Methodology

Our in-house power systems simulation framework EnerPol 19, 06 is applied in this work. Within EnerPol, a detailed geo-referenced representation of the central European power systems is the basis of hourly, chronological AC optimal power flow simulations. The generation model comprises 5,000 individual power plants, distinguishing between conventional and renewable power plants. Detailed thermodynamic models are applied to incorporate conventional power plants’ baseload and cycling operation, as well as costs and constraints of ramping 07. The exact geographic locations of the power plants are used in the model, as shown in Figure 1. The hourly energy production of solar and wind power plants is derived from in-house meso-scale weather simulations. The transmission system model consists of 3,000 high voltage transmission lines and 2,000 substations, as shown in Figure 2. The hourly cross-border power flows to the non-simulated neighboring countries are derived from ENTSO-E statistics 04 and are distributed across the tielines proportionally to the tielines’ capacity ratings. The hourly end-user electricity demand is differentiated by the consumer groups of households, industry, commerce and transportation and is spatially allocated with respect to each consumer group’s density distribution. The resulting hourly electricity demand is aggregated to the nearest substation. More information about the applied simulation framework is given by the present authors in 07.

Figure 2.

High voltage transmission system model of the central European power system that is used in this work.

Figure 1.

Power plant model of the central European power system used in this work.

Improved partload and baseload efficiencies for selected power plants

To investigate the impacts of increased partload and baseload efficiencies on the operational and financial performance of conventional power plants, a set of 45 coal and natural gas combined cycle (NGCC) power plants, whose characteristics are summarized in Table 1 and whose locations are shown in Figure 3, are considered to be upgraded in 2030. These power plants cover a wide range of nameplate capacities so that the effects of increased penetration of renewables on representative conventional power plants can be investigated. To minimize the influence of one upgraded power plant on other upgraded power plants, the power plants have geographically dispersed locations, as shown in Figure 3. In total, the upgraded power plants represent 5% of the total installed generation capacity in central Europe.

Figure 3.

Locations of the coal and NGCC power plants that are upgraded in 2030.

Within the simulation framework, merit order dispatch is used for each hour to determine the most cost-effective power plant portfolio that satisfies the corresponding power demand. An increase in either partload or baseload efficiency reduces a plant’s marginal cost of power production; hence the power plant is dispatched more frequently as its marginal cost is lower on the merit order curve. While an increase in baseload efficiency results in the power plant being dispatched more in baseload, an increase in partload efficiency enables the power plant to gain market shares during periods of low prices, whereas without the increase of the partload efficiency, it would not have been profitable to dispatch the power plant. This simulation framework therefore represents an operational response to increased efficiency, which realistically simulates power plants’ operation in the real wholesale market.

For each of the upgraded 45 power plants, the efficiency curves are adapted to reflect the assumed increase in partload or baseload efficiency, as shown in Figure 4. To enable an assessment of the impacts of different extents of efficiency improvements, the baseload and partload efficiencies are increased by 1%, 2% and 3%. The efficiency curves for the baseload efficiency increase scenarios are adapted only in the 100% load point, to enable a clear distinction in simulation results between partload and baseload efficiency increases.

Figure 4.

Illustration of change in efficiency curves for representative coal power plant. Curves for baseload efficiency scenarios offset by 0.1% to increase legibility.

Power system of central Europe in 2030

The 2030 power system model is derived as follows from the 2013 model that is described above. The power demand is scaled proportionally with the projected developments in population for all simulated countries 22. Projections for increased penetrations of renewables are obtained for the year 2020 from the national renewable energy action plans of the simulated countries 01, 02, 14, 15, 16, 17, 03. These 2020 projections are extrapolated to 2030 to obtain the generation model of the 2030 baseline scenario. A summary of actual and derived renewable capacities and predicted population developments is given in Table 2.

