Distributionally Robust Chance Constrained Optimization Method for Risk-Based Routing and Scheduling of Shared Mobile Energy Storage System With Variable Renewable Energy

Abstract

This paper proposes a pricing and scheduling method for shared mobile energy storage systems (SMSs) in coupled power distribution and transportation networks. Different from existing shared energy storage studies, which mostly focus on stationary resources, the paper investigates the SMS operation considering the negotiation of rental prices as well as mobility and charging/discharging among SMS owners and different users. Specifically, the SMS pricing and scheduling with variable renewable energy are established as a bilevel mixed-integer chance-constrained distributionally robust optimization problem. In the upper-level problem, the SMS owner determines pricing and day-ahead mobility strategy to maximize its payoff. In the lower-level problem, the SMS users, i.e., distribution grid operators, determine the SMS charging/discharging power according to the SMS day-ahead pricing results and intra-day distribution grid operation strategies for accommodating variable renewable energy. The distributionally robust chance constraint is designed to cope with the intra-day operational risk caused by the variability of renewable power generation. To cope with the solution difficulty in the proposed bilevel optimization problem, the chance constraint is reformulated as second-order cone constraints, which are further transformed into a set of linear constraints, and then the reformulated bilevel mixed-integer linear programming problem is decomposed and iteratively solved to avoid enumerating lower-level integer variables. Simulation results show that the utilization rate of SMS batteries is increased and the excess renewable power is fully consumed when SMSs are shared among different distribution grids. The proposed distributionally robust optimization achieves higher revenue for the SMS owner and smaller operating costs of distribution grids than robust optimization under uncertain environments.