Stochastic Programming Versus Chance-Constrained Optimization for Optimal Rescheduling of Microgrids in Hierarchical Multi-Microgrid Systems

Abstract

Multi-microgrid systems (MMSs) can pave the way for the development of microgrids (MGs) in distribution networks (DNs), contributing to renewable energy exploitation and carbon footprint reduction. Notwithstanding, the emergence of MMSs complicates the day-ahead optimal energy management of DNs since both private MGs and distribution system operators (DSOs) are involved in the decision-making process compared to conventional DNs. Hence, hierarchical structures as practical solutions have attracted the attention of many researchers. Nevertheless, in those structures, MGs have to reschedule their generation and consumption patterns based on the orders received from DSOs. In this paper, two-stage stochastic and chance-constrained models for the rescheduling of an MG in an MMS are deployed and compared to embrace the uncertainties linked to demand and renewable energy generation. Results for the modified IEEE 33-bus test system showed that the proposed chance-constrained model can reduce the total cost by 6.68% compared to the stochastic one. Thus, the proposed model is well-suited for a fair rescheduling of the MG by avoiding excessive costs associated with extreme cases.