buildings. A homeowner earning a tax-free return of 10percent on capital invested in solarenergy equipment may be well satis-fied. This advantage is moot, of course,if the individual is unable to raise anycapital for the project at al 1.

Utilities are already in the business ofselling energy and have the required in-frastructure for billing, marketing, andrepairing equipment, Some potentialowners of onsite devices have beenwary of investing in equiprnent whichmight lead them to unfamiIiar mainte-nance problems or the hiring of special-ized personnel,

Utility ownership or marketing of solarequipment and a willingness to standbehind the equipment once installedcould increase consumer confidence inthe equipment.

There is some ambiguity, however, aboutwhether utility ownership of smalI solar en-ergy equipment would be permitted by Fed-eral antitrust statutes. The legal issues of

“Pau l j Garfield and W F LoveJoy, Publ ic Ut i l i tyEconomics, Prentice Hall, Inc , Englewood Cllffs, N J ,1964, p 25


ownership are discussed in greater detail inchapter V 1, Legal and Regulatory Issues.

UTILITY ATTITUDES

There is no industrywide position eitheron the issue of onsite generation or on thequestion of utility ownership of such facil-ities, Industry attitudes vary company bycompany, and the diversity of attitudes isdue, at least in part, to the fact that manyutility companies simply have not taken aclose look at the issue.

Natural gas utilities have expressed thegreatest recent interest in onsite solar facil-ities since supplies of gas are diminishingand the companies are looking for new ener-gy sources to replace natural gas in thefuture. Southern California Gas Co., for ex-ample, has tested solar-assisted gas heatingfor apartment buildings as one means ofconserving supplies and extending the Iife ofthe company’s resources .28 Interest is notconfined to gas companies. A recent surveyfound nearly 100 electric utilities whichwere initiating projects in solar heating andc o o l i n g . 29 30 The EIectric Power Research In-stitute has an extensive program in solarenergy equipment.31 At least one electr ic

utility has entered into an agreement with alocal installer of solar hot water heaters, anda gas utility has proposed changes in regula-tions that would permit it to operate as acombined gas and solar utility.32 33

‘aAlan Hlrshberg, Pub//c Po/icy for So/ar H e a t i n gand Cooling, October 1976, p 37

“Solar Energy Intelligence Report, Feb. 14, 1977, p31

‘“E Iectrlc Power Research Institute, Survey of E/ec-tric Utility So/ar Prolects, Palo Alto, Callf , ER 321 -SR,1977

J‘ E Iectrlc Power Research I nstltute, E/ectric PowerResearch Institute: So/ar Energy Program, fall of 1976,E PRI RP 549, 1976

“E S Dav is , Comrnercia/izing Solar E n e r g y : TheCase for Gas Uti//ty Ownership, mlmeo, Jet PropulsionLaboratory report, California Institute of Technology,June 1976

“General Electrlc Corporation, conceptual designstudy.

On the other hand, most of the utilitycompanies surveyed for a General EIectricstudy referred to earlier said they were re-luctant to enter the business of selling ther-mal energy systems. Some indicated that itwou ld requ i re too much d ivers i f ica t ion;others c i ted problems wi th meter ing andother technical diff icuIties.

A study conducted for the Federal EnergyA d m i n i s t r a t i o n f o u n d m i x e d o p i n i o n samong utilities on expanding capacity byusing conventional onsite generating equip-ment, primarily cogeneration systems in fac-tor ies and other generators of processheat. Examples include:

●

●

●

●

A west-south-central utility which sellsboth steam and electricity.

A Pacific coast utility, which has in-stalled turbines at a paper mill, returnslow-temperature steam to the mill andpays $0,01 per kWh for the electricitygenerated.

A Vermont utility actively searching forcogeneration opportunities.

A Texas utility which stated flatly thatthey were “not in the business of sellingsteam, ” and which turned down severalopportunities.

