Optimization Design of Electromagnetic Structure of Traction Transformer Based on NSGA-II Optimization Algorithm

As the core component of the electrified railway traction power supply system, the high work efficiency, low loss, and high reliability of three-phase $\mathrm{V} / \mathrm{v}$ traction transformers are important goals for optimization design. However, in the process of optimizing the electromagnetic structure of traction transformers, there are conflicts between multiple optimization objectives that are difficult to balance, thereby increasing the complexity of optimization design. Therefore, this article first analyzes the main structural parameters that affect transformer efficiency, loss, and power density, and proposes a calculation model for transformer core loss, winding loss, and leakage reactance. Secondly, in response to the high coupling problem in multi-objective optimization, a traction transformer optimization design process based on non-dominated sorting genetic algorithm (NSGA-II) was established, and a multi-objective collaborative optimization design scheme for traction transformers was formulated. Finally, based on the optimization results of NSGA-II, a finite element simulation model was established using the ANSYS simulation platform, and the effectiveness of theoretical analysis and optimization results was verified using analytical methods.