Study on a novel fault diagnosis method based on information fusion method

For the low accuracy and calculation speed of traditional fault diagnosis methods, the chaos optimization algorithm (COA), quantum particle swarm optimization (QPSO) algorithm and support vector machine (SVM) are introduced into the fault diagnosis to propose a novel fault diagnosis (CQPSMFD) method in this paper. In the proposed CQPSMFD method, the COA is used to initialize the parameters of the QPSO algorithm in order to obtain the CQPSO algorithm with the better convergence speed. Then the CQPSO algorithm is used to optimize the parameters of the SVM model to construct a high-precision SVM model (CQPSM) with the higher accuracy and stronger generalization ability. Next, the CQPSMFD method based on CQPSM method is proposed for motor. Finally, the experiment data from Case Western bearing dataset and actual motor are selected to verify the CQPSMFD method. The results show that the CQPSO algorithm can obtain the optimal parameter combination and the CQPSMFD method can effectively improve the fault diagnosis accuracy and speed.