Although utility attitudes may be chang-ing, there have been scattered complaintsthat utilities have tried to thwart privatecompanies intending to instalI onsite gener-ating equipment. The Dow Chemical Co. re-ports that onsite industrial cogeneration“has been consistently discouraged by long-standing policies on the part of most pri-vately owned electric utilities that have dis-couraged in every way possible the genera-tion of electricity by any other type of or-ganization. Relevant here are rate schedules

34 Dow Chemical Co , et al , Energy /ndustria/ CenterStudy, June 1975, p 23

jSThermo E Iectron Corp., A Study of /rip/ant ~lectr;cPower Generat ion in the Chernica/, Petro/eum Refin-ing, and Paper and Pti/p /ndustries, June 1976

Ch. V The Relationship Between Onslte Generation and Conventional Public Utilities . 151

that favor large indust r ia l users (whether if no power is used) that make it uneconoml-justif ied or not by “cost of service”), and cal to use the utility as a standby sourceheavy demand charges (charges levied even backing up industrial power generation “‘b

THE COST OF PROVIDING BACKUPPOWER FROM AN ELECTRIC UTILITY

There is no easy way to compute the costof providing backup power to an onsite sys-tem from a conventional electric utilitysince electric utility costs are very sensitiveto the cost of equipment and fuels in thearea being served and to the times whenelectric backup power is demanded. A sim-ple technique for computing these costs ispresented here to illustrate some of the ma-jor trends, to exhibit several different waysof looking at the major trends, and to exhibitseveral different ways of looking at the issueof rates. It must be emphasized from the on-set that the method cannot be taken to be aprecise calculation of the costs of electricutiIities actualIy operating in the regionscovered. The results are so sensitive to localconditions that each utiIity must make itsown analysis of the costs. The followinganalysis shows that utiIity costs are extremeIy sensitive to four variables:

● The regional cost of equipment, theavailable financing, and the local costof fuels;

● Local climatic conditions [for example,solar backup costs are lower if peakheating and cooling periods are cor-related with periods of clear skies);

● The type of solar equipment installed(collector area, storage capacity, etc.);and

Ž The number of buiIdings in a utiIityservice area equipped with solar energydevices (a relatively small number ofsolar facilities will contribute to loaddiversity, but a large number will re-verse this resuIt).

The details of the technique used to com-pute utility costs, and detailed assumptionsmade about the costs of equipment andfuels experienced by each utlIity, are de-scribed in detail in appendix A of thischapter, but the basic method IS straight-forward:

1.

2

3

A “baseline” utility was constructed bycombining the electric demands of anumber of buildings and industrialprocesses using a mixture of energy-consuming equipment which was typ-ical for the city under examination. Thecost of providing electricity for thecombined utility load was then com-puted, assuming that the utility was op-timized to meet the demands. (Both thecost of energy attributable only to thegenerating units, the so-called “busbarcosts, ” and the cost of energy deliveredto customers were computed).

The cost of meeting a new set of elec-tric demands resulting from adding aspecified number of solar and nonsolarbuildings to the utility was computed,assuming that the utility was optimizedto meet the new demand pattern.

The effective cost of providing backuppower for different types of buildingscould then be computed by examining

the incremental costs and the incre-mental ut i l i ty costs which resultedwhen the new buiIdings were added,

Since it was assumed that all of the equip-ment in the utility was new, and the costscomputed are all essentialIy marginal costs,the actual utility costs in the region wouldbe lower because some fraction of the ener-

“ DOW Chemical Co , et al , Energy /ndustr/a/ CenterStudy, June 1975, p 23

152 Ž Solar Technology to Today’s Energy Needs

gy would be generated from less-expensive,older plants. There are a number of other ar-tificial ities in the technique:

●

●

●

The choice of an “optimum” set of util-ity equipment does not use a detailedanalysis of overall system reliabilityand maintenance schedules— it insteadsimply assumes that the utility will pur-chase 20 percent more in each gener-ating capacity than the peak required in ,that category to meet the load;

Utilities will seldom be able to deployequipment which is optimally suited toload patterns because of regulatory de-lays, an inability to precisely predict de-mands, and the need to use older equip-ment; and

I t was assumed that none of the utilitiesevaluated owned storage equipment. Inthe future, utility storage devices mayplay a major role in replacing the oil-and gas-burning devices now used tomeet real demands. The impact of stor-age on the cost of backup power forsolar systems will need to be carefullyexamined. One would expect that low-cost storage would minimize the nega-tive impacts of solar equipment.

