A distributionally robust collaborative scheduling and benefit fallocation method for interconnected microgrids considering tail risk assessment

The uncertainty of load and renewable energy poses a huge challenge to the optimal economic dispatch of interconnected microgrids. In this paper, a distributionally robust optimization (DRO) collaborative scheduling and cooperative benefit allocation method is proposed. First, an improved ambiguity set is constructed to characterize the uncertainty of load and renewable energy to reduce unnecessary conservatism of the scheduling strategy. Then, the day-ahead collaborative scheduling problem of interconnected microgrids is constructed as a DRO model based on the conditional value at risk (CVaR) to accurately assess the tail average risks of strategies. Furthermore, due to the difficulty of solving the double-layer definite integral optimization model, this paper equivalently transforms the original model into an easily solvable single-layer mixed-integer second-order cone programming (MISOCP) model through dual transformation and reformulation of interval constraints. Subsequently, a benefit allocation strategy based on the improved Shapley value is proposed, which considers energy supply and demand fluctuations to encourage microgrids to participate in energy sharing. Finally, the case study demonstrates that the day-ahead risks and actual costs of the microgrid cluster are reduced by 20.19 % and 15.07 %, respectively, and the proposed method can achieve more fair benefit allocation under source and load uncertainty.