• regenerative braking
• rail transit system
• optimization
• genetic algorithm
• energy saving
• simulation
• storage

This PhD thesis is composed of 3 parts generally related to the problem of electro-mechanical modelling and simulation techniques and energy saving measures of railway systems with energy storage systems.
For the first part, the electro-mechanical modelling and simulation techniques for smart energy systems were investigated. The simulators for DC electrified railway system and the AC electrified railway system were built by using two different simulation tools, the open language Modelica and one of the most influential commercial software Matlab-Simscape. The regenerative braking process of the rail transit systems and the energy utilization efficiency under different conditions are analyzed. The modelling and simulation techniques of the two tools were compared in terms of systematic modelling, flexibility, and expandability. The simulation results show that the capability of the Modelica and Matlab-Simscape to obtain accuracy results and the simulation speed are nearly equivalent. In terms of flexibility and expandability, Modelica has an open language platform for the reuse of custom components which makes it more flexible compared with Matlab-Simscape.
For the second part, the application of energy storage systems for enhancing energy efficiency of railway systems was discussed, especially the rechargeable electric energy storage systems (REESS) lithium battery and supercapacitor, which are used as stationary energy storage devices or on-board energy storage devices. The railway system in which all the regenerative braking energy is consumed by the on-board resistor is taken as the reference system to compare and analyze the energy-saving effect and energy efficiency of the railway system with energy storage. Simulation results show that the energy storage systems obviously contribute to the effective use of regenerative braking energy and enhance the energy efficiency of the railway systems. A real-life line was selected to analyze the cost-benefit of the energy storage system to verify that the energy storage system is recommended for use in practical projects.
In the final part, a methodology combining a genetic algorithm and a simulator is proposed to solve the multi-objective optimization design problem of the system. The system model is a DC power supply railway system which is constructed in Matlab-Simscape. The genetic algorithm adopts a multi-objective optimization algorithm, named nondominated sorting genetic algorithm II (NSGA-II). The proposed optimization methodology can get a better system layout design plan of traction power substations and energy storage systems, and valuably reduce the system operation and installation cost by optimize the position of the substations, and the number of energy storage systems and their positions.