Research on Fatigue Testing Machine Control Algorithm Based on Neural Network PID

In order to improve the control accuracy of the fatigue testing machine, reduce the tracking error and improve the stability, an adaptive PID control system based on neural network is designed. The mathematical model of the fatigue testing machine and the control system is established in matlab/simulink, and the model simulation results show that: compared with the traditional PID control, the use of different control gains for different target responses can reduce the steady state error, overshooting and oscillation of the actual response; the genetic algorithm is used to optimize the control parameters p1, p2, and the parameters of different target responses and the corresponding optimal control gains are saved in the database; the target response parameters in the database are taken as the input of the network, the control gains are taken as the output, and the root-mean-square error (RMSE) is used as the evaluation index to train the neural networks with different structures. Among them, network 5 exhibits the smallest RMSE and can correctly predict the optimal control gains corresponding to different target responses.