In response to the pressing practical requirements associated with new energy locomotives （NELs） integrated into photovoltaic energy storage （PVES） microgrids，a methodology was proposed for optimizing the configuration of PVES capacity within the locomotive depot microgrid，utilizing a flood algorithm. Firstly，by integrating the flexible load characteristics associated with NEL batteries and energy storage stations，a mathematical model was developed to optimize the configuration of PVES capacity. Secondly，four typical days in spring，summer，autumn and winter were selected for example analysis，utilizing the load shortage rate，light rejection rate and photovoltaic energy penetration rate as key evaluation metrics. Finally，the flood algorithm was employed to solve the model，followed by a comparative analysis of three NEL charging schemes alongside four optimization algorithms. The findings from the simulation indicate that the flood algorithm is capable of attaining the optimal solution in fewer iterations，thereby demonstrating its efficiency and superiority in addressing complex optimization challenges. The optimized configuration scheme effectively harnesses the flexibility of the flexible load of NEL battery charging system，achieving “load following source”. The average penetration rate of photovoltaic power generation attains 93.98%，thereby promoting the low carbonization and economic viability of railway transportation in mining regions.