This technique has been used to computethe cost of providing electric power to anumber of different single family houses,and the results are summarized in table V-5.This table compares the cost per kWh ofproviding electricity to the building de-scribed to the cost per kWh of providingelectricity to a similar building using anelectric heat pump. AlI comparisons are be-tween delivered costs. Greater detail on theutility costs and the capacity of equipmentinstalled is presented in volume I I.) For ex-ample, the table indicates that the electrici-ty required by a single family house usinggas for heating, water heating, and air-conditioning in Albuquerque, N. Mex., coststhe utility 2 percent more than the electric-ity used to provide power for a conventionalsingle family house in the same city using aheat pump and electric hot water.

An examination of table V-5 reveals sev-eral

Ž

●

●

●

features:

Electricity for conventional houses us-ing gas for everything other than fansand miscellaneous electric loads coststhe ut i l i ty approximately the sameamount as electricity for a heat-pumphouse.

Electricity costs are lower for housesusing electric resistance heat (presum-ably because they use more electricityduring periods of mild winter weatherwhen utility demands are relativelylow) and are higher for houses using gasheat and electric air-conditioning. (Allof the utilities examined have peaks inthe summer, and these peaks are in-creased by the added air-conditioning.But the added equipment is underuti-lized since the new houses do not usemuch electricity during the winter. )

The houses using solar energy for heat-ing and hot water and a heat-pumpbackup cost the utility more per kWhthan the conventional houses usingbaseboard heat , but less than thehouses using gas heat.

The photovoltaic houses cost somewhat more than the houses equippedonly with solar heating and hot water,but in no case is the utility cost sig-nificantly larger than the cost of pro-viding backup power for a heat-pumphouse– in several instances, the utilitycosts are lower in the solar cases. Therelat ively favorable appearance ofthese solar systems results in part fromthe fact that some of the utility air-con-ditioning peaks can be reduced by thesolar devices. (About 50 percent of thetotal building energy requirement isprovided by the photovoltaic devices,and about 30 percent of the total ener-gy requirement of the house is providedby the heating and hot water systems.)

Utility costs per kWh of backup powerdelivered are increased slightly if salesto the utility are permitted.

Ch. V The Relationship Between Onsite Generation and Convention/ Public Utilities ● 1 5 3

Table V-5.—The Fractional Difference Between the Utility Costs [¢/kWh] Required to Provide BackupPower to the Systems Shown and the Costs to Provide Power to a Residence Equipped With an

Electric Heat Pump [see note for explanation]

1.

2

3.

4.

5.

Albuquerque Boston Fort Worth Omaha

Single family house with gas heat,hot water, and alr-conditioning” . . 0.02 -0.09 -0.15 0.03

Single family house with gas heatand hot water, and central elec-tric air-conditioning ● . . . . . . . . . . . 0.26 0.28 0.15 0.32

Single family house with base-board heat, electric hot water, andwindow air-conditioning” . . . . . . – 0.14 -0.14 -0.15 -0.10

Single family house with solar heatand hot water backed up with aheat pump and electric hot water* 0.01 -0.13 0.06 -0.07

Single family house with extra in-sulation, electric hot water, andheat pump with: ● ●

a. Pho tovo l t a i c sys tem w i t h nobattery and no sale to the uti-lity . . . . . . . . . . . . . . . . . . . . . . . . -0.06 -0.27 ‘0.02 -0.07

b. Photovoltaic system with nobattery and sales to utility per-mitted, . . . . . . . . . . . . . . . . . . . . 0.07 -0.23 0.03 -0.01

