A combined diagnosis method for electric drive system faults

The fault diagnosis of electric drive system is studied, and a combined diagnosis method is proposed; In order to detect the existing faults in the system, a modeling method based on nonlinear analytic redundancy is proposed. In this method, based on the nonlinear system theory, the residual construction fault detection and isolation method using nonlinear analytic redundancy is first obtained. Then, the obtained redundant residuals are applied to the electric drive system, and the corresponding residuals calculation formulas are generated according to the its state space equation, so as to realize the fault detection in the system; In order to locate the fault, a multi-layer perceptual artificial neural network recognition method is proposed. The principle of fault recognition based on neural network, the architecture design of neural network, the training and testing of neural network are discussed in detail; Finally, based on the simulation results of a real electric drive system composed of three-phase asynchronous motors, the feasibility of the proposed fault diagnosis scheme is verified.