Distributionally Robust Dynamic Economic Dispatch with Energy Storage and Renewables

Zhongjie Guo,Wei,Jun Yu,Haiji Zhao,Shengwei Mei
DOI: https://doi.org/10.1109/icpet55165.2022.9918217
2022-01-01
Abstract:The integration of renewable generation which is uncertain and fluctuates over time challenges the economic dispatch of power systems. This paper proposes a distributionally robust dynamic programming framework to make economic dispatch decisions in a sequential manner. Compared to the stochastic DP that assumes the stage-wise independence of uncertainty and applies the sample average approximation, the proposed framework is improved from two main aspects: first, the temporal dependence of uncertain variable is exploited and the value functions in Bellman’s equation are taken conditional expectation; second, the inexactness of estimated conditional distribution is compensated by considering the worst distribution within an ambiguity set. A sampling-based algorithm with efficient samples is proposed to calculate the value functions. Case studies conducted on the modified IEEE 118-bus system verify the effectiveness of the proposed method.
What problem does this paper attempt to address?