Two-stage Optimal Dispatch Strategy for PV and HESS Co-Phase Traction Power Supply System Considering Prediction Uncertainty

To further improve the utilization of regenerative braking energy (RBE), perform peak shaving and valley filling for traction load, address power quality issues, and reduce carbon emissions, integrating photovoltaic (PV) and hybrid energy storage system (HESS) into the co-phase traction power supply system (TPSS) has become a key development trend for future smart railways. However, the uncertainties of traction load can impact the safe and economic dispatch of TPSS. To enhance the operational efficiency of TPSS and compensate for short-term supply-demand imbalances, a traction load ultra-short-term power prediction method and a two-stage optimization dispatch method are proposed. The latter consists of an offline stage and an online stage. The offline stage, considering HESS degradation and performing global optimization, provides a reference for the next day’s dispatch to avoid myopic decision-making. The online stage is divided into a model predictive control (MPC) process and real-time decision. The MPC optimizes based on the predicted values and dispatch reference values, while the real-time decision adjusts the MPC outcomes based on real-time observational data and predefined rules to address the impact of prediction uncertainty. Finally, the accuracy and effectiveness of the proposed prediction method and dispatch strategy are validated using actual electrified railway data, while a StarSim hardware-in-the-loop platform is adopted to validate its practical deployment capability.