A multi-objective robust optimal dispatch and cost allocation model for microgrids-shared hybrid energy storage system considering flexible ramping capacity

In this paper, a microgrid groups with shared hybrid energy storage (MGs-SHESS) operation optimization and cost allocation strategy considering flexible ramping capacity (FRC) is proposed. Firstly, a joint system containing MGs with SHESS is constructed and its operation modes are analyzed. Secondly, Gaussian mixture model (GMM) and Latin Hypercubic Sampling (LHS) are used to obtain the confidence intervals of prediction errors for wind power, PV, and load. And then the FRC demand and supply of MGs-SHESS are quantified to construct the FRC sufficiency index. Furthermore, a multi-objective optimization model considering FRC sufficiency and operating costs is constructed, and the range of uncertain variables is described by confidence intervals, which transform the model into confidence gap decision theory (CGDT) model form. Further, the basic Shapley method is improved by considering the electricity interaction factor, carbon reduction factor, and FRC supply factor. Then use improved Shapley method to construct the MGs-SHESS two- layer cost allocation model. Finally, the case study results show that: (1) The total costs of SHESS are reduced by 5.89% and the FRC sufficiency is increased by 8.43% compared with decentralized energy storage system (DESS), which indicates that SHESS is able to achieve the co-growth of economy and flexibility. (2) Multi-objective dispatch plan considering FRC sufficiency is able to significantly improve FRC sufficiency by 12.56% at an additional cost of only 1.66% compared to conventional economic dispatch. It is worthwhile because it reduces the risk of power curtailment and load shedding during actual operation. (3) The costs of the CGDT model are reduced by 1.26% and the FRC sufficiency is increased by 1.20% compared with the IGDT method. Moreover, the total costs of the CGDT model are only increased by 1.84% compared with the day-ahead stage after substituting multiple random scenarios. It indicates that the CGDT is more robust in dealing with uncertainty. (4) The costs allocation method considering multi-factors improvement can satisfy the interests of multi-subjects. Even if the decision maker has a slight or moderate tendency to a certain factor, it can maintain the satisfaction of each subject to the allocation results, which improve the stability of the joint operation.