A dynamic nonlinear fault tolerant control algorithm and its application for motor reliability

A dynamic fault-tolerant control system based on the improved CMAC (cerebellar model articulation controllers) neural network is presented in this paper. In the conventional CMAC learning scheme, the correcting amounts of errors are equally distributed into all addressed hypercube, regardless the credibility of those hypercube. The proposed improved learning approach is to use the learned times of addressed hypercube as the credibility, the correcting amounts of errors are proportional to the inversion of the learned times of addressed hypercube, with this idea, the learning speed can indeed be improved. Based on the improved CMAC fault learning approach for motor, the effective control law reconfiguration strategy is presented. The system stability and performance are analyzed under failure scenarios. The numerical simulation demonstrates the effectiveness of the Improved CMAC algorithm and the proposed fault-tolerant controller.