Cogeneration potential in Indian sugar mills: a case study

Nuruml Rc.mnrce.r Fonm. Vol. 22. No. 4. pp. 245-251. 1998

Pergamon PII: S0165-0203(Y8)00019-1

9 1998 Unated Nations. Published by Elrsv~erScience Lid. All rights reserved. R i m e d in Great Britain 01 6s-n~o1~n~~1~.oo+o.oo

Cogeneration potential in Indian sugar mills: a case study Anindita Roy Saha

The Indian sugar industry has massive potential for the cogeneration of electrical power. The sugar manufacturing process generates bagasse as a byproduct from cane-crushing. This bagasse can be used as a fuel f o r the boilers employed in sieam raising for the process use and cogeneration. In this study, the potential for power cogeneration in a typical plant under the existing boiler-turbine configuration has been estimated. An alternative configuration requiring capital investment in machinery has also been studied. The cost of cogenerating power, including the cost of fuel, operations and capital services wherever necessary, has been calculated for both these cases, using a linear optimisation method. The exercise allows for the use of multiple fuels, namely bagasse and coal, for cogeneration throughour the year including the cane-crushing season and the off-season. The per unit supply price of cogenerated power thus computed has been compared with the utility 's own cost with a view to selling ihe surplus cogenerated power to the grid. 0 1998 United Nations. Published by Elsevier Science Ltd. All rights reserved

The sugar industry has been identified as one of the cheapest sources of cogenerated power. Cogeneration in its simplest sense means the simultaneous generation of process heat and electricity. Technically, it means the production of electrical power and thermal energy from the same fuel. Sugarcane waste, namely bagasse and other waste such as fibre residue, crushed sugar and leftovers in the field can be used as a fuel for raising steam in the boiler house. This steam, in turn, can run power turbines after which the exhaust steam can be put to use in the sugar manufacturing process. The Indian sugar industry, being the largest in the world with 394 operating plants, has been identified as one with tremendous cogeneration potential, especially in the context of the grim power scenario in India (CEI, 1991). The cane-crushing season in India is from November to April, which will be referred to as the in-season period in the rest of this article. This coincides with the period of peak agricultural demand and minimum hydel power supply, and cogenerated power from sugar manufacturing plants can be used to supplement the energy supply from conventional sources. Moreover, the same infrastructure may be used during the off-season (May to October) utilising alternative conventional fuel and stored bagasse and barbojo (top and leaves of the sugarcane plant) to contribute to the country's peak summer load. The study conducted by RCG/Hagler, Bailley, Inc. (1987) estimated the potential for extra power generation in the sugar industry in only two Indian states, The author was at the time of writing a Research Scholar at the Centre for Economic Studies and Planning, School of Social Sciences, Jawaharlal Nehru University, New Delhi, and is at present Lecturer of Economics at the Indrapnstha College of the Delhi University, New Dethi, India

Gujarat and Maharashtra, as 42s MW. The Task Force for Formulation of the National Programme on Biomass-based Cogeneration in India, Ministry of Non-Conventional Energy Sources, The Government of India (1993) assessed a vast cogeneration potential of 3500MW in the Indian sugar industry. The National Federation of Cooperative Sugar Factories (NFCSF, 1992) estimated 3200 MW for the same. Such macroeconomic estimates may be explained through microeconomic exercises camed out on Indian sugar mills. A typical sugar plant, the Daurala Sugar Works, Uttar Pradesh, India, has been studied for this purpose. An attempt has been made to estimate the potential for suplus power and to compute the supply price of cogenerated power in the case of sales to the utility power grid. While the base case considers the existing boilerturbine combination already in operation in the plant for the above estimation, another possible case has been analysed with alternative configurations. The two cases thus reflect the existing potential and the possibility of capital investment in order to extract the maximum surplus power generation capacity both during the season and in the offseason.

The manufacturing process of sugar, plant capacity and machinery The manufacturing process in a typical sugar factory involves five basic processes: juice extraction, clarification, evaporation and concentration of juice, crystallisation of sugar and crystal separation. Bagasse results from the first step when sugarcane is crushed. The production of bagasse with 50% moisture amounts to around 28-30% of the

245

246

Cogeneration potential in Indian sugar mills: A. R. Saha

32KgIcm

2

prd2s

PADSU2

1

L

19.34Kqlcm wd4s

I b4s

PRDSU4 2

b 7s

-

Mill

PRDSX3

Turbines

'

Plant

p5

I

Y

Refinery

32s (Exhaust & Extraction)

b6s

prd3s

?

11 Kslcm

b5s

P3S

Chemical Plant

:pex~

u

1

Sugir Plant p5:entrifugals (Sulphur Burning. Drylng etc I

Sugar Plant

(Juice Healing. Evaporator. Pan Sraiion etc )

Figure I Steam balance diagram for sugar phnt operations is p& madel I .

cane crushed with respect to the total weight of the cane crushed'. The gross calorific value of bagasse is 2340 kcal kg-I, sufficient to generate between 2.02 and 2.20 kg of steam per kilogram of bagasse burnt. The use of bagasse in the boiler for captive power generation to meet the-in-house load of the plant is an existing practice. However, utilisation of the entire available bagasse to generate exportable surplus power is yet to be practised. Most of the existing sugar plants are equipped with boilers generating steam at parameters varying between 22 ata (atmospheres; 1 ata = 1.033 kg/cm2), 340°C and 45 ata, 440°C at the rate of 32-40 t h-' (tph). The high-pressure steam outlet from the boiler drives the turbo alternators and the low-pressure exhaust steam (at 1.5 ata) from these is used in the processes for juice heating and evaporation, and for driving fans and vacuum pans. The total consumption of high-pressure live steam is around 46.7% cane for the prime movers, 3% cane for direct process use in centrifugal stations, 0.1% cane as make-up steam and 2.2% cane as loss, giving a total steam consumption of 52% cane. The prime movers exhaust low-pressure steam of the same extent for use in the process. The captive power requirement of a typical plant with a cane-crushing capacity of 2500 t of cane per day (tcd) is around 3-3.5 MW while that of a 5000 tcd plant is 77.5 MW. The in-house power load is less than what can be optimally produced with the available bagasse. 'The unit of measurement in a sugar manufacturing process, i.e. the percent cane (Ti cane), is the percentage of usage in the process with respect to the total cane crushed.

The sugar plant under study has a capacity of 5000 tcd with allied organic and chemical plants, and a distillery. The ancillary units also draw power as well as low- and mediumpressure steam from the plant network. The net bagasse availability is 1650 t per day on average during a 165-day crushing season, with an average operational time of 23 h per day. There are seven boilers in total and five power turbines. Two spreader stoker travelling grate boilers (B 1 and B2) supply 40 tph steam each at 42 kg/cm' and 400°C. One spreader stoker travelling grate boiler (B3) supplies steam at 32 kg/cm2 and 400 2 12°C to the steam header while three 30 tph boilers (B4, B5, B6; one horseshoe furnace and two spreader stoker) supply steam at 19.34 kg/cm2 and 315°C. There is also one 10 tph boiler (B7) with steam parameters of 20 kg/cm2 and 300°C. The capacity utilisation on the whole is around 69% and the efficiency is around 6568%. No boiler has a bagasse drier and all can be operated with multifuel, e.g. bagasse and coal. The plant has three backpressure steam turbines (BPST) each with 2.5 MW capacity, one 3 MW BPST and one 4.5 MW condensing extraction steam turbine (CEST). Four pressure-reducing stations (PRDS) are used to step down the steam pressure wherever necessary in the process. The 42 kg/cm2 header supplies steam to the two 2.5MW BPSTs, the 32 kg/cm2 header through the pressure reducer PRDS1, and the 11 kg/cm2, header through PRDS4. The 32 kg/cm2 header supplies steam to the 3 MW BPST, the 4.5 MW CEST and the 19.34 kg/cm2header through PRDS2. The next steam header at 19.34 kg/cm2 supplies steam to the mill turbines at the rate of 35.96 tph, the organic plant at

Cogeneration potential in Indian sugar mills: A. R. Saha

42KqIcm

32KgIcm

bls

2 r

247

b2s 1

2

2

b 7s

-

19.34Kq/cm

PRDS#4 '

I

Plan! 4

I

1 1 Kq/cm

I

2

-

I

: pexrr 1 SKg/cm

(Exhaust & Extraction)

.

I

Relinery , 8 . Dislillery

Chemical Plant

I

;

I

I I

Sugar Plant (Sulphur BurnincJ Centrifugals Dryi.q c.T.)

pexr3

Sugar Plan1 (Juice Healing. Evaporalor, Pan Slalion eic 1

Figure 2 Stcani balance diagram for sugar plant operation as per model 2.

2.75 tph, one 2.5 MW BPST and the I 1 kg/cm2 header through ,PRDS3. The low-pressure steam supply at 11 kg/cm- goes to the centrifugal station, the sulphur burning and sugar drying units of the sugar plant at 6.69 tph, the chemical plant at 3.09 tph and the refinery and distillery at 0.6 tph. The total generation of exhaust steam at 1.5 kg/cm2 from the various power and mill turbines is around 138.96 tph. The various power turbines and mill turbines generate exhaust steam according to their steam inlet and power generation. Figure 1 describes the steam supply and consumption flow (steam balance) under the current machine configuration. The power load of the plant under study is 9.4 MW, of which 7.2 MW goes to the sugar plant and 2.2 MW to the allied plants. At present the plant does not generate surplus power. The case study presented here thus aims at finding 1. the potential for additional power cogeneration with the present infrastructure; 2. the potential for power generation during the off-season with alternative fuel; and 3. the possibility of capital investment in improving machine designs in order to attain a higher energy efficiency and power generation. An alternative configuration with all turbines of the con-

densing extraction type has also been studied. The use of CEST can avoid the wastage of exhaust steam from back pressure turbines when there is no demand for the same in the process, i.e. during the off-season. The advantage of

CEST over BPST lies in the facility of extracting steam at the relevant pressure in case there is a demand during the inseason. CEST also offers the facility of condensing steam in the absence of a demand for low-pressure exhaust steam in the off-season. Therefore, another model (model 2) has been conceived as one with four CESTs each with 4.5 MW capacity, giving a total of 18 MW, in contrast to the existing 15 MW capacity model (model 1). The same seven boilers have been retained in both the models, since these are aIready capable of multifuel operation. Coal may be used for firing these during the off-season while bagasse and coal combinations may be used during the in-season for both the models. Figure 2 describes the steam balance conceived for model 2. While the boilers, pressure-reducing stations and the steam load points remain the same as in model 1, the configuration of the power turbines has been altered. One CEST is fed by the 42 kg/cm2 steam header and two others are fed from the 32 kg/cm2 header. The fourth CEST is located at the 19.34 kg/cm2 header. These two models have been analysed as alternative options for cogeneration in which the former is the existing one and the latter is a possibility.

The optimisation problem The method used for the estimation of cogeneration potential and computation of the cost is linear optimisation (Hadley, 1969). The former has been formulated as a power output maximisation problem, constrained by the steam and power

248

Cogeneration potential in Indian sugar mills: A. R. Saha

balances between generation and consumption in the process. Further constraints are placed by the availability of fuel and capacity limits of the machines. A subsequent optimisation exercise minimises the fuel cost for generating power only to meet the in-house requirement, subject to the same constraints. While the latter exercise gives the minimum cost of power generation for the in-house demand, the results of the maximisation exercise may be used to compute the fuel cost of generating the maximum quantity of power under a given model. The cost differential gives the additional fuel cost of cogenerating power over that of generating for inhouse use only. By dividing the total additional cost by the amount of surplus cogenerated power, one may compute the per unit fuel cost of the exportable surplus power. The final supply price of the cogenerated power may be computed by adding the per unit fuel cost to the costs of capital, operations and maintenance. The cogenerator's investment is justified if the sale of extra power earns a profit or at least breaks even. The optimisation methodology (Kambo, 1984) may also be used as a valuable tool in estimating the optimal fuel mix of byproduct bagasse and purchased coal (if any) which would maximise power output or minimise the fuel cost of generating power for process use only. It is easy to visualise that one may further extend this method to compute optimal fuel mixes in order to minimise cost to meet any given load or to maximise power output in any operating environment of cogeneration. The cost minimisation problem does not exist during the off-season when there is no power requirement in the plant. Accordingly, the optimisation problem for the off-season reduces to only the estimation of the potential for maximum power output based on stored bagasse andor purchased fuels. The fuel cost of cogenerated power during the offseason and the corresponding supply price may be computed directly. The power output maximisation exercise is to maximise Pi 1

where p i denotes the output of turbine i. The subsequent optimisation problem is to minimise

their respective capacities. The steam balances put further constraints as the process steam demands must be met at the relevant pressure headers. The availability of the fuel offers the last set of constraints. The fuel constraints for bagasse and coal state that the total fuel consumption in the seven boilers does not exceed the total availability of bagasse ),(f and coal (cf). In symbols, one may write

(3) and -&JSCf, i

i = l , 3, ..., 7

(4)

While fmVr is determined by the plant's cane-crushing capacity, cf can be purchased in any amount. The multifuel boilers generate steam according to the steam-bagasse andor steam-coal ratio. The steam output (bi)of the boilers may thus be written as

bi=aif;++icJ, i = 1, 2, ..., 7

(5)

where bi is the steam output of the boiler i, ai is the steambagasse ratio and is the steam-coal ratio. The power output ( p i )of turbine i can be expressed as (Tribus, 1961)

+;

p i = pipsi

+ B,pexri + di

(6)

where psi is the rate of flow of steam at the inlet of turbine i, pexfi is the extraction from turbine i, p i and hi are the respective power generation coefficients and d i is a constant. However, the term pexti vanishes for back pressure turbines where there is no extraction. The constraints due to the operational capacity of the boilers and turbines have been derived from standard industrial practices (Spalding and Cole, 1958). The boilers are assumed to operate between 30 and 100% of the maximum capacity while the turbines operate between 20 and 100%. Accordingly, one may write the capacity constraints as 0.3(bim,,) 5 bj 5 bjmax

(7)

and where r c is the price of coal, xb is the price of bagasse, cfiis the coal fuel feed to the boiler i and f i is the bagasse fuel feed to the boiler i. The price of bagasse, which is a byproduct of the process, has been considered as a component of the cost because bagasse has an alternative sale price in the market. Therefore, for the producer, bagasse has an opportunity cost of being otherwise sold, say, to a paper plant and thereby earning revenue. The cogenerator foregoes this revenue but would try to recover it while setting the supply price of power. Hence the market price of bagasse constitutes a part of the cost to the cogenerator. The alternative sale price of bagasse has been taken as Rs. 250 t-' as in April 1995 at the time of the field study. The price of boiler coal is Rs. 850 t-'.

0.2t Oim ax) 5 Pi 5 Pimax

(8)

where bimpnand are the maximum capacities of the boilers and turbines, respectively. The crucial constraints in the optimisation problem come from the steam load balancing at each header. Due to the difference in turbine configurations between model 1 and model 2, the steam balances also vary (see Figures 1 and 2). The respective steam balance equations state that the supply of steam from various boilers andor pressure-reducing stations at specific headers have to meet the demand for steam at the inlets of the power turbines, pressure-reducing stations and consumption points in the proc,ess. The fiv: steam balance equations for the 42 kg/c,m', 32 kg/cm-, 19.34 kgkrn', 11 kg/cm2 and 1.5 kg/cm- (exhaust and extraction) headers in model 1 may be formulated as

+ b2 1psl +psz + p r d s l +pr&

Constraints

bl

The optimisation exercises are constrained by the inputoutput relationships of the boilers and turbines, and by

63 +prds, 2 ps3 +ps4 +prds2

(9) (10)

Cogenerationpotential in Indian sugar mills: A. R. Saha

prds3 +prdsJ

If + sp + cp

2

(12)

i = 1, 2, 3,5

ci 5 psi,

c4 5 pcxt4

c, Imt

z c i 2 138.96,

i = 1, 2, ..., 5, m

1

where bi is the output of boiler i, pi is the output of turbine i, p s i is the input for and also the outlet steam of turbine i, prdsi is the steam passing through pressure-reducing station i, op is the steam demand at the organic plant, mr is the

steam demand at the mill turbine, rf is the steam demand at the refinery, sp is the steam demand at the sugar plant at 11 kg/cm2, cp is the steam demand at the chemical plant, ci is the steam consumption-point at i at 1.5 kg/cm2, and p a t i is the extraction at 1.5 kg/cm2 from turbine i . The operations of the plant under study specify up+mt=38.71 tph in Equation (ll), I f + s p + c p = 10.38 tph in Equation (12) and c,=mt=35.96 in Equation (13). The corresponding steam balance equations in model 2 (see Figure 2) may now be written as b,

+ 62 2 psl +prds1 +prds4

(14)

where all the symbols have the meanings specified earlier.

+ +ps3 +prds? b4 + b5 + h6 + 67 +prdsl 2 ps4 +prds3 + 38.7 1 b3 prds, 2 ps2

pr& +prds4 2 10.38 pexti

5

psi

1 ci+c,,

2

138.96 i = 1,2,3,4

(15)

(16) (17)

i

The cost minimisation problems have an additional constraint due to captive power load, namely

z

pi 2 9.4

i

Results Model 1 shows a maximum power output of 15 MW which implies an excess of 5.6MW over the in-house load of 9.4 MW. All turbines operate at the respective maximum capacities. All the boilers except B6 generate steam at near maximum, giving a total live steam output of 189.37 tph. This includes the steam for captive power generation, process use, and for the additional cogeneration. A total pass-out steam of 149.6 tph meets the low- pressure steam load of the plant. Thus, the solution provides an excess cogenerated power after adequately meeting the plant's live and exhaust steam load. The most striking feature of the optimal solution lies in the optimal use of bagasse and near-zero use of coal (8.56 tph in Bl). The definite advantage of having by-

249

product bagasse as fuel is thus established, together with having an optimal fuel mix. The cost of fuel consumption per hour has come out to be Rs. 24562 for 15 MW power output. The subsequent fuel cost minimisation exercise gives a cost of Rs. 17 460.75. The cost differential of Rs. 7101.25 may be accounted for by the additional 5.6 MW of cogenerated power. This gives a fuel cost of Rs. 1.26/per kWh. The power output of 15 MW during the off-season is associated with a cost of Rs. 26 879.23 per hour due to the use of purchased coal. This gives a fuel cost of Rs. 1.79 per kWh. However, the optimal solution gives a total of 113.62 tph backpressure steam which has no demand in the process during the off-season. Such a massive loss calls for the alternative options of having all CEST-type turbines. CESTs provide the facility of extracting low-pressure steam during the season and no extraction in the off-season. Model 2 with 4 X 4.5MW CESTs gives a maximum power output of 18 MW in season. All the boilers and turbines operate at near optimum. The optimal fuel mix consists of total available bagasse of 7 1.76 tph and a marginal amount of 11.13 tph of coal. The total fuel cost is Rs. 27388. The subsequent fuel cost for generating 9.4 MW is Rs. 20265.29. One may now compute the cost of the additional 8.6 MW of cogenerated power as Rs. 0.83 per kWh. Using all CESTs of higher installed capacity lowered the cost of fuel per kilowatt hour. The fuel cost of generating 18 MW power during the off-season has been obtained from the optimisation exercise as Rs. 1.14 per kWh, which is much lower than that computed using model 1. Therefore, in terms of fuel cost, it may be said that model 2 (with all CEST-type turbines) offers definite cost-economy over model 1 with four BPSTs and one CEST during the in-season as well as the off-season. It may be mentioned here that the results of the optimisation exercises have been found to be sensitive in the range of 225%. However, the overall comparison of the two options would depend on the final supply price of cogenerated power, which includes a capital cost component and other operating costs. This analysis is highlighted in the next section.

The cost of capital and final supply price of cogenerated power The two models under study may be considered as options for greenfield (i.e. newly constructed) plants as well as brownfield plants2 (i.e. already in operation). While greenfield options would require capital investment for fhe entire machinery, the brownfield plants need partial replacement of the turbines only. The greenfield options for the two models would mean capital investment for 15 MW and 18 MW respectively. On the other hand, model 1 will have no investment in case it is considered as a brownfield option. Model 2 will require an investment for 13.5 MW capacity due to the installation of three CESTs over the existing one. The cost of fixed capital services has been computed as capital investment amortised over the lifetime of the power plant of 15-20 years. A 16% rate of return with a debt'Greenfield plants refer to newly constructed plants whereas brownfield plants signify plants already in operation.

250

Cogeneration potential in Indian sugar mills: A. R. Saha

Table 1 Cost and potential of cogenerated power under alternative options in a typical 5000 tcd sugar plant ~~

Model 1

Model 2

5.6

8.6 18.0

1. Export of power (MW) In-sexion

Off-season 2. Fuel C O . Y ~(Rs.per kWh) In-season Off-SCdSOll

3. Cost of capital (Rs.per kWh) ( a ) In-season Greenfield Brownfield (b) Off-season Greenfield Brownfield 4. Total cost (Rs.per kWh) (a) In-season Greenfield Brownfield ( b ) Off-season Greenfield Brownfield

15.0

I .26 1.79

0.83 1.14

0.82 0.10

0.86 0.55

0.86 0.14

0.89 0.58

2.1 I

I .73 I .42

1.37

2.68 I .94

2.07 1.75

equity ratio of 1:l has been considered as the opportunity cost of capital3, as applicable to the Indian power sector. A three year gestation lag and a 5% salvage value has been assumed with an investment phasing of 20%, 30% and 40%, respectively, in the first three years, and 5% in the eighth year. A further 5% investment in the 14th year extends the plant life to 20 years. The generation of output is phased as 70, 90, and loo%, starting from the fourth year. The installation cost of power plants in the sugar industry is Rs. 30m/MW (BHEL). The cost of BPST is Rs. 5m/MW, that of CEST is Rs. 7.5m/MW (approximately) and that of a high-pressure spreader stoker boiler is around Rs. 20m. The cost of working capital has been considered to be three months’ block of works cost (materials, fuel, nonfuel utilities, operations) at a 16% rate of interest. The total cost of capital is the sum of fixed and working capital servicing charges. While the former varies from greenfield to brownfield plants, the working capital cost varies with the season as the fuel cost is seasonally dependent. The other elements of cost include those of repair and maintenance at a rate of 2.5% of the total project cost, and another 2.5% as salaries, wages and administration of cogeneration. Table 1 summarises the components of the costs, supply price and the potential of the cogenerated power under the two models.

Conclusions The major result of the exercise is an estimate of the potential for cogenerating exportable surplus power with ?he Government of India has declared a 16%rate of return for the power sector as a whole. The non utility co-generators are expected to fetch equity capital at a I : 1 debt-equity ratio under the various attractive schemes of facilities. incentives and concessions offered by the government in the transactions of cogenerated power in particular. While a 4 : 1 debt-equity ratio may be applicable to the Indian power sector in most cases, a 1 : 1 ratio may be justified for the economically viable cogeneration schemes only.

the two cogeneration options considered. Moreover, the entire amount of cogenerated power is available during May to October (the off-season) through the use of alternative fuel while the byproduct bagasse is usable during the in-season. The final supply price of bagasse-based cogenerated power is as low as Rs. 1.37-1.42 per unit for the existing plants with partial retrofit and replacement. The supply price in-season is still moderate at Rs. 1.73-2.11 per unit even if the plant is constructed afresh. The supply price of coal-based cogenerated power during the off-season is still not high. It varies between Rs. 1.75 and Rs. 1.94 per kWh for brownfield plants and between Rs. 2.07 and Rs. 2.68 per kwh for new options. Comparing the two models of cogeneration considered, one may infer that an all-CEST-type turbine combination is more cost-effective except in the case of the existing infrastructure during the in-season. It is also seen that model 1 no longer remains the cheaper option during the off-season even for the brownfield plant because model 2 avoids wastage of backpressure steam during the off-season. Thus, model 2 emerges as a more economic option, if one considers both from the perspective of year-long power generation. While the in-season operation may be termed cogeneration, the off-season one is simply power generation with the same infrastructure. The implementation of cogeneration projects in Indian sugar mills, however, will depend not only on its internal cost economics but also on the terms laid down by the utility. The utility’s offer price, which may be based on its own avoided cost of power generation, has to be at least equal to the cogenerator’s supply price in order to make it attractive. The results presented in this exercise have revealed costs of cogenerated power to be less than the Government of India’s declared rate of Rs. 2.25 per kWh (1994). Provided such a price is ensured by the electricity boards, buying of cogenerated power is guaranteed and other facilities like banking, wheeling and backing-up etc. are provided, the Indian sugar mills may demonstrate vast cogeneration potential. In view of the country’s grim power situation, the sugar industry needs to be encouraged to contribute significantly to the grid by supplying cheap cogenerated power.

Acknowledgements The case study presented in this article was undertaken as part of the research work camed out towards the author’s doctoral thesis, entitled “The Economics of Energy Conservation and Industrial Cogeneration of Electricity: Case Studies of Certain Selected Industries in India”, accepted by the Centre for Economic Studies and Planning, School of Social Sciences, Jawaharlal Nehru University, New Delhi, India, 1997.

References CEI ( 199 1) Proceedings of the National Conference on Cogeneration (Technical Set). New Delhi. January. Confederation of Engineering Industries. Government of India (1993) Report of the Task Force for Formulation of National Programme on Bio-mass-based Cogeneration in

C o g e n e r a t i o n potential in I n d i a n s u g a r mills: A. R. Saha India, October. Ministry of Non-Conventional Energy Sources,

New Delhi. Government of India (1994) Report on the National Programme on Bagasse-based Cogeneration, January. Ministry of Non-Conventional Energy Sources, New Delhi. Hadley. G.(1969) Linear Programming. Addison-Wesley. Kamho. N.S. (1984) Mathematical Programming Techniques. Affiliated East-West Press Pvt. Ltd.. New Delhi. NFCSF (1997) Report on Cogeneration. National Federation of Cooperative Sugar Factories, New Delhi.

25 1

Spalding. D. B. and Cole, E. H. (1958) Engineering Thermodynamics. Edward Arnold. London. RCGIHagler, Bailley. Inc. (1987) Non-utility Power Generation in the Indian States of Gujarat and Maharashtra: Potential Impediments and Policy Issues, prepared for the Department of Non-Conventional Energy Resources, June. Sponsored by the US Agency for International Development, Washington DC. Tribus. M. (1961) Thermostatics and Thermodynamics. Van Nostrand. Princeton, NJ.