A complete phase-out of nuclear power by 2030 is considered for Germany, which is in accordance with political targets 10. For Switzerland, a phase-out of all nuclear power plants except for NPP Leibstadt is used. The installed capacities of coal, lignite and gas power plants for the 2030 reference scenario are not varied from the 2013 capacities. A coal price increase of 26% and a gas price increase of 14% compared to 2013 are anticipated for 2030 12. As more stringent CO₂ restrictions are anticipated for 2030 08, the price of CO₂ certificates is considered to increase from 6 €/tCO₂ (in 2013) to 20 €/tCO₂ (in 2030).

The total annual costs of cycling are derived from unit-specific cycling costs, which are based on actual US power plants 13 and are summarized in Table 3. The cycling cost factors of Table 3 are applied to derive the annual cost of cycling for each upgraded coal and NGCC power plant.

Comparing results for coal and NGCC power plants

Table 5 provides a summary of annual revenues, averaged over all upgraded power plants, distinguished by coal and NGCC power plants. In this work the revenues are presented per MW of nameplate capacity, termed “capacity-specific revenue,” in order to provide a consistent comparison of the upgraded power plants.

The annual revenues of the upgraded power plants in the 2030 reference scenario are 0.27 m€/MW and 0.32 m€/MW for coal and gas, respectively. This is at the upper end of today’s price ranges, which historic data shows to be in the order of 0.27 million €/MW for coal 11, and 0.24 to 0.34 million €/MW 20 for NGCC, since price increases are assumed for coal, gas and CO₂ emissions certificates, which increase the electricity price across the European power system. It can be seen, that the selected NGCC power plants achieve on average higher capacity-specific revenues than the selected coal power plants. This is because the coal price is assumed to increase more (+26%) than the gas price (+14%) until 2030, and because the increase in CO₂ prices shifts market shares from coal towards NGCC plants, as the capacity factors in Table 4 also indicate. Furthermore, as can be seen from Table 5, there is a fundamental difference in operational requirements between coal and NGCC power plants in 2030: While the upgraded coal power plants profit significantly from increased baseload efficiencies (0.27 to 0.41 m€/MW), there is only a small increase in revenues for coal plants from increased partload efficiency (0.27 to 0.29 m€/MW). On the contrary, NGCC plants profit slightly more from increased partload efficiencies (0.32 to 0.40 m€/MW) than from increased baseload efficiencies (0.32 to 0.38 m€/MW).

This effect is can be seen in more detail in Figure 6 and Figure 7, which shows the capacity-specific revenue for all efficiency scenarios for the investigated coal (Figure 6) and NGCC (Figure 7) power plants as a function of the capacity factor in the reference scenario.

Figure 7.

Annual capacity-specific revenue (m€/MW) as function of reference scenario capacity factor for upgraded NGCC power plants in 2030, for increased baseload (Figure 7a, top) and partload (Figure 7b, bottom) efficiency.

Figure 6.

Annual capacity-specific revenue (m€/MW) as function of reference scenario capacity factor for upgraded coal power plants in 2030, for increased baseload (Figure 6a, top) and partload (Figure 6b, bottom) efficiency.

As Figure 6 shows, there is an increase in revenues for coal power plants with increasing baseload efficiency. This is mainly because coal power plants, which had only achieved intermittent load capacity factors in the range of 0.2 to 0.4 in the 2030 reference scenario, now have more opportunity to run in baseload, which increases their revenue. The few coal power plants, which have not achieved any significant market penetration in the reference scenario (capacity factor below 0.15), do not profit from increasing baseload efficiencies, as they are not cost effective enough to penetrate the market at all, independent of any efficiency increase. Figure 6 shows that there is consistently little impact of increased partload efficiency on coal power plant revenue. This is because the partload efficiency increase slightly increases the profitability during partload operation, yet does not enable the plants that run in intermittent operation to move towards more baseload operation.

Figure 7 and Figure 7 show an opposite effect for NGCC plants compared to coal plants (Figure 6). There is only a moderate increase of revenues for NGCC plants with increasing baseload efficiencies (Figure 7); the main increase is seen for NGCC plants that ran in a capacity factor range of 0.2 to 0.4 in the reference scenario. However with increasing partload efficiency (Figure 7), it can be seen that there is a significant increase in revenue for several NGCC plants, while many other plants are not affected at all. Since this discrepancy cannot be explained only in terms of the operating regime of the NGCC plants, it is interesting to assess the role of the power plants’ geographic location in this trend.

Impact of instalment location of NGCC power plants on revenue increase

The results of Figure 7 are disaggregated to the country level in Table 6. It can be seen that NGCC power plants in Germany benefit most from increasing baseload efficiency (+29% revenue increase), while the highest increase in revenue from a partload efficiency increase is seen for NGCC power plants in Italy (+37%). This effect is explained in more detail in Figure 8 and Figure 9. Figure 8 shows the average weekly fluctuation of power production for the upgraded NGCC power plants in Germany and Italy in the reference scenario for 2030. Figure 9 shows the relative fluctuation of combined wind and solar power production for an average week in Germany and Italy. It is evident that the NGCC power plants in Italy have to adjust to market fluctuations more frequently and more heavily than the NGCC power plants in Germany (Figure 8). This is because there is a higher relative fluctuation of renewable power production in Italy compared to Germany (Figure 9). The direct influence of renewables on the operation of NGCC can be seen most clearly during noon hours, where NGCC in Italy have to reduce their power production to accommodate the peaks in solar power production. The relative fluctuation of renewables is lower in Germany than in Italy because in Germany, the share of renewables is split evenly between wind and solar (see Table 2) whereas in Italy, there is three times more solar power than wind power. Since the power production of wind power plants varies less than for solar power plants, the overall renewable power production is less variable in Germany than in Italy, as Figure 9 shows. The standard deviations of relative power production are 0.42 for Germany and 0.82 for Italy. Consequently, the standard deviation of relative weekly power production by NGCC power plants is 0.14 in Germany, and 0.19 in Italy. Therefore, the increased variability of renewable power production requires NGCC in Italy to cycle more, which leads to a higher benefit of increased partload efficiencies compared to NGCC in Germany, as seen in Table 6.

Figure 9.

Relative weekly power production of renewable capacities in Germany and Italy, for 2030 reference scenario.

Figure 8.

Relative weekly power production of NGCC power plants in Germany and Italy, for 2030 reference scenario.

In Figure 10, the annual average locational marginal electricity prices (LMPs), in blue coloured contours, and the relative increase in revenue for the 3% partload efficiency increase case of all upgraded NGCC power plants, red coloured symbols, are shown. The plotted average LMP values are calculated as averages of the hourly LMPs at each substation throughout the year; values for locations between two substations are interpolated accordingly. It is clear that the NGCC power plants most positively affected by the partload efficiency increase are located in northern Italy, which coincides with the locations with the highest electricity prices. This is because in the high-price region of northern Italy, NGCC power plants have to cover most of the load, which includes frequent ramps to adjust to the fluctuations in demand and solar power production. Since the NGCC power plants in northern Italy encounter more ramps and more partload operation than NGCC power plants in other regions of Europe, NGCC power plants in Italy benefit most from a partload efficiency increase.

Figure 10.

Annual average LMP [€/MWh] and increase in revenue [%] for all upgraded NGCC power plants in 3% partload efficiency increase scenario.

Financial comparison of revenue effects and cycling effects

As seen in Table 4, the conventional power plants will not only encounter new revenue opportunities in 2030, but also significantly increased numbers of starts and ramps to accommodate the increased penetration of non-dispatchable renewables in the central European power system. The relative change in annual cycling cost compared to the 2030 reference scenario is shown in Figure 11 together with the relative change in revenue for each efficiency increase scenario. It is evident that for the most beneficial efficiency improvements (baseload for coal power plants and partload for NGCC power plants), the revenue increase outweighs the increase in cycling cost, yielding an overall positive balance. Only for the case of an increase in baseload efficiency of NGCC power plants is the increase in cycling cost higher than the increase in revenue; this is due to a significant increase in load-following ramps of the corresponding NGCC plants. This underlines again, that for NGCC power plants an increase in partload efficiencies is more beneficial than an increase in baseload efficiency.

Figure 11.

Relative increase in annual revenue and cycling costs for coal (top) and NGCC (bottom) power plants for all baseload and partload efficiency increase scenarios.