Ship power system is an independent power system, its stability is greatly affected by the excitation control system of generator. The adjustment of control parameters depends on the experience accumulated by engineers, which has some limitations. The high cost of real ship test and test results. Therefore, this paper uses MATLAB/Simulink to model and simplify the ship power system, and then uses differential evolution algorithm to optimize the parameters. Finally, the typical operating conditions such as sudden load operation and sudden asynchronous motor operation under a single generator are simulated. The performance of the control system is evaluated using ITEA index. The simulation results show that the differential evolution algorithm has better control effect.