A framework for joint management of regional water-energy ... - Orbit

More documents

Recommendations

Info

Equivalent reservoir level [%] a Very dry 80 60 70 60 60 65 40 150 65 70 1300 900 150 70 20 80 60 60 40 20 65 80 50 60 40 20 80 40 60 50 40 20 80 40 60 50 40 20 70 150 65 Dry Average 60 65 900 70 70 60 60 70 150 70 65 900 0 60 70 65 40 50 60 65 Wet 150 Very wet 70 70 J F M A M J J A S O N D 70 65 65 60 60 50 40 50 40 50 b 50 50 50 58 40 40 40 40 70 145 70 145 50 Very dry 58 Dry 58 1300 900 145 70 900 50 50 58 58 0 40 50 58 50 58 Average Wet 70 145 900 Very wet 70145 J F M A M J J A S O N D 70 58 70 58 58 50 50 40 50 40 c 51.6 51.6 60 51.6 40 150 60 Very dry Dry 1300 900 60 150 150 900 60 51.6 51.6 51.6 60 900 150 60 40 0 51.6 51.6 40 Average Wet 60 150 Very wet 40 51.6 40 40 51.6 J F M A M J J A S O N D 60 Figure 16. Water values [€/MWh] from optimization a) considering the power market, b) assuming a monthly-varying hydropower price, and c) assuming a constant price. occurred during the irrigation season, when the total benefits (constant hydropower plus irrigation) were highest (Figure 17c). Representing hydropower benefits through a power market provided more realistic results. The reservoir was not completely emptied during the autumn —as it occurs in the other two approaches— because of high water values at low reservoir levels (Figure 18c). Furthermore, reservoir operation changed depending on inflow conditions, resulting in low storage during dry years and high storage during wet years. These results indicate that using constant power prices (e.g. Cai et al., 2003) does not reflect the inflow and power demand seasonality, which affects the availability (and therefore the value) of hydropower. Monthly-varying power prices (e.g. Tilmant and Kelman, 2007) capture seasonal effects, but can cause unrealistic operation rules that can only be avoided through additional constraints. Using a simple power market to represent hydropower benefits provides reservoir operation policies that are more realistic, and that adapt better to changing inflow conditions. 38
15000 a Power market 20000 a Power market 10000 15000 5000 10000 Monthly energy releases [GWh/month] 0 15000 b 10000 5000 0 15000 c Monthly price Constant price Energy storage [GWh] 5000 20000 b 15000 10000 5000 20000 c Monthly price Constant price 10000 15000 5000 10000 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 17. Monthly energy releases [GWh/month] from simulations using water values a) considering a power market, b) assuming a monthly-varying hydropower price, and c) assuming a constant price. Thin color lines represent individual years; black lines represent the average year. Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 18. Energy storages [GWh] from simulations using water values a) considering a power market, b) assuming a monthlyvarying hydropower price, and c) assuming a constant price. Thin color lines represent individual years; black lines represent the average year. 5.3 Optimization of regional water-energy systems Combining SDP with the water value method proved to be a useful approach to assess some of the linkages between water and energy systems. However, because of the computational limitations faced by SDP, those linkages could only be studied at an aggregated level. To overcome this issue, SDDP was used to optimize a coupled model of the power system and the seven major river basins of the IP. An aggregated model (at the peninsula level) was used to evaluate the impact of aggregation. Figure 19 shows the convergence of the SDDP algorithm for the aggregated and the disaggregated models. The aggregated model, which has one reservoir, converged faster than the disaggregated one, which has seven. 5000 39
Page 1:
A framework for joint management of
Page 4 and 5: Silvio Javier Pereira-Cardenal A fr
Page 7 and 8: Acknowledgements I will always be g
Page 9 and 10: with the highest price, which can o
Page 11 and 12: og dermed værdien af vandkraft på
Page 13: Table of contents 1 Introduction ..
Page 16 and 17: xii
Page 18 and 19: Recognizing the problem’s complex
Page 20 and 21: whereas medium- and long-term tradi
Page 22 and 23: allocations contribute to the compe
Page 25 and 26: 3 Case study: The Iberian Peninsula
Page 27 and 28: Production of biofuels, which is en
Page 29 and 30: Figure 2. Mean annual irrigation re
Page 31 and 32: Price [EUR/MWh] 80 60 40 20 Price P
Page 33 and 34: Table 4. Characteristics of generat
Page 35: Figure 7. Temperature change factor
Page 38 and 39: The constraints are mathematical ex
Page 40 and 41: The weekly transition probability m
Page 42 and 43: values θ (€/MWh) for each inflow
Page 44 and 45: generate the backward openings (see
Page 46 and 47: The change in the objective functio
Page 48 and 49: where represents the standard dev
Page 51 and 52: 5 Results and Discussion The main f
Page 53: ment decision (and their constraint
Page 57 and 58: Relative storage 1 0.8 0.6 0.4 Tajo
Page 59: eservoirs, although some level of a
Page 62 and 63: The suggested method relies on math
Page 65 and 66: 8 References Al-Sunaidy A, Green R
Page 67 and 68: Hoffman AR (2010) The Water-Energy
Page 69 and 70: Schaefli B, Hingray B, Musy A (2007
Page 71: 9 Papers I. Pereira-Cardenal, S.J.,
show all

A framework for joint management of regional water-energy ... - Orbit

Create successful ePaper yourself

Delete template?

Save as template?