Self-tuning control of large-capacity friction brake based on RBF neural network

A large-capacity friction brake has been analyzed in this paper. The nonlinear model has been given and its working principle has been described quite clear. In order to realize the precise control of the large-capacity friction brake, a PID control method based on Kalman filter and the dynamic identification of radial basis function neural network is proposed. The extended Kalman filter algorithm can reconstruct the feedback signal of the whole frequency interference, which can be used for neural network identification and PID control input. The algorithm can handle hostile work environment, and its anti-interference is good. The training time of the radial basis function network algorithm is less than the back-propagation algorithm's. In short, using this algorithm, it is easy to realize online identification, what's more, it is easier to approach the local characteristics of the controlled object, and the identified model is more precise. Compared with traditional control methods, the control method proposed in this paper can deal with the non-linear and slow time-varying characteristics, and it has been verified through the prototype test, which has the guiding significance for the precise control of the large-capacity friction brake.