-291-
DYNAMICS OF POWER SUPPLY AND DEMAND
Krishan Kumar Garga N.K. Gupta
Electrical Engineering Department Engineering and Economics
Punjab Engineering College Research Inc. Suit 300,
Chandigarh - 160 012 1951 Kidwell drive, vienna
India VA 22180 U.S.A.
B. Thapar
Electrical Engineering Department
Punjab Engineering College
Chandigarh - 160 012
India
ABSTRACT
Power demand forecasting methodologies which are currently being used by electricity
authorities are end use method, trend method and Scheer's formula. These methodolo-
gies being static in nature, do not take into account the future power supply
position, while becoming an important instrument of economic change the growth of
power generation activity itself is totally dependent upon the overall economic
development thus forming an important feedback loop in the economic system.Present
paper discusses a power economy system dynamic model for estimation of future demand
and supply position of Power,
INTRODUCTION
Unlike early period of development of electric power generation activity, when
electric power was primarily used for lighting homes, electricity is now a major
input to most of the economic activity. While becoming an important instrument of
economic change the growth of power generation activity itself is totally dependent
upon the overall economic system. Any shortfall in the availability of electric power
has an inhibiting effect on the economic growth, whereas excess power generation is
a drain in the limited available resources which could be gainfully used in other
sectors of economy. It is, therefore, imperative that the dynamic interdependence
of all sectors should be considered while estimating the demand and supply position
of power for future.
Thus when palnning for an important sector like power which is both capital intensive
and is basically a long gestation activity, methods which give a long prospective
should be used. Currently available power planning methods, in contrast have a much
shorter time perspective. Forecasting methodologies which are currently being used
by electricity authorities in India are end use method and trend method(working
group 1979). Parikh did an excellent work in improving upon the work of working
group and prepared a ‘more comprehensive model (Parikh 1981, pp. 6-13). Pachauri's
model (Pachauri 1975) for a region in U.S.A is also a good attempt to improve upon
the conventional methodologies. Though these models go a step ahead then the
conventional methodologies, but the feed back loop of supply is not integrated in
the model. Rajadhayksha has highlighted the importance for an integrated approach.
"It is one thing to arrive at a set of forecasts but it is quite another to arrive
at some deeper analysis. As an example one could perhaps relate a number of economic
happenings with the occurance of sun spots which may provide us with a good way of
forecasting but such a method have little explanatory power. What we really need,
it seems is a vehicle that could enable us to find out in which direction the demand
-292-
for power would move as a result of various changes taking place in the economy of
the country. It would, therefore, be desirable to develop detailed and regorous
models of demand to give us not only a set of reliable forecasts but @ medium through
which the effect of various economic demographic changes on the demand for electric
energy can be assessed (Rajadhyaksha 1977, pp. 4).
System dynamics developed by Forrestor (Forrester 1961), present a viable alter-
native by making possible the comprehensive power economy model capable of simulat-
ing the total economy of a region to provide reliable demand and supply forecasts
(Garga et al. 1982, pp. 59-61). In addition to becoming a medium through which the
effect of various economic and demographic changes on the demand and supply of
electrical energy could be assessed. This new methodology is able to generate long-
term projections and reflect the sensitivity of power supply and demand to various
economic demographic variables and vice-versa. System dynamics has been used here to
formulate the power economy model to understand the dynamics of power supply and
demand for Punjab in India.
Situated as it is, in the northwest corner of India, the state of Punjab is far from
the coal-fields of East India. As a result of its proximity to lower Himalyas,
however state got a major irrigation-cum-hydroelectric scheme namely Bhakra dam
complex which was finally commissioned in the date 1950's. Considering the then
economic development of the state, power from this hydroelectric project was found
to be far in excess of demand. This succeeded in triggering increased economic acti-
vity resulting in the now famous green revolution with matching increase in Indus-
trial activity, which in turn resulted in tremendous increase in demand for electric
power with Consequent power shortages. With similar situation also prevailing in
other parts of India, shortage of electric power has now become a major constraint
on the economic development, despite the fact that a large portion of the Govt. funds
now diverted to capital expenditure needed for creating new power generating capacity.
As an example, an amount of Rs. 1890 million out of the annual plan budget of
Rs. 3850 million for the state of Punjab was spent in 1982-83.
