Case Study of the Indian Power Sector

Present energy scenario

As the world's fourth-largest economy that is still growing significantly, the electricity demand in India has been constantly increasing. At present, India's power capacity is nearly 30% less than its demand (Eugelmeier, 2015). India is largely a fossil fuel-based economy with coal and petroleum being the key energy sources. Indian Coal has a higher than average ash content and. combined with ineffective combustion, this results in increased air pollution, a higher concentration of particulate matter in the atmosphere, and global warming (Mishra. 2014). The dependence on fossil fuels is high and on the rise. Between 1990 and 2009. the share of electricity generation from coal increased by 10%, making fossil fuels responsible for 85% of electricity generation. Of this, coal alone is responsible for 71% generation in 2016. Among CO,-free green energy, wind energy is significant in India, accounting for 14% of the total installed capacity (MNRE. 2016).

The initial state of the Indian power sector is as described in Table 1 in terms of installed capacity, capacity factor and operating emissions. In the Indian energy mix, the renewable includes Wind (25.18 GW), Solar (5.25 GW), Small Hydro (4.19 GW) and Biomass (2.68 GW) (MNRE, 2016).

Table 1. Details of existing power plants m India.

Source

Limit (GW)

Operating-capacity factor

Tonnes of CO;/MWhd

Coal

175.24*

©

CO

1.08s

Gas

24.51*

0.8*

0.45d

Diesel

1.00*

0,9

0.65d

Nuclear

5.78*

0.82f

0

Wind

25.1Sb

0.14*

0

Solar

5.25»

0.18*

0

Hydro

42.66b

0.5*

0

SHP

4,19b

0.5*

0

Biomass

2.68 ”

0.7*

0.11

Compiled from data available on Ministry of Power, 2016 (a); MNRE, 2016 (b); MNRE, 200S (c); Tan et al., 2009 (d); NTPC, 2015 (e); NPCEL, 2014 (f); CEA, 2008 (g).

Future of Indian power sector

India’s electricity consumption has increased by 64% duiing the ten-year period between 1997 and 2007. This trend is likely to continue, and it is safe to assume that India's need for electrical energy will continue to rise at a rate of around 8 to 10%. This would make India one of the largest electricity consumers in the world (ICLEI South Asia, 2007). India's energy demand for 2030 from electricity is predicted at 2940 TWh (Kliosla and Dubash, 2015). The need for increased electricity generation has to be balanced with the need to reduce greenhouse emissions. In this study. India's emission target is set to

1420 Mt of CO,. This hypothetical target is a 25% reduction in emission factor from 2012 levels. India is committed to lowering its CO, emission per unit GDP by 33-35% of 2005 levels by 2030 (UNFCCC, 2015). As a signatory of the Paris Agr eement, India is committed to lowering emissions by concentrating on renewable energy resources. Projections based on the existing National Energy Plan predict that India will reach its target of 40% electricity generation from non-fossil fuel sources by 2030. However, there is substantial uncertainty with regards to the translation of these policies into reality as the coal consumption continues to rise at 4.8% as of 2017 (Climate Action Tracker, 2019). Additionally, sector-specific targets, such as those for electricity generation, are not available. The actual CO, emission targets will be set by the Indian Government, considering various economic and societal parameters. A recent committee on energy highlighted this shortage, pointing out that the current allocation to Natural gas power plants is 70% short of allocation (Ministry of Power, 2019) and no new gas-based power plants are recommended for the next three decades (Pandey. 2018). Similarly, diesel-based power generation in India decreased 37% in 2018 due to fuel shortage (Rise in diesel, 2018). Under these conditions, this illustrative example assumes that diesel and natural gas will not be used for electricity generation in the future, given the present fuel crisis. The present case study is used primarily to illustrate the applicability of the proposed methodology; though realistic data has been considered.

Table 2 gives details related to resources available for future power plants. The lifecycle emissions are considered here for all renewable sources. For coal and biomass-based power generation, lifecycle emission and operating emission are treated as the same as the major contributing component is the fuel consumed. Also, all new coal and nuclear power plants are assumed to operate at a relatively high capacity factor of 0.9.

Table 2 lists the capital costs, in terms of its installed capacity, of various power plants. The option of carbon capture is also included with new coal power plants. Importance of CCS technology for the future power system cannot be neglected (Morgan, 2019). With regards to the cost of CCS technology, varying estimates are available in literature as no commercial plant is operational. In general, it is estimated that CCS enables power plants will cost 40-80% more than their traditional counterparts, though these percentages are expected to lower by 10-30% over the next decade (Heuberger et al., 2016). For the purpose of this example, it is estimated that CCS enabled power plants will have a 30% higher capital cost than conventional power plants, as suggested by World Energy Council report (Breudow, 2007).

It should be noted that the effective capital cost, per unit energy delivered (and not installed capacity), is considered in the analysis. This is because of varying capacity factors of renewable power plants. 1 MW coal power plant is expected to deliver 0.9 MWy of energy annually. On the other hand, a Solar PY plant of the same capacity is expected to deliver 0.2 MWy of energy every year. Therefore, in terms of energy delivery, the effective capital costs of renewable plants are significantly higher. The effective

Table 2. Future power sources for the Indian power sector.

Resource

Limit

(GW)

Operating capacity factor

Tonnes of CO/MWh'

Water

(inJ/MWh)<

Land use

(m2/MWh-y)i

Capital cost (io's/mw)

Coal

NA

0.9'

1.08

2.56

0.433

0.67?

Nuclear

9.5Sb

0.82b

0.02

2.80

0.029

0.79?

Hydro

148.70’

0.50’

0.12

5.30

3.70

1.17’

Wind

47'

0.14'

0.07

0015

0.006

1.131

Biomass

19.50'

0.70'

0.11

1.80

0.126

0.50b

Small hydro

15'

0.50'

0.12

5.30

0.003

2.09‘

Solar PY

NA'

0.20'

0.15

0.004

0.164

1.19*

Solar thermal

NA

0.20

0.185

1.08

0.366

1.92“

Coal with CCS

NA

0.9'

0.10

5.03

0.626*

0.87?

Compiled from data available on NHPC, 2012 (a); NPCIL, 2012 (b); MNRE, 2008 (c); NTPC, 2012 (d); Tan et al., 2009 (e); Wilson et al., 2012 (f); CEA, 2004 (g); Banerjee et al., 2006 (li); Nouni et al., 2008 (i); Fthenakis and Kim, 2009 (j); Carter, 2007 (k); CERC, 2015 (1); IRENA, 2012 (m) (IS = 60 Rs ), capital cost of a power plant can be calculated by dividing the installed capital cost with corresponding capacity factor. Based on the capacity factors and installed capital costs, as provided in Table 2, the effective capital costs of these plants are calculated and used for further analysis.

It should be noted that the case study is used purely to illustrate the methodology, though realistic data has been used. The choice of one power plant over another is a complex one, involving socioeconomic factors, environmental considerations, and policy decisions in addition to constraints such as land and water availability, capital investment requirements, etc. This example considers three-objective functions related to land use, water requirement and capital investment. While the example is limited to three- objective functions, the methodology can support any number of objective functions, provided they can be quantified. Additionally, since the method can be employed to generate a Pareto optimal front of all possible solutions, policy makers can employ the same to identify discrete energy mixes based on complex requirements.

Two-objective optimization

As an example, the two-objective functions considered for the Indian power sector are the minimisation of capital investment (Ф) and minimisation of water footprint associated with new power plants ( KF). This is considered purely as a hypothetical example as the deciding on one type of power plant over another involves multiple environmental, economic and societal parameters. The presently available power plants are already listed in Table 1 and the available energy sources that can be converted to new power plants are listed in Table 2. Pinch point of the existing system of power plants is identified as 1.08 tCO,/ MWli. Weighting factor corresponding to the capital cost objective function is introduced. When a = 1, the problem is purely cost minimisation and when it is zero, the objective is simply to minimise water footprint. The relevant portion of the MOPC vs. weighting factor plot is shown in Figure 6.

The intersection points of the MOPC lines give the weighting factors at which the prioritizing sequence of the system may change. For low values of weighting factor (a < 0.24), it is seen that solar energy is the resource with the lowest prioritised cost. Solar energy, which has the lowest emission factor, also has the lowest prioritised cost in this region. As solar energy is considered as an unlimited resource, in region O-A (0 < a < 0.24), the entire energy requirement is supplied by solar energy. Beyond point A, Biomass becomes more cost effective than solar energy, and it too enters the solution mix and the prioritising sequence changes to [Biomass-Solar], In region A-B (0.24 < a < 0.32), biomass and solar energy supply the demand.

Prioritised cost of available types of power plants

Figure 5. Prioritised cost of available types of power plants.

Regions with discrete solutions

Figure 6. Regions with discrete solutions.

In this case, the maximum available capacity of biomass is limited when compared to the demand. Therefore, biomass is used to its maximum capacity, and solar energy supplies the rest of the demand. At point B, Nuclear energy becomes cost effective. Nuclear energy has a lower prioritised cost and emission factor than biomass. This changes the prioritizing sequence to [Nuclear-Biomass-Solar]. In region B-C (0.32 0.48), the solution mix consists of nuclear energy, biomass and solar PV. The major change in the composition of energy mix occurs at a weighting factor of C (a = 0.48), where CCS enabled coal power plants becomes more cost effective than solar PV. From this point onwards, the energy that was being supplied by solar PV is supplied by CCS enabled coal power plants, and the emission factor of CCS enabled coal is lower than that of biomass, meaning the prioritising sequence changes to [Nuclear-CCS enabled Coal-Biomass],

Energy generated from the existing coal power plants decreases as the value of ir increases, i.e., the amount of unused energy (or waste) varies from 15.4% to 9.6% from point A to C. This is because, as the stress on water minimisation increases, the share of resources with low water footprint, such as solar energy, increases. In this example, solar energy has a slightly higher emission factor than CCS enabled coal, leaving less room for emissions from existing power plants. Note that not all points of intersection alter the constitution of the resource mix. For example, at a weighting factor of around 0.27, the MOPC lines of wind and nuclear energy intersect. As wind energy is not a part of the mix, this intersection point is of little significance. The Pareto optimal front in Figure 7 shows the four possible solutions for this system for a net electricity generation of 2940 TWli.

In essence, if there is stress on cost reduction, as opposed to water footprint minimization, the solution mix is dominated by fossil fuels (specifically new coal power plants with CCS). In fact, biomass is the only renewable that is pail of the cost minimal energy mix. Nuclear energy also plays a significant part. However, for most weighting factors, the bulk of the energy requir ement is met by solar PV and at lower weighting factors when water footprint minimization is more critical, Solar PV alone satisfies the entire energy requirement along with existing coal power plants or coal power plants with CCS. Improving the cost-effectiveness and water footprint of Solar PV and CCS enabled coal power plants can significantly affect the cost and water consumption associated with power generation in India.

Three objective optimizations of the Indian power sector

The three objectives considered in this example are the minimisation of capital investment (Ф,) and minimisation of water footprint associated with new power plants (Ф.) and minimisation of laud footprint

Pareto optimal front for the Indian power sector

Figure 7. Pareto optimal front for the Indian power sector.

for the power plants (Ф,). Two weighting factors a and ft corresponding to Ф1 and Ф, are also introduced. The power plants available and new energy sources are presented in Tables 1 and 2 along with their characteristics. The pinch point of the system is at 1.08 tCO,/MWh. Knowing the pinch emission factor and the characteristics of the new energy sources, their MOPCs can be calculated for the entire range of weighting factors. It can be seen that solar thermal has a higher cost coefficient for all three objectives than solar PY (which is the energy source with least emission factor) and. therefore, will never be part of the optimal solution mix. Hie MOPCs of such energy sources can entirely be avoided, further simplifying the solution process. As explained earlier, it is possible to generate a graphical solution space for three- objective optimization problems. For this purpose, the lines along which the MOPC planes of any two energy sources are functions of a. and ft and they intersect when the prioritised cost of the two energy sources become equal. The lines along which they intersect are plotted as prioritised cost intersection lines or PCILs. Since the two planes intersect along a line, the prioritising sequence of the system is only likely to change along these lines. The PCILs for all new energy sources are provided in Figure 8. The sum of weighting factors cannot exceed one. Therefore, only the relevant and viable portion of the solution space needs to be considered as the sum of a and ft cannot be greater than one. The solution space is divided into a number of regions by these lines. In each region, it is possible to eliminate certain PCILS as then MOPC is significantly higher than that of others.

Through proper understanding of the interactions between the PCILS, the prioritizing sequence for any combination of weighting factors can be identified. In order to better illustrate the process, consider an enlarged region, provided in Figure 9. The PCIL of wind and solar PV starts at a weighting factor of a = 0 and ft = 0.94 and meets the x axis at a = 0.1 and /3=0. It means that wind energy is cheaper than solar energy for weighting factors nearer to the origin. However, for most of the solution space, solar PY is preferred to wind.

In the region surrounding a = 0 and /3 = 0.1 in Figure 9, had wind energy been an unlimited resource, it alone would supply the entire demand requirement. As that is not the case, the prioritising sequence in this region is [Wind-Solar PV], It can be seen that at weighting factors of a = 0 and ft = 0.054, the PCIL involving solar PV and nuclear energy enters the solution space. It divides the space into two regions, such that nuclear energy has a higher prioritised cost than solar PV in one region and a lower MOPC in the other. In this example, nuclear energy has a lower emission factor than wind energy. Therefore, depending on the PCIL between wind and nuclear energy, two prioritising sequences are possible here. If the prioritising cost of wind energy is lower than that of nuclear energy, then the prioritising sequence

Prioritized cost intersection lines for Indian power sector (Krishna Priya, G. S., and Bandyopadhyay, S. 2017b. Multi-objective pinch analysis for power system planning. Applied energy, 202

Figure 8. Prioritized cost intersection lines for Indian power sector (Krishna Priya, G. S., and Bandyopadhyay, S. 2017b. Multi-objective pinch analysis for power system planning. Applied energy, 202: 335-347).

is [Nuclear-Wind-Solar PV]. Conversely, if the prioritising cost of wind is higher than that of biomass, it will no longer be part of the solution mix, and the prioritising sequence is [Nuclear-Solar PV]. However, in this particular example, the demand to be supplied is much greater than the maximum capacity of nuclear energy. Therefore, it gets utilised to its maximum capacity and is no longer part of the available resources. The prioritising sequence then reverts back to include wind energy. In essence, in the second region, the prioritising sequence is [Nuclear-Wind-Solar PV]. This means that nuclear energy and wind energy are used to its maximum capacity first. The next low emission resource available is solar PV, which is capable of supplying the entire load. In essence, both wind and nuclear energy are used up and the rest of the demand is supplied by solar PV. The next major change occurs at weighting factors of a = 0 and /3 = 0.025, where biomass becomes cheaper than solar PV. Beyond the PCIL involving Biomass and solar PV, the solution consists of [Nuclear-Wind-Solar PV-Biomass], As biomass has a higher emission factor than nuclear, two prioritising sequences are possible here. The sequence can either be [Nuclear-Wind-Solar PV-Biomass] or [Nuclear-Wind-Solar PV] if biomass has a higher MOPC than nuclear energy. If nuclear energy or biomass had a maximum capacity' comparable to that of the demand, it would have been necessary to study their interaction. However, as nuclear energy has a very limited capacity, it is used to its maximum capacity in either case, making its interaction with biomass unimportant.

The unique prioritizing sequence is unnecessary, as biomass, wind and nuclear energy will be exhausted in tiying to supply the load. The solution in this region contains [Nuclear-Wind-Solar PV-Biomass]. Similarly, bey'ond a = 0 and p = 0.03, the PCIL involving SHP and solar PV enters the solution space, and the solution mix changes to [Nuclear-Wind-Solar PV-SHP-Biomass], It should be noted that in a small triangular region in the solution space with base p = 0.025 and p = 0.03, the solution contains SHP, but no biomass as biomass has a higher prioritised cost than solar PV in this region. At a = 0.096 and /3=0, the PCIL relating wind and Solar PV intersect the x-axis. Beyond this line, wind energy is no longer a part of the solution mix. The next major change is at a = 0.16 and /3=0, where the PCIL involving coal power plants with CCS and solar PV enter the solution space. Beyond this line, the energy being supplied by solar PV is replaced by coal power plants with CCS. Multiple low emissions energy sources are part of the prioritising sequence.

In this particular problem, the demand for electricity is significantly higher than the maximum capacity of most energy sources, such as wind, SHP and nuclear energy. The two energy sources of unlimited capacity are solar energy and coal power plants with CCS. Therefore, major changes in the composition of the optimal energy mix will be dependent on these two energy sources. At this stage, it

Identifying prioritising sequence using PCIL plot (Krishna Pnya, G.S. and Bandyopadliyay, S. 2017b. Multiobjective pinch analysis for power system planning. Applied Energy 202

Figure 9. Identifying prioritising sequence using PCIL plot (Krishna Pnya, G.S. and Bandyopadliyay, S. 2017b. Multiobjective pinch analysis for power system planning. Applied Energy 202: 335-347).

is necessary to study the MOPCs of wind, SHOP, nuclear energy and biomass against that of coal power plants with CCS. In this case, by studying the complete plot of PCILs, it can be seen that they all have a lower MOPC than coal power plants with CCS for the specified weighting factors. The solution beyond this line contains [Nuclear-Wind-SHP-Coal with CCS-Biomass], The PCILs along which the solution mix changes are highlighted in Figure 9.

Not all PCILs alter the solution mix. For example, the PCIL involving nuclear energy and wind is not as significant because both these energy sources have a small maximum capacity when compared to the demand, and will be used up entirely. For these resources, given their limited maximum capacity, they will be completely absorbed into the optimal energy mix, provided their MOPC is lower than that of solar PV. Using a similar analysis, it is possible to analyze the entire solution space and identify unique solution mixes. This is shown in Figure 10.

Complete solution space for Indian power sector (Krishna Priya, G.S and Bandyopadhyay, S. 2017b. Multi- objective pinch analysis for power system planning. Applied Energy 202

Figure 10. Complete solution space for Indian power sector (Krishna Priya, G.S and Bandyopadhyay, S. 2017b. Multi- objective pinch analysis for power system planning. Applied Energy 202: 335-347).

A total of thirteen unique solutions are possible for this particular combination of energy sources. All these solutions have been identified using the method developed. Some of the regions are already explained in Figure 9 and have simply been marked in Figure 10. It can be seen that as solar PV and CCS enabled coal are treated as unlimited energy sources, they supply a significantly large portion of the overall demand requirement.

 
Source
< Prev   CONTENTS   Source   Next >