Vehicle-Operation-Driven Capacity Optimization and Dynamic Control of Urban Rail Transit Energy Storage Systems

This study proposes a novel two-stage optimization architecture (TSOA) for ultracapacitor energy storage systems (UCESS) in urban rail transit systems (URTS), aiming to enhance both economic returns and operational performance simultaneously. In the design phase, a swarm intelligence-based Cheetah Algorithm (CA) is employed to optimize UCESS capacity allocation alongside vehicle dwell time scheduling. During the operational phase, the system leverages these pre-optimized parameters to plan the state-of-energy (SOE) trajectory, ensuring precise real-time coordination between UCESS and vehicle demand. Simulation results indicate substantial improvements: the levelized cost of electricity (LCOE) decreases from 0.56 to 0.42 yuan/kWh compared with standalone UCESS optimization. Additionally, the TSOA elevates average energy savings of 6.12% and 4.07%, comparing the traditional fixed-threshold strategy (TFT-EMS) and the genetic algorithm-optimized fixed-threshold strategy (GAFT-EMS), respectively. Validation through a 1.5 MW engineering prototype implemented in a Chinese subway system confirmed these benefits, demonstrating practical average energy savings of 5.46% and 3.54% over TFT-EMS and GAFT-EMS. The proposed architecture exhibits broad applicability, independent of city, line, or passenger flow characteristics. It provides valuable guidance for promoting green transformation and low-carbon development in URTS through large-scale deployment of UCESS.