C. Photovoltaic system with bat-tery and no sales to utility. . . . . -0.30 -0.27 0.01 -0.05

“ Compared with single family house with electric hot water and heat Pump.“’ Compared with single family house with extra insulation, electrlc not water, and heat pump,

NOTE: let Cr = Incremental utlllty costs resulting from adding 1,000 reference houses with heat pumps.

let Kr = the Incremental number of kWh generated when 1,000 reference houses with heat pumps are added to the utilitylet Ct and Kt be the equivalent quantities resulting from adding 1,000 houses with a different kind of energy equi Prnent

Then the fractional change JJlustrated above Is given as follows: F = Ct/Kt - Cr/Kr

Cr/Kr

Table V-6 indicates the Ievelized monthlycosts which would be experienced by con-sumers if they were charged electric ratesreflecting the marginal cost of providingenergy to several types of energy systems.Levelized costs assuming a moderate rate ofincrease in electric prices are shown f o rcomparison. In these cases, it is assumedthat an electric utility will purchase elec-tricity from the onsite generating facility for50 percent of the price at which it sells elec-tricity. (In the cases when the electric price

28-842 () - 11

was assumed to be the marginal cost of pro-viding backup — giving credit to the value ofelectricity sold to the utility— it was as-sumed that electricity is bought and sold atthe same price. )

Except for Boston, the two techniques forcomputing Ievelized energy costs producesimilar est imates of Ievel ized energycharges. (It is Iikely that the techniques usedto compute the marginal costs of electricityare overly optimistic about the costs of new


Table V-6.—Levelized Monthly Costs for a Well-Insulated Single Family House Showing the Effectof Marginal Costing for Backup Power

I. ALBUQUERQUE

—no solar. . . . . . . . . . . . . . . . . . . .—59 m2 silicon photovoltaics,

no batteries, no sales to thethe utility . . . . . . . . . . . . . . . . . .

—59 m2 silicon photovoltaics,no batteries, sales to utilitypermitted . . . . . . . . . . . . . . . . . .


Il. BOSTON

—no solar. . . . . . . . . . . . . . . , . . .—59 m2 silicon photovoltaics,

no batteries, no sales to theUtility . . . . . . . . . . . . . . .


—59 m2 silicon photovoltaics,no batteries, sales to utilitypermitted ., . . . . . . . . . . . . . . . .

Ill. FORT WORTH

—no solar. . . . . . . . . . . . . . . . . . . .

—59 m2 silicon photovoltaics,no batteries, no sales to theutility . . . . . . . . . . . . . . . . . . . . .



IV. OMAHA

—no solar. . . . . . . . . . . . , . . . . . . .—59 m2 silicon photovoltaics,

no batteries, no sales to theutility . . . . . . . . . . . . . . . ., . . . .



Electricity rates assumed to Electricity rates reflect marginalincrease by BNL forecast utility rates

(see appendix for methodology)

No credits 20% ITC on No credits 20% ITC ongiven solar equipment given solar equipment

183

213(255)

197(240)

216(267)

300

319(362)

305(348)

324(376)

190

228(272)

218(263)

240(293)

211

241 (284)

231 (275)

253(305)

183

202(246)

187(231 )

204(257)

300

308(353)

294(339)

311 (365)

190

216(262)

207(253)

226(281 )

211

230(275)

220(265)

239(294)

204

21 4(257)

190(232)

194(245)

215

215(259)

200(244)

227(279)

191

227(272)

212(257)

242(295)

219

238(282)

223(267)

250(303)

204

204(248)

179(223)

181 (234)

215

204(249)

189(234)

21 3(267)

191

216(262)

201 (247)

228(283)

219

228(273

21 2(258

237(291 )

NOTES: (1) All houses use heat pumps for space conditioning and electric resistance hot water heaters(2) Parenthesis ( ~ indicates utility ownership.(3) ITC = Investment Tax Credit

Ch. V The Relationship Between Onsite Generation and Convention/ Public Uti/ities Ž 155

equipment in that region. ) More important-ly, however, the ranking of the Ievelizedcosts of the four systems appears not to de-pend on whether marginal costs or averagecosts are used to make estimates.

