In view of such issues as weak disturbance resistance tracked by traditional PI control current

inaccurate current tracking prediction and large delay due to local optimum predicted by neutral network when the cascade-based H-bridge active power filter achieves the harmonic suppression

a kind of improved GA-optimized cascaded H-bridge APF neural network predictive current control strategy is proposed in this paper to improve the GA-optimized neural network change factor of selection and crossover operation

which improves the global search ability and robust performance of current tracking control. Simulation results show that the proposed control strategy

compared with traditional PI control

improves the robustness of current tracking control. Compared with the neural network control method based on traditional GA optimization

the tracking speed of the method proposed in this paper is improved by about 0.04 s

the control error is reduced by about 0.2 A

and the harmonic distortion rate is reduced by 0.36%. It can be seen that the response speed and control accuracy are better

and the tracking control capability is better.