Fault diagnosis based on EEMD-IGSA-IPNN for motor bearing

To improve the accuracy of fault diagnosis for motor bearing, an ensemble approach for fault diagnosis based on Ensemble Empirical Mode Decomposition (EEMD), Improved Gravitational Search Algorithm (GSA) and Incremental Probabilistic Neural Network (IPNN), called EEMD-IGSA-IPNN, is presented. Firstly, EEMD is used to extract the fault feature from the data of the motor bearing, then IGSA optimizes the thresholds of IPNN to classify the type of fault signal, and finally the goal of fault diagnosis is achieved. The experiment shows that the proposed method can improve the accuracy for motor bearing fault diagnosis.