In order to solve the problems of parameter uncertainty, large disturbance of unbalanced torque and time-varying internal parameters in artillery AC servo system, a control strategy combining neural network and fuzzy control is designed. The mathematical model of AC servo system is established, the neural network fuzzy controller is designed, and the particle swarm algorithm is used to optimize the parameters of the neural network. The controller model is established in Simulink, and the classical PID controller, the general neural network fuzzy controller and the neural network fuzzy controller optimized by particle swarm algorithm are compared. The simulation results show that the optimized controller has faster response speed, better robustness and stability.