OVERVIEW OF THE MODEL
Before discussing the power demand and supply projection mechanism in some detail,it
is useful to present a brief overview of the model. The model may be described as a
closed-loop-power-economy-model. Power model projects both demand and supply position
having feed back from economy model. Economy model having feed back from power supply
and demand model based on the power supply/demand ratio affects the output capital
ratio. Output capital ratio in turn affects the state domestic product. Power supply
model generates future power supply position from the state plan funds, share of
power plan and respective share of generation and transmission. Ratio of money in
generation and transmission will increase or decrease the power losses, which will
affect the power supply/demand ratio.
The economy sector generates consistent forecasts of variables essential to power
demand forecasting viz. Land requiring irrigation, diesel tubewells to be replaced,
rural and urban households, rural and urban houses, Industrial capital, state
domestic product, savings and capital formation etc. Each sector attempts to capture
the basic elements that shape the future dimensions of a region's power and economy
position.
In the power demand sector, the projections of various economic parameters are
coupled with the power consumption pattern of electricity usage to produce projec-
tions on yearly basis. The power demand sector is divided into basic consumer
catagories: Domestic, Agriculture, Industries, Commercial and Public Lighting. In
the following discussion, the basic loops of the power supply,demand and economy
~293-
GROUND WATER
CULTURAL COMMERCIAL,
Pare “Bera Sector
AGRICULTURE ———~a TUBE WELLS +
x
POWER SUPPLY
+ “POWER DEMANO OY
POWER DEMAND
= <p GOOMESTIC i = CAPITAL.
ima” / Rous
“eo
~£ "+ cee
senile my ZZ
y PRODUCT
aerate Neen CAPITAL
FORMATION
nin VAR J
wah
wie
MONEY TAXES
Sa POWER mat plan a
SHARE
FIG. 1 CAUSAL LOOP DIAGRAM
model are discussed as shown in the causal loop diagram Fig.1. The model has been
simulated on DEC 2050 digital computer using Mini Dynamo compiler for running the
model.
THE MODEL
The model consists of the following sectors: 1) Population, 2) Agriculture, 3)
Manufacturing, 4) Services, 5) Capital Sector, 6) Government revenue and 7) Power
Supply and Demand.
Population creates demand of power for domestic sector, for illumination, heating,
cooling and other household needs. In the population sector,population is divided
into 15 age groups and the birth rates in the infant age group and death rates in
all age groups are calculated dynamically from related parameters from all the
sectors. As power demand for rural and urban households differ substantially,
population is further divided into rural and urban population. As the power is to
be used in a residential premises, demand is based on the incoming houses both in
rural and urban sectors. Houses are constructed depending upon the capital formation
for construction in residential sector and the requirement of houses depending upon
the population and family size. New houses constructed each year having feedback
from capital sector shown in the cusal loop diagram 1 will determine the demand for
new load to be connected. Total connected load with increasing per capita income
and consumption for urban and rural domestic sectors computes the demand for
domestic sector.
Agriculture sector creates the demand of power for irrigation. Irrigation is one
of the most important inputs for the growth of agriculture. High yielding
varieties of crops and chemical fertilisers require assured irrigation for giving
results. With the increasing population pressure on land creates more demand
~294-
for food, creates increasing demand for irrigation. Agriculture sector computes
total yield of cereals, non cereals and other farm related products, based on
yield per hactare and land under cultivation and land double cropped having feed-
back from capital sector for providing capital for agriculture. Yield per hectare
is a function of yield per hactare normal, multiplying factor from irrigation and
multiplying factor from agriculture capital, thus completing its major feedback
loop from state domestic product.
Major irrigation is provided from tubewells both electric and diesel. As the
operating costs of electric tubewells is less than the diesel tubewells, demand
for electric connections to convert existing diesel run tubewells into electric
tubewells is yet to be fulfilled. This will also result in saving of diesel which
is in actual shortage.
Demand for power in agriculture sector is being computed by the number of electric
tubewell connections provided and power consumed per tubewell is a function of
agriculture capital per hactare of land sown. With increase in agriculture capital
per hactare,consumption of power increases. Electric tubewell connections is a level
which is increased by the incoming connections. New connections are provided based
on the incoming power installed capacity and power share of agriculture which is
function of percentage of land irrigated. Ground water potential is a limiting
factor for providing the total number of connections for tubewells.
Industrial sector have been partitioned into two main heads: power intensive indus-
try and other Industry. Capital from the capital formation sector as shown in the
causal loop diagram 1 is fed back to the two types of industries. Capital is divided
among the two industries depending upon the power supply/demand ratio. As power
shortage will persist, power intensive industry will be discouraged or it will come
with captive power plants. Capital in these sectors converts the capital into value
added in Industrial sector, depending upon the output capital ratio. Output capital
ratio has been assumed to be a function of power supply/demand ratio and output
capital ratio normal. Output capital ratio varies from .5 of the output capital
ratio normal OCRN, to one when power supply/demand varies from 0 to 1. Here the
demand in industrial sectors is a function of power required per unit of ‘capital and
expected variations in power intensiveness of capital,compute the demand of power.
Value added of industrial sector is fed to state domestic product.
Service sector creates demand for power in commercial sector for lighting, heating
and airconditioning. As industrial sector overtakes the agriculture sector in
economy, more and more people start migrating to the urban areas. With increasing
urban population, service sector activity starts increasing thus creating more
demand for power. Model computes the power demand for commercial sector based on
the urban houses and percentage increase in state domestic product.
With increasing economic activity and population migrating to urban areas,density
of road use starts increasing with increasing use of road transport, more and more
public lighting is required. But in state of power deficits, public lighting is the
first to bear the brunt of power cuts. Demand of power for public lighting is a
function of state domestic product and power shortage factor. Power shortage factor
is a function of power supply/demand.
Capital sector generates capital formation, delayed function of savings. Savings
are generated after deducting the consumption from state domestic product. This
capital is fed back to Agriculture sector, Industry sector and service sector. Part
of the capital is invested in residential construction for housing. Government
sector generates revenue from taxes. Taxes determine the plan funds. Share of
-295~
money from plan funds for power sector is determined by the normal power share and
power supply/demand as shown in the causal loop diagram 1.
POWER SUPPLY
Power supply = Power installed capacity x Load factor x (1 - losses) x ( 1 - power
used in auxilaries) x Kwh/Kw.
Power supply is a function of power installed capacity, load factor, losses and
power house auxilary consumption. Power installed capacity is a level which is
increased by incoming power installed inrate and decreased by derating rate with
time. Power in rate is increased by the power in pipe line which is due to the money
invested before the starting year of the model and the power inrate from money for
power generation. Money power is a level which is increased by money allocated to
power sector from plan funds. Money power outrate is a delayed function.of level by
9 years gestation period assumed to be for becoming the money effective in generat-
ing power. This money outrate is divided into money for generation and money for
transmission as per the feed back loops shown in the causal loop diagram 1. Money
generation divided by rate for conversion per MW gives the power installed capacity
inrate. The money for transmission decides the losses in transmission lines. As
losses increase there is much hue and cry to increase the money in transmission as
shown by the feedback loop from losses to money generation. Power supply/demand
ratio regulates the money for power and further increases the share of money for
generation as shown by two feedback loop from power sypply/demand, thus counter-
acting the effect of power losses loop. Delay in construction has been kept
constant as having exhausted all the available medium and large hydroelectric
sources, further demand has to be met from thermal power houses and agro based
mini-thermal power houses having delay period of two years as discussed by authors
(Garga et al. 1984, pp. 1250-52). Load factor has been assumed to be 608%.
RESULTS
Model has been validated by comparing the graphical results of various variables
viz:~ Power Installed Capacity, Power Supply, State Domestic Product, Tubewells
Connections etc. from 1970-71 to 1982-83. It seems that the basic structure of power
demand and supply feed backs has been captured in the model:
Graphical output of computer results for the demand and supply variables are shown
in the figure 2, 3 and 4. Variables plotted are, Demand for agriculture = DEMPW = 1,
Demand for industries = DMPIN = 2, Domestic demand = DODE = 3, Commercial demand =
COMDE = 4, Public Lighting = PBLIGT = 5, Power installed capacity = PIC = 6,
Required installed capacity = RINC = 7. Results are plotted from 1971 to 2011.
Figure 2 shows the results for the model base run, when all the loops are active.
Agriculture sector demand increases upto 1999, when the ground water potential
limits the tubewell connections to be provided. Thereafter demand for agriculture
remains almost constant. Demand for industry overtakes all the demands as the more
and more capital starts diverting to industrial sector as the agriculture capital
requirements saturates. Power installed capacity chases the demand but gap between
supply and demand increases every year. Despite the power supply/demand to money
power plan and money power generation loops are active, but power installed capacity
remains short of the demand.
Figure 3 shows the results for the normal run, when the feedback loops of power
supply/demand and power plan and money generation remains inactive. Here the demand
for agriculture sector goes on increasing as the power installed capacity is less
~296-
27-mARAB
MilbevYWAma @ VERSTOW 1,00
Pe 10 RUN@L
?
a Bihere bye 2% re
a 3
Popks3 CONDE=4 PELTGTES
1
1
rat
ore rH
° 7
= &
a a
= =
ObAPwe)
secon.