The same methods can be used to esti-mate the price a utility should be able topay for electricity produced by an onsitedevice which exceeds onsite demands. This

can be done simply by computing the dif-ference in utility costs which results when agroup of solar buildings is permitted to sellenergy and dividing this cost by the totalamount of kWh sold. The results are pre-sented in table V-7 as a ratio between thevalue of the electricity available for pur-chase and the average cost of electricitygenerated by the utility. Since all costs inthe utility used in this analysis are effective-ly “marginal costs,” the onsite devices are

Table V-7.—Ratio of Price Utilities Can Pay for Solar Energy Generated Onsite to the PriceCharged by Utilities for Electricity

Albuquerque Boston Fort Worth Omaha

Purchase Purchase Purchase Purchase Purchase Purchase Purchase Purchasebase reference base reference base reference base reference

1. Single family

house with noonsite bat-teries . . . . . .

2. Single familyhouse withonsite bat-teries . . . . . . .

3. Single familyhouse withextra insula-tion and noonsite bat-teries . . . . . . .

4. High riseapartment withno onsite bat-teries . . . . . . .

5. High riseapartment withonsite bat-teries . . . . . . .

0.67 0.65 0.62 0.55 0.31 0.28 0.57 0.59

0.40 0.39 1.09 0.98 0.29 0.26 0.64 0.66

0.64 0.62 0.58 0.49 0.66 0.64 0.50 0.50

0.64 0.66 0.51 0.48 0.29 0.29 0.42 0.44

0.41 0.43 0.38 0.36 0.29 0.28 0.36 0.38

Let x n (utlllty cost supplylng buildlng not selling electricity minus utility costs for building selling excess electricity to the utility)

Let y = (kWh generated by util ity In supplying buildlng not selling electricity minus util ity costs in supplying building selling excesselectrlclty)

Let z n (added ut i l i ty cost incur red in supply ing bu i ld ing wi th no so lar equ ipment) d iv idedthe bulldlng)

Let w = (total utlllty costs) dlvlded by (total kWh produced by the utility)—no additional buildings

P u r c h a s e = Xly P u r c h a s e = Xly

Base w Reference x

All utlltty costs are dellvered costs

by (added kWh required to supply

156 . Solar Technology to Today’s Energy Needs

not given credit for the difference betweenthe imbedded average utility costs used todetermine selling prices and the fact thatnew solar systems will displace relatively ex-pensive “new” generating facilities.

Table V-8 shows information for high riseapartment buildings which is equivalent tothe data for single family houses shown intable V-7. It can be seen that the solar de-vices are less attractive to the utility in thesecases, in part because the reference casechosen for the high rise building uses elec-tric baseboard heating—which produces amore even load than the heat pumps usedfor the single family reference case.

As noted earlier in this chapter, the costsof providing electricity can be reduced if on-site storage is available at each building sitewhich permits the buiIding to purchase ener-gy during periods when the demand on theutility is relatively low. Solar equipment, ofcourse, is typically already equipped with athermal storage device, and these devicescan be converted to allow the system to pur-chase energy only during off peak periodswith a relatively simple change in their con-trol systems. The result of installing off peak

storage devices in a number of differenttypes of buildings is shown in tables V-9 andV-10. Chilled water can also be producedduring the night and stored in tanks toreduce cooling loads during the day. The useof “off peak cooling” has the additional ad-vantage of allowing the chilling equipmentto operate at night when it is more efficient.In computing the load pattern, it was as-sumed that the storage devices are chargedbetween midnight and 5 a.m. and that theamount stored is equal to the amount ofbackup energy for heating, hot water, andcooling which would have been required forthe previous day with no off peak storage.

