(3) In view of the fact that KELM is more sensitive to the settings of the nuclear parameter Z and the penalty parameter C, a method for optimizing the nuclear extreme learning machