Chapter-9 PARAMETRIC SENSITIVITY ANALYSIS

Transcription

1 151 Chapter-9 PARAMETRIC SENSITIVITY ANALYSIS 9.1 Introduction The MARKAL Business As Usual (BAU) case presented in Chapter 5 provides a projection of the evolution of the Indian energy system from 25 through 245. The BAU case was generated using best estimates for the values of model inputs, such as the characteristics of current and future technologies, energy service demands, and regulations on criteria pollutant emissions. Since the true values for many of these inputs are unknown, the BAU case represents only one of many possible outcomes. Further, it does not itself convey information regarding the sensitivity of the energy system to input variations and assumptions. This chapter describes the application of formal sensitivity analysis techniques to evaluate the model s response to changes in input assumptions. The results aid in characterizing and communicating the drivers that leads to such outcomes as: the penetration of particular technologies, the increase in fuel prices, efficiency improvement in new power generation technology, or an increase in availability factor etc. Sensitivity analysis also allows one to view the BAU case in the context of the range of possible future energy scenarios that may occur. 9.2 Sensitivity Analysis Background In broader sense there are two types of sensitivities: Global and Parametric. Global sensitivity analysis techniques typically are applied in practice when the goal is to characterize the relationships among model inputs and outputs over a wide range of input conditions. In contrast, parametric sensitivity analysis, also known as local sensitivity analysis, is used to evaluate the response to a change in a single input, holding all other inputs constant [1]. To evaluate sensitivities within the Indian database and model, parametric sensitivity techniques have been used in this chapter. Global sensitivity analysis involves perturbing multiple model inputs simultaneously and evaluating the effects of each input or combinations of inputs on model outputs. Inputs are often perturbed via Monte Carlo simulation. In Monte Carlo simulation, statistical or

2 152 empirical distributions are assumed for inputs of interest. A value is sampled from each distribution, and the resulting set of values is fed into the model. The values of relevant outputs are recorded. The combination of a set of inputs and the corresponding outputs constitutes one potential realization. Typically fifty to several hundred realizations are generated in a Monte Carlo simulation. These realizations are then evaluated using visualization and statistical techniques to characterize the nature and strength in the relationships among inputs and outputs. Parametric sensitivity analysis, in which one input is perturbed while others are held constant, is very useful in characterizing incremental responses to changes in inputs from a base or reference case. These responses can be characterized quantitatively, such as with a sensitivity metric or empirical derivative, or graphically. While parametric techniques do not characterize responses over combinations of inputs, they often play an important role both in preliminary analyses, as a cursory means to identify sensitivities of interest, and in more detailed analyses of input-output responses. Global and parametric sensitivity techniques can be used independently or together. Such an analysis is required to provide detailed information about impacts of changes in specific inputs in the context of the BAU case. 9.3 Inputs for Sensitivity Analysis in MARKAL Inputs included in the analysis are: discount rate, the future costs of natural and, the hurdle rate for new electric generation technologies, future investment cost, the availability factor and efficiency improvement of clean technologies. Each of these inputs was expected to have an impact on the outputs i.e. future resource allocations. The next step in the analysis was to characterize ranges for the inputs. Each of the inputs, which were assumed to be independent of each other, was represented with a uniform distribution. Discount rate is the most important parameter for sensitivity analysis because this affect whole of the energy system cost and allocations. A range from 8% to 15% has been taken in consideration for MARKAL simulations. The considered range for other inputs is summarized in Table 9.1.

3 153 Table 9.1 Input parameters Considered for sensitivity analysis in MARKAL Input Parameters Units Range Discount Rate percentage Efficiency percentage percentage 34-5 Availability Fraction Fraction.8-.9 Hurdle Rate percentage 5-25 Fuel Prices Oil US$/barrel 3-15 Gas US$/thousand cubic feet 3-8 Nuclear Investment Cost US$/kW The hurdle rate for new electricity generation technologies ranged from.5 to.25[2]. Nuclear power investment cost has been taken from 116 $/kw to 3$/kW [3-5].The wide variation in investment cost is taken because of different reports in India as well as in the world. The clean technologies and still under test in India. For the efficiency range has been taken from 38 to 58% while efficiency varies from 34 to 43% [6].The availability factor will improve for these technologies in future. On the basis of literature studies it has been taken from.86 to.89[7]. 9.4 Parametric Sensitivity Analysis and MARKAL Results Parametric sensitivity analysis was carried out to evaluate the effect of incremental changes in each of these inputs on the primary outputs of interests. By evaluating changes parametrically, sensitivities around the baseline run were characterized. Results of the parametric runs are provided in next subsections.

4 Discount Rate This parameter has its importance not only in this work but in every economic activity having dynamic aspects. This factor governs the time dependent value of money which means the worth of money at any point of time can be calculated through this factor. Value of this factor is usually taken equal to the interest rates offered by banks on deposits and for calculating worth of money at any point of time approach of calculating compound interest can be used. For the reverse case, i.e. for converting costs at any point of time in future, this factor is used in a reciprocal way and its worth of money is reduced at this rate when bringing it backwards. In MARKAL, as minimization of total discounted system cost is the objective function, this factor becomes most important. Therefore, variation in results with four other values of discount rates has been obtained. One on the lower side 6.5% as this rate is equal to the rate of inflation in costs taken in this analysis. This lower rate brings inflation and discount rate equal leading to a situation where effectively there would be no change in costs as the increase in costs due to inflation is equal to the increase in worth of money if forwarded to future time period. This situation is quite possible in future as the interest rates paid by banks in India have come down as compared to the rates few years back. On the other hand the changes in economic conditions of the country may force the banks to revert to higher interest rates as there is a reduction in saving tendency with reduction in interest rates. Therefore, on the higher side 1%, 12% and 15% discounts rate have been considered. The results are shown in Fig.9.1. With the low discount rates of 6.5%, the share of in the energy mix decreases from 283GW in the base case(discount rate 8%) to 218GW in the year 245.The another major change in the energy mix is the proportion of and other renewable. The other renewables share was 134 GW in the base case.however; these do not get allocations at all in 6.5% discount rate scenario. On the other hand the gets negligible (1.9GW) allocation in base case with a share of 123GW with 6.5% discount rate. The emerges as largest power production technology in this scenario while was the largest power production technology in the base case. Thus the conventional based technology is not able to overtake as the disadvantage of relatively high O&M cost due to fuel costs remains equally important all the times in the projected period.

5 Coal Gas Oil Large hydro Nuclear wind Small hydro Other renewable Installed Capacity Disc.1% (G W ) Coal Gas Oil Large Hydro Nuclear Wind Small Hydro Other Renewable (a)resource wise installed capacity with 6.5 % (b) Resource wise installed capacity with 1 % discount rate scenario discount rate scenario 8 8 Installed Capacity Disc. 12% (G W ) In s ta lle d C a p a c ity D is c.1 5 % ( G W ) Coal Series2 Series3 Series4 Nuclear Wind Small Hydro Other Renew able Coal Gas Oil Large Hydro Nuclear Wind Small Hydro Other Renewable (c)resource wise installed capacity with 12 % (d) Resource wise installed capacity with 15 % discount rate scenario discount rate scenario Fig.9.1 Resource wise installed capacity with various discount rate scenarios. In case of high discount rates, the share of technology remains almost constant with as shown in Fig.9.1 (b), (c) and (d). This means conventional power plants does not remain important for future points of time due to heavy discounting. The was the most preferred electricity production technology in 6.5% discount rate scenarios. But with higher discount rates this technology share gets reduced i.e GW in the all three (1%, 12% and 15%) scenarios for the year 245. The which requires higher initial investment as

6 156 compared to does not get any allocation as its advantage of low O&M cost during its life does not get importance due to large discounting. The other renewables (solar, biomass and waste to energy etc.) becomes major component of the energy mix from 23 onwards in high discount rate scenarios. The contribution of other renewables is 2GW (27 %) for 1% discount rate. The Oil thermal was not getting any allocation in base case becomes important in higher discount rate cases. The share of power production technology increases with increasing the discount rate Efficiency Improvement in and The clean technologies are still under development in India as discussed earlier.only demonstration plants have been installed. Due to prototype technology, and show varying efficiencies in different countries. In India at a presentation [8] it has been estimated to be 4.6 %on the average. In an earlier study [9] a range of 38 to 58% for and for the 34 to 5% was presented. The results of MARKAL simulations are shown in Fig.9.2.The first three figures show the results of variations in efficiencies of an power production technology. With increase in efficiency of the technology share in energy mix decreases slightly. This may be the consequence of the efficiency improvement. The other renewable allocations slightly increase with increased efficiency of. The technology in these cases does not get any allocation in the energy mix due to high investment cost. All other allocations remain same as in the base case. The next three figures (Fig.9.3) show the resource allocation in the case of efficiency variations in the. In these scenarios the share in energy mix remains same. The drastic change is in proportion of other renewables. The other renewable share is about 4% of total electricity generation in the year 245. The power technology does not get any allocation where as, gets around 5GW out of 752 GW allocations.

7 advance In s ta l l e d C a p a c i ty ( G W ) advance (a) Resource wise installed capacity with 38% (b) Resource wise installed capacity with 44% efficiency scenario. efficiency scenario. Installed C apacity(gw ) advance (c)resource wise installed capacity with 58% efficiency scenario. Fig. 9.2 Resource wise installed capacity with various efficiency scenarios. In s ta l l e d C a p a c i ty ( G W ) advance In s ta l l e d C a p a c i ty ( G W ) advance (a)resource wise installed capacity with 34 % (b) Resource wise installed capacity with 43 % efficiency scenario. efficiency scenario

8 advance (c)resource wise installed capacity with 5 % efficiency scenario. Fig.9.3 Resource wise installed capacity with various efficiency scenarios Hurdle Rate for New Technologies Hurdle rates refer to the discount rates applied to the investment cost of new technologies which are meant to mimic hesitancy on the part of the purchaser to invest in a newer technology over an established technology. Therefore, hurdle rate is basically the technology specific discount rate which is implemented to make a new technology attractive for power generation. Three hurdle rates applied (5%, 18% and 25%) and the resource allocation are shown in Fig 9.4(a),(b) and (c).the first two i.e. 5% and 18% show no change in the proportion of technologies. This means that up to 18% no new technology becomes more attractive options. The 25% hurdle rate shows slight increase in the proportion of new technologies (~ 55GW for ) advance In s ta l l e d C a p a c i ty ( G W ) advance (a) Resource wise installed capacity with 5% (b)resource wise installed capacity with 18% hurdle rate scenario. hurdle rate scenario.

9 159 8 Installed C apacity(gw ) advance (c)resource wise installed capacity with 25% hurdle rate scenario Fig. 9.4 Resource wise installed capacity with various hurdle rate scenarios Availability Factor for and Availability refers to the percentage of the year that a technology is online and available accounting for forced and scheduled outages. Installed C ap acity (GW ) advance advance (a)resource wise installed capacity with 86% (b) Resource wise installed capacity with 89% availability of scenario. availability of scenario. Installed C apacity(gw ) advance advance (c) Resource wise installed capacity with 8% (d) Resource wise installed capacity with 8% availability of scenario. availability of scenario. Fig. 9.5 Resource wise installed capacity with various availability factor of and.

10 16 Two scenarios for, with 86% and 89%, have been developed to see the effect of availability in energy mix. There is not much change in the energy mix due to availability variation in the two scenarios as seen in Fig.9.5 (a) and (b).coal remains the same as in the base case but proportion in energy mix changes slightly about 1.2 %. Two scenarios for availability has also been developed as shown in Fig.9.5 (c) and (d). Very little (about.8%) allocation of has been seen in the energy mix in the year 245. There is a small decrease in the contribution to the power generation Increasing Fuel Prices due to Uncertain Market The present costs of as compared to are higher which is restricting it to get any allocations in the mix power source scenario. There exists a possibility of reduction in prices of due to the success in the ongoing exploration activities. Therefore, prices of were lowered at 3$/barrel scenario to see the effect on the model. The prices are also taken as high as 15$/barrel because of any embargo to happen in future just like in the year 28. The results have been shown in Fig.9.6.There is a slight change in proportion in energy mix. As the price of rise the little decrease in technology and corresponding increase in. Small hydro contribution to the energy mix increases with lower prices. Wind and remains as same as in the base case. In Indian power sector the contribution to the power is very negligible i.e.1.2 GW out of 127 GW in the year 25.Due to negligible proportionate very little effect can be seen in the energy mix In s ta l l e d C a p a c i ty ( G W ) (a)resource wise installed capacity with price increases to 15$/barrel scenario. (b) Resource wise installed capacity with price increases to 8$/barrel scenario.

11 (c ) Resource wise installed capacity with price 3$/barrel scenario Fig.9.6 Resource wise installed capacity with various price scenarios. Gas thermal, on the other hand, amounts to be 11.9GW in the installed capacity in the year 25.Therefore; a significant variation can be seen due to price fluctuation of. The installed capacity has been shown in Fig.9.7. Decreases in natural costs led to an increase in natural use in electricity generation, at the expense of and as shown in Fig.9.7 (a). Increasing natural costs appeared to drive other renewable use in electricity generation while at the same time resulting in decreased use of. Natural use decreased with cost, although the decrease was comparatively modest. These observations suggest that the increased natural cost may have triggered an emissions-related modeling tipping point that favors use over. At decreased cost (3$/thousand cubic feet) of the becomes preferred power production technology after the year 23.Other renewable come into picture only after year 24.as in the base case. Nuclear remains the same as in base case. The increased cost of preferred other renewable rather than to use. Gas thermal completely disappear from the energy mix.

12 162 In sta lle d C a p a city(g W ) (a)resource wise installed capacity with Gas price 3$/thousand cubic feet. (b) Resource wise installed capacity with Gas price 8$/thousand cubic feet. Fig.9.7 Resource wise installed capacity with various Gas price scenarios Nuclear Power Plant Investment Cost Due to Indo-US deal, the technology suddenly becomes important in the Indian power sector. The investment cost per unit of installed capacity varies in different countries due to labour, operation and maintenance cost and sometimes due to immature technology. Therefore, the price variation available (Table 9.1) in literature has been considered for the sensitivity analysis [1]. The results are shown in Fig.9.8. A slightly lower cost with respect to base case gives allocation of in the energy mix. Coal remains same as in the base case. A rise in the investment cost of the power plant decreases the use of proportion in the energy mix. Further increase in the investment cost makes completely disappear from the energy mix as shown in the Fig.9.8(c).On the other hand, the emerge as major electricity producer. Large hydro installed capacity is about 5% of the total installed capacity in the year 245.Other renewable remains constant in all three cases.

13 In s ta l l e d C a p a c i ty ( G W ) (a) Resource wise installed capacity with investment cost 116 $/kw. (b) Resource wise installed capacity with investment cost 2 $/kw (c) Resource wise installed capacity with investment cost 3 $/kw. Fig.9.8 Resource wise installed capacity with various investment cost scenarios. 3.5 Summary of Observations from Sensitivity Analysis The parametric analysis of the BAU case provides considerable insight into the innerworkings of the MARKAL model and the response of the model to alternative input assumptions. In most cases, the results of these analyses confirmed expected behavior in the model. Additionally, the results provided insight into the complicated response of the system to criteria pollutant emission limits indirectly. In this section, we summarize many of the key observations from the sensitivity analysis.

14 164 The lower discount rates as compared to BAU case prefers as the main component in energy mix while experience stagnation in the planning period. In higher discount rates the is the major source of energy production technology. Increasing efficiencies of prefers the other renewable energy technologies while in case of increasing efficiencies of prefers as energy technology. Availability factor increase of new technologies does not change the proportion in energy mix of various technologies. Coal remains almost constant in each case. Due to market fluctuations, if the prices increases the proportion becomes zero in the energy mix and the also decreases. The electricity generation hurdle rate has an additional impact on future-year energy sector technologies. Increasing this rate effectively makes it more difficult for new, more efficient technologies to penetrate the electricity generation market. The uncertainty in investment cost of affect the penetration in the power sector. Very high cost gives no allocation in energy mix. The becoming most competitive and covers about 5% market of total installed capacity. Sensitivity analysis such as is presented here has been a useful component of the MARKAL model database development and quality assurance. By identifying key drivers and interactions, sensitivity analysis facilitates an understanding of how the model responds to alternative input assumptions which, in turn, aids model refinements. Reference [1] Timothy L.J.et al (26) MARKAL Scenario Analyses of Technology Options for the Electric Sector: The Impact on Air Quality United States Environmental Protection Agency Office of Research and Development Washington DC 246. [2] Shay, C. et al (26) EPA U.S. National Database: Database Documentation, U.S. Environmental Protection Agency, Office of Research and Development, Washington, DC 246. [3] Nuclear Power's Missing Fuel (26),Business Week magazine, Retrieved

15 165 [4] Bruce Power New build Project Environmental Assessment - Round One Open House (26). (Appendix B2). BrucePower,Retrieved on [5] NYTIMES(27) s/1energy.html NY Times, July 1, 27, "Costs Surge for Building Power Plants. [6] Bansal N.K. (1998) CO 2 Reduction Potentials Through Renewable Energy Sources and Rational Use of Energy in India, Edited by Hake J. F., Bansal N., Kleemann M.,Ashgate Publishers, England. pp [7] Chikkatur, Ananth P. and Ambuj D. Sagar (27) Cleaner Power in India: Towards a Clean-Coal-Technology Roadmap. Discussion Paper 27-6, Cambridge, Mass.: Belfer Center for Science and International Affairs, December 27. [8] BARC (27) Clean Coal Technologies, Energy Systems Engineering, Presentation at BARC, 4 th December 27. [9] Bansal N.K., Hake J.(2) Energy needs and Supply Options for Developing Countries,Proceedings of the World Engineer s Convocation (Energy section), Hannover; pp [1] Sinha, R.K. (25) India s Energy Security The Role of Nuclear Energy Guest Lecture at Petroleum Federation of India, New Delhi, May 27, 25