Stochastic Energy Scheduling for Urban Railway Smart Grids Considering Distributed EVs Charging and PV Output Uncertainty

In order to improve the utilization rate of regenerative braking energy (RBE) and reduce the operation cost of railway system, this paper proposed an urban railway smart grids (URSG) with efficient and flexible energy management strategy is proposed, which integrates a traction power supply system (TPSS), electric vehicle charging station (EVCS), PV and battery storage. In this study, a bi-level stochastic optimization (BLSO) model is employed to determine the sizing of battery storage and the energy scheduling of URSG. The upper level aims to determine the optimal size of battery storage and minimize the comprehensive cost, considering the degradation of the battery capacity. In the lower level, based on piecewise linearization method, the power transmission loss of catenary is formulated as a linear mathematical model, and a stochastic mixed-integer linear programming model of URSG is established to schedule the power flow to minimize the electrical cost. Furthermore, regarding to the stochastic nature of EV charging behavior and PV output, the constraints with uncertain parameters are posed as chance constraints, for which Sample Average Approximation (SAA) method is introduced to transform chance constraints into deterministic constraints. Based on the actual urban rail transit line and traction load data, the simulation results show that the above system can reduce the cost by 18.16 %.