Worst-case conditional value-at-risk based bidding strategy for wind-hydro hybrid systems under probability distribution uncertainties

• Published in: Applied energy Vol. 256; p. 113918
• Main Authors: Liu, Yangyang, Shen, Zhongqi, Tang, Xiaowei, Lian, Hongbo, Li, Jiarui, Gong, Jinxia
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 15.12.2019
• Subjects: Bidding strategy; Distributionally robust optimization; Hybrid renewable energy system; Mixture distribution; Risk management; Worst-case conditional value-at-risk; Risk management; Worst-case conditional value-at-risk; Bidding strategy; Distributionally robust optimization; Mixture distribution; Hybrid renewable energy system
• ISSN: 0306-2619; 1872-9118
• DOI: 10.1016/j.apenergy.2019.113918

•Bidding strategy proposed for the wind-hydro hybrid systems in electricity markets.•Partial information available for probability distributions of random variables.•Distributional uncertainties of multiple variables modeled with mixture distribution.•Worst-case conditional value-at-risk used as a risk measure and its optimization.•Generation companies should obtain more information about probabilities of predictions. It is challenging for renewable power (such as wind power) to participate in electricity markets, because of various uncertainties in terms of prices and power generation fluctuations. Further, the exact probability distributions of random variables are difficult to specify, leading to problems and errors with respect to the bidding strategy and risk management conducted by power generation companies. To overcome these issues, a risk averse bidding strategy is proposed to allow a wind-hydro hybrid system to participate in an electricity market when only partial information is available about the underlying probability distributions of random variables. A mixture distribution structure is employed to model multiple distributional uncertainties for the hybrid system, and the worst-case conditional value-at-risk is used to measure the hybrid system’s risk considering the distributional uncertainties. This bidding strategy provides a solution that allows power generation companies to manage their distributional uncertainties in electricity markets, especially for renewable power with low accuracy forecasts. This method can estimate the benefits of forecast accuracy improvement and predictions’ probability information on generation companies. Compared with the stochastic bidding strategy, the proposed bidding strategy obtains robuster results for distributions to achieve better risk management, as illustrated by the study case.