An examination of tables V-9 and V-10shows that the savings to the utility can beconsiderable if off peak storage is used on-site. In the case of off peak storage of cool-ing, utility costs per kWh attributable to thehouse are reduced by nearly 50 percent. Inall cases, the reduction in costs is lower ifsolar equipment is installed, but in many in-stances the difference is not very large. Typ-ically, the utility cost per kWh to providebackup power for the solar houses with off-peak storage is about 10 percent greater

Table V-8.—The Fractional Difference Between the Utility Costs Required To Provide Backupto the Systems Shown and the Costs of Providing Power to a High Rise Apartment

Equipped With Central Electric Air-Conditioning, Electric Hot Water, and Baseboard Resistance Heating[see notes on previous table for explanation of how these fractional changes are computed]

Building equipment Albuquerque Boston Fort Worth Omaha

1.

2.

3.

Gas heat, gas hot water, and gasair-conditioning. . . . . . . . . . . . . . . . . . . . . . . . . . 0 -0.11 -0.12 -0.05

Gas heat, gas hot water, electric air-conditioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.02 0.32 0.21 0.26

Electric hot water, electric baseboardheat, central electric air-conditioning, anda photovoltaic system:

a. No batteries onsite, no sales to grid . . . . . 0.27 -0.01 0.21 0.03

b. No batteries onsite, sales to grid allowed . 0.31 -0.02 0.23 0.09

c. Batteries onsite, no sales to grid . . . . . . . . 0.28 -0.02 0.22 0.08d. Batteries used onsite, sales to grid

allowed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.31 -0.02 0.22 0.08

Ch. V The Relationship Between Onsite Generation and Conventional Public Utilities ● 157

Table V-9.—The Impact of Off peak Storage on Utility Costs[fractional increase or decrease in backup costs per kWh–see notes]

A l b u q u e r q u e Boston Fort Worth O m a h a

Nonsolar houses● Off peak storage for heat and hot water . . . . . . . -0.37 -0.38 -0.29 -0.34

● Off peak storage for heat, hot water, andcooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.47 -0.45 -0.48 -0.44

Houses with solar heating and hot water

. No off peak storage . . . . . . . . . . . . . . . . . . . 0.03 -0.11 0.12 -0.06

Ž Off peak storage for heating and hot water . . . -0.11 -0.24 0.003 -0.21

. Off peak storage of heating, hot water, andcooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -0.30 -0.35 -0.32 -0.36

Notes The t‘ reference house’ is a single family house using electric resistance heating and hot water and window air-conditioners

All solar houses generate only heating and hot water from solar energy.

Let Cr = added u tI I It y costs resu It I ng from the add It ton of 1,000 reference houses

K r = added kWh resu It I ng from the add It Ion of 1,000 reference houses

% = added u tI I Ity costs resu It Ing from the addlt Ion of 1,000 test houses (type noted In left column above)

K = added u tI I Ity costs resu It I ng from the add!t Ion of 1,000 test houses

The fractional change ratto shown above IS calculated as follows:

(Ct/Kt) - (Cr/Kr)F=

(Cr/Kr)

than the utility cost per kWh attributable toconventional houses with off peak storage.

It is apparent that the solar systems havemore difficulty competing with convention-al electric heating systems if the conven-tional devices use off peak storage and areable to buy electricity during off peak peri-ods at reduced rates. There are, however,still a number of cases in the examplesshown where solar devices are able to com-pete. Off peak storage equipment, while notas expensive as solar devices, can still becostly: they require the installation of heat-ing (or chilIing) equipment which is larger incapacity than conventional equipment bythe ratio of 24 hours to the number of hoursused to charge storage; with existing tech-nology, heat pumps cannot be used tocharge off peak heat storage and resistanceheating must be used; and the devices re-quire more sophisticated controls than or-

dinary heating systems. Another difficultywith storage of off peak electricity in theform of thermal energy to be used for spaceheating is that, in the climates examined inthis study, the use of off peak storage Ied toa significant increase in the electricity con-sumed by each building—which would re-duce the economic advantage to the build-ing owner of using off peak storage. It mustalso be recognized that the costs shown inthe tables implicity assume that an ideal“marginal rate” is charged in which the util-ity is able to charge each customer preciselythe real incremental cost incurred by theutility in supplying that customer. In thissense, the costs represent a “best case” foroff peak power. The rates also do not in-clude additional charges which might resultfrom additional metering and bilIing. The re-sults may indicate, however, that in thelong-run conventional electric utilities mayprove to be poor choices for providing back-up power to onsite solar instalIations.