BASE RUN
2
FIG.
27-raR-8
9-19 RUN-RORMPL MINI-DYNAMG # VERSION 1.00
Rines7
Picas
OMPIMs2 DODE=3 COMME PBLIGT25
DENPWS1.
NORMAL RUN
3
FIG.
-297-
than the base run and so the tubewell connections provided have still not reached
the ground water potential limitation. Power installed capacity reduces drast-
ically, thus affecting the output capital ratio and reducing the state domestic
product drastically. Demand for power or required installed capacity has reduced
slightly from the base run as the capital which could have gone to power sector is
diverted to other industrial activity as there is no feed back from Power supply/
demand thus making a substantial part of the capital unproductive.
F- 28 KubeLuss —-MINDADYUANO © WERSICH 2.00 aTomakew
gue
DEMPw=) CAP4N2? DODES3. COMDESE PuLiGI=D nea?
G.uvye v7 vues 12549
929001 ra 3
ymca as eee oe e aweew 3 SED
fous . : :
4312 6 : :
cae : : :
a3 iz : : : :
312 : : F :
Gr. : : : :
43 048 3 : : :
$3 °1°2 . : : :
$3. ete : : 2 :
1961.0 S39 SaesZ se See me eee se ee ee wes ee eee eet
33° “81‘52° : . = °
$371 "9g 3 : 2 :
nr ve : 5 3
830 Gh iat > g :
33 er rad : ‘ :
33 er rat : : 3
ea3 é1 akg és : z
$y ee : > 3
333 at ; E : :
999.9 $a] = > ebede e)~ ee ee ee ee
$33 ed . S :
333 8 : : : :
3 : : 2 :
333 . : : :
i . 3 :
a 3 : :
2 7 z g :
: os : :
: 4 2
: : 3
2u1d 0
FIG. 4 LOAN RUN
Figure 4 shows the results for a run 3 based on assumption that state will get a
loan of Rs. 2000 million in 1990 at 1970-71 prices. This has been able to reduce
the gap between demand and supply but gap still persists. Power installed capacity
increases substantially but demand for power also increases more than the previious
2 runs.
CONCLUSION
In this paper a system dynamics model to understand the dynamics of power supply
and demand has been presented. Feed back loop from power supply/demand for affecting
the money in power sector is an important loop, which when effective regulates the
money to power sector and generation sector, thus increasing the power supply and
increasing the output capital ratio which inturn increases the state domestic
product. As even this loop has not been able to bridge the gap between demand and
supply, it seems that an alternative to bridge this gap and divert the money from
industrial capital to power sector as discussed by the authors (Garga et al.,1984,
pp. 1250-52) will be-worth mentioning here. That instead of starving the capital
-298-
of power, capital from the industrial sector will have to be diverted to power sector
may be even in private sector. Secondly agrobased mini power thermal stations will
also be economical to compensate for'the required increasing amount of coal for
thermal power houses. It is proved that economy, power supply and demand feedback
model can help in understanding the future behaviour of power and economy sectors.
Further runs can show the effect of other alternatives viz-a-viz change in power
supply,demand and state domestic product.
REFERENCES
Forrester, J.W. Industrial Dynamics, M.I.T. Press, 1961
Govt. of India, Planning Commission, “Report of the working group on energy policy",
Planning Commission, New Delhi, 1979.
Garga, K.K., Gupta, N.K. and B.Thapar, "Electric Energy Planning through System
Dynamics", I.E. Journal, Vol. XI, No. 2, 1982, pp. 59-61.
Garga, K.K., Thapar, B. and N.K. Gupta, “Economic Viability of small Hydro and
Thermal Power Houses in Punjab - A System Dynamics Study", Proceedings IEEE
International Systems Man and Cybernatics Conference New Delhi, 1984, pp.1250-52.
Pachauri, R.K., "The Dynamics of Electrical Energy supply and demand: An economic
analysis, New York: Praeger,1975.
Parikh, Jyoti. K.,"Modelling energy demand for policy analysis" New Delhi :Planning
Commission, 1981.
Rajadhyaksha, V.G., “Panel discussion on methodology for power planning Proceedings
of the Seminar on Planning for Electrical Energy, Central Electricity Authority,
New Delhi, 1977, pp.4.
