A CVaR-based stochastic framework for storm-resilient grid, including bus charging stations

Abstract

This paper proposes a two-stage stochastic framework for improving distribution system resilience against storms, in which the uncertainties associated with load demands, solar irradiance after a storm event, and maximum gust wind speed are considered. In the first stage, the available idle electric buses (EBs) are optimally allocated to the charging stations prior to storm arrival. In the second stage, critical loads are restored through island formation after the storm. In this framework, the solar-powered charging stations of EBs and the distributed generators (DGs) are part of the electric energy resources. The solar charging stations contribute to supplying the critical loads in two ways: 1- The electricity generation through rooftop-installed photovoltaic modules 2- By discharging the batteries of the pre-allocated EBs. The objective of this optimization model is to minimize the expected priority-weighted curtailed energy. At the same time, the risk imposed by the uncertain parameters is taken into account through the conditional value-at-risk (CVaR) index. The proposed framework is expressed in terms of a stochastic mixed-integer linear programming (MILP) optimization problem. This framework is implemented on the 33-node distribution test system, and its superior resilient performance is demonstrated through four case studies.