Fault tolerant control for a class of nonlinear system based on active disturbance rejection control and rbf neural networks

In this paper, a fault tolerant control method based on active disturbance rejection control (ADRC) and radial basis function neural network (RBFNN) is proposed for a class of multi-input-multi-output nonlinear system with actuator faults, components faults and sensor faults. The proposed method does not rely on the plant model. By regarding the faults and plant uncertainties as the disturbance, through the observation of extended state observer and the compensation of feedback control signal, this method achieves the fault tolerance control of the plant with component fault and actuator fault. For sensor faults, in this work, radial basis function neural network is applied to estimate the real output of the system. Then this output estimation is utilized by active disturbance rejection control to achieve the fault tolerance of sensor. Finally, the effectiveness of the proposed method is validated by the simulation results of the three-tank system.