Table V-10.—Levelized Monthly Costs for a Single Family House Showing the Effect of MarginalCosting and Buying of Off peak Power

Electricity rates assumed Electricity rates reflectto increase by BNL forecast marginal utility rates

(see appendix for methodology)

200% ITC on 20% ITC onNo solar & off peak No solar & off peak

credits equipment credits equipment

i. Albuquerque

—No solar or off peak buying:. Electric resistance heat. . . .● Heat pump heat . . . . . . . . . .

—Solar only:● Low cost coils . . , . . . . . . .Ž High cost . . . . . . . . .

—Offpeak heating, & hot wateronly . . . . . . . . . . . . . . . . . . . . . .

—Offpeak heating, cooling &hot water only. . . . . . . . . . . . . .

—Solar & off peak heating & hotwater● Low Cost coils . . . . . . . . . .● High cost coils . . . . . . . . . .

—Solar & off peak heating, cool-ing & hot water● Low Cost coils . . . . . . . . . . .● High cost CO iIS. . . . . . . . . . .

Il. Omaha

—No solar or off peak buying:● Electric resistance heat. . . .● Heat pump heat . . . . . . . . . .

—Solar only● Low Cost coils . . . . . . . . . .Ž High cost CO iIS . . . . . . . . . . .

—Offpeak heating & hot wateronly . . . . . . . . . . . . . . . . . . . . . .

—Offpeak heating, cooling &hot water only. . . . . . . . . . . . . .

—Solar & off peak heating & hotwater● Low COSt Coils . . . . . . . . . .. High cost colts. . . . . . . . . . .

—Solar & off peak heating cool-ing & hot water● Low Cost Coils . . . . . . . . . . .Ž High cost coILs. . . . . . . . .

239204

185(21 1 )205(240)

N / A

N 1A

N / AN / A

N / AN/A

278250

252(282)278(320)

N/A

N/A

N/AN/A

N/AN/A

239204

179(205)196(232)

N / A

N / A

N / AN / A

241229

183(209)203(238)

185(205)

198(232)

177(207)197(236)

N/A 194(242)N/A 214(272)

278250

244(276)267(31 1 )

N/A

N/A

N/AN/A

N/AN/A

278268

234(265)261 (303)

219(241 )

237(278)

221 (258)247(296)

244(303)270(341 )

241229

177(203194(230)

181 (201 )

189(225)

169(200)186(227)

181 (231 )198(258)

278268

227(259)250(294)

21 4(237)

227(270)

211 (250)234(284)

228(289)250(324)

NOTE: Solar systems are thermal only; Housea are SF-3

Ch. V The Relationship Between Onsite Generation and Convention/ Public Utilities Ž 159

Table V-1 O also shows that when contem-porary rate schedules are used, it costs lessto heat a single family house with a heatpump than with electric resistance base-board heat. The difference in costs narrowsconsiderably, however, if marginal costs arecharged. The resistance system actually isless expensive if off peak storage is used inconnection with the resistance heating. Thedisadvantages of heat-pump systems wouldbe magnified if a large number of customersin the regions examined used electric heat-

BACKUP POWER

ing and the utility peak occurred in thewinter (both utilities examined have summerpeaks). The heat-pump systems would bemore attractive, if their performance wereimproved or a technique could be devel-oped for storing off peak energy for use dur-ing periods when the heat-pump capacity isinadequate to meet the load; in current sys-tems, straight resistance heating is used insuch situations. Clearly, more analysis is re-quired in this area.

WITH ENERGYSOURCES OTHER THAN ELECTRICITY

It was noted earlier that the cost of pro-viding backup for solar equipment from rel-atively expensive electric-generating equip-ment may be so great that it would be pref-erable to provide backup by using seasonalstorage systems or relying on gas I n fact, ithas been suggested that electric utilities,with their high capital costs, are uniquelyunsuited to providing backup for solarequipment. 37 Analysis presented in volumeII shows that for large buildings (and forsingle famiIy structures which can pipe ther-mal energy to a central storage site), solarheating and hot water systems capable ofproviding 100 percent of local requirementsmay become economicalIy competitive with

1‘J oseph C Ashbury and Ronald 0 A!LJel Ier, “solar

E n e r g y and E Iectrlc Ut I I It le~ Shou Id They be In te r -tdceci?” \c/ence, 1954127, p 445 (1977]

electric heating and hot water in many partsof the country in the relatively near future.

Table V-11 indicates some comparisonsbetween gas and electric backup. In the gascases, it is assumed that the system is notconnected to an electric grid. Backup is pro-vided by a small, 32-percent efficient engineburning natural gas to power a heat pumpElectricity for lighting and other uses is pro-vided from a generator attached to the heat-pump engine The table indicates that thegas backup alternative may be attractiveeven if gas prices increase drasticalIy overthe next few decades This possibility maymake it interesting to consider the possi-bility of granting preferential allocation ofgas to facilities using gas to backup solar fa-cilities and permitting new gas hookups inregions where such hookups are now per-mitted if the gas is used as a solar backup


Table V-n .—Levelized Monthly Costs of Several Kinds of Energy Equipment in aSingle Family Detached Residence in Albuquerque, N. Mex.

(dollars per month)

Gas price in 2000 (¢/kWh) . . . . . . 0.005 0.011 0.015 0.03

Electric price in 2000(¢/kWh) . . . . . . . . . . . . . . . . . . . . . 3 4.5 10 —

Incentive ... , ., . . . . . . . . . . . . . . none

1.

2.

156

3.

4.

5.

6.

7.

Heat pump, electric hotwater . . . . . . . . . . . . . . . . . . . . . . .

Solar heat and hot water, electricheat pump backup—High. . . . . . . . . . . . . . . . . . . . . .—Low . . . . . . . . . . . . . . . . . . . . . .

Extra insulation, 59 m2 photo-voltaics at $500 kW, electric heatpump backup—No batteries . . . . . . . . . . . . . . .—20 kWh batteries at$70/kWh . . . . . . . . . . . . . . . . . . . .

Gas heat and hot water, centralelectric air-conditioning , . . . . . .

Solar heat and hot water backup,electric air-conditioning—High. . . . . . . . . . . . . . . . . . . . . .—Low . . . . . . . . . . . . . . . . . . . . . .

Extra i n s u l a t i o n , 5 0 m2

p h o t o v o l t a i c s a t $500/kW,gasfired heat pump/generatorbackup. . . . . . . . . . . . . . . . . . . . . .

Solar heat with central seasonalstorage, homes connected withhot water piping, electric air-conditioning in each house—High. . . . . . . . . . . . . . . . . . . . .—Low . . . . . . . . . . . . . . . . . . . . . .

187(223)158(1 83)

169(207)

190(228)

116

172(21 O)1 43(1 70)

157(1 96)

21 4(292)164(21 2)

none

203

213(250)184(210)

196(233)

215(253)

173

201 (239)172(1 99)

177(21 5)

233(31 1 )184(231 )

20%tax credit

395

309(354)284(315)

294(338)

303(349)

287

276(323)251 (284)

182(230)

290(383)249(306)

20%tax credit

—

——

—

—

500

422(469)379(430)

203(277)

——

Percent of totalenergy usagesupplied bysolar energy

o

2828

52

45

0

4141

50

6565

Notes: All costs In 1976 dollars

( ) = utlllty ownership

Documents

THE RELATIONSHIP BETWEEN ONSITE GENERATION AND