Game-theoretic risk-averse day-ahead optimal bidding strategy of virtual power plant aggregated with heterogeneous distributed resources

The rapid development of distributed energy resources imposes challenges to their scheduling and management. The virtual power plant provides solutions as an intermediate entity between the power grid and local distributed resources aggregators. This paper exploited a game-theoretic risk-averse day-ahead optimal bidding solution for a virtual power plant aggregated with heterogeneous distributed resources to address the technical challenge of optimal market participation and internal energy interactions. A day-ahead bidding strategy is proposed towards risk aversion in the wholesale electricity market, where the conditional value-at-risk is incorporated as a risk criterion to quantify and avoid the risk in the market with classified real-time electricity price scenarios. A local market is developed for electricity trading within the virtual power plant in a Stackelberg game manner. Considering privacy protection, the distributed iterative algorithm is adopted to solve the game model with limited information exchange, i.e., the electricity trading price and curves, where the bisection method is incorporated to guarantee the equilibrium solution. The carbon emissions and corresponding costs are considered in the decision-making processes of each entity in the virtual power plant. For the long-term cooperation among entities within the virtual power plant aggregation, a carbon emission intervals optimization method and award mechanism for internal electricity trading are established, considering the distributed resource aggregators’ satisfaction. The case studies and comparative experiments have evaluated the risk-averse performance of the proposed bidding strategy, with an 80.7 % reduction in the worst-case real-time operational costs and a 3.5 times increase in the worst-case profits. In addition, the proposed solution can provide reasonable pricing and ensure the benefits of multiple distributed resource aggregators, with the operational cost and carbon emissions reduced by 8.9 % and 9 % on average, respectively.