Feature Selection and Parameter Optimization of Optimized Extreme Learning Machine for Motor Fault Detection

In view of the disadvantage of existing motor fault detection algorithms that they cannot adapt to complex and nonlinear fault categories and the accuracy is relatively low, we proposed one method for feature selection and parameter optimization based on simulated annealing-based whale optimization algorithm (SAWOA) optimized extreme learning machine (ELM) are proposed. Based on the ELM classification model, SAWOA is used to optimize the selection of network input variables and parameters of hidden layer nodes of ELM. To verify the effectiveness of the proposed method, it is applied to the real motor fault detection example, and compared with the existing methods under the same conditions. The experiment results show that the method achieves good performance in aspect of the classification accuracy and reliability, and show its effectiveness and application potential.