JonasVaičys
Kruonis Hydro Pumped Storage (KHPS) is a power plant located in Lithuania with the installed capacity of 900 MW. A floating photovoltaic (FPV) system of total 200 MW capacity is going to be installed on the upper reservoir of KHPS. One of the main objectives is to adjust the KHPS generation/ consumption schedule to solar generation forecast in order to reach the maximum efficiency in terms of profit gained from participation in various electricity markets.
Optimization for an efficient use of energy
There are several markets to utilize generated electricity (DA, ID, mFRR, aFRR and FCR). Efficient participation of non-dispatchable renewable energy sources (RES), largely depending on metrological and oceanic condition [1], in some of these markets (for e.g. mFRR) is quite challenging. Nevertheless, a strategy optimizing the use of RES can be developed [2]. An optimization algorithm maximizing the revenue (similar to [3]) is the strategy to be used for KHPS and FPV.
One of the most important input variables of the objective function of the algorithm is the previously discussed FPV generation forecast. In addition to other input variables (DA and mFRR market prices forecasts, imbalance volume forecast, KHPS water level and day type), the optimization algorithm will also be provided with constraints specific for each power source. The optimization objective function is as follows:
Fig. 1 Flowchart of the optimization process
max[∑_(t=1)^T▒〖〖(β〗_t^DA 〖∙P〗_t^DA+β_t^ID 〗 〖∙P〗_t^ID)+∑_(t=1)^T▒〖〖(γ〗_t^mFRR"↑" 〖∙∆〗_t^mFRR"↑" )-∑_(t=1)^T▒〖〖(γ〗_t^mFRR"↓" 〖∙∆〗_t^mFRR"↓" )+∑_(t=1)^T▒〖〖(δ〗_t^FCR 〖∙∆〗_t^FCR)〗〗〗]
where:
〖β_t^DA,β〗_t^ID -DA, ID market price for particular hour;
〖P_t^DA, P〗_t^ID -DA, ID market volume for particular hour;
〖γ_t^mFRR"↑" ,γ〗_t^mFRR"↓" -mFRR market price for particular hour;
〖∆_t^mFRR"↑" ,∆〗_t^mFRR"↓" -mFRR market volume for particular hour;
δ_t^FCR- FCR market price;
∆_t^FCR- FCR market volume.
Conclusions
Artificial intelligence enables combination of different image data and other numerical information in order to increase the forecast horizon and reduce prediction errors in hybrid energy power systems. These forecasts can then be used as key input for the optimization algorithm of a hybrid energy system, so that solar and hydro energy would be used more efficiently and RES will gain participation flexibility in various electricity markets.
References
[1] L. Bird, M. Milligan and D. Lew, "Integrating Variable Renewable Energy: Challenges and Solutions," National Renewable Energy Laboratory, 2013.
[2] H. Algarvio, F. Lopes, A. Couto and A. Estanqueiro, "Participation of wind power producers in day‐ahead and balancing markets: An overview and a simulation‐based study," WIREs Energy and Enviroment, 2019.
[3] C. Olk, D. U. Sauer and M. Merten, "Bidding strategy for a battery storage in the German secondary balancing power market," Journal of Energy Storage, vol. 21, pp. 787-800, 2019.
[4] Jose L. Crespo-Vazqueza, C. Carrillo, E. Diaz-Dorado, Jose A. Martinez-Lorenzo and Md. Noor-E-Alam, "A machine learning based stochastic optimization framework for a wind and storage power plant participating in energy pool market" Applied Energy vol. 232, pp. 341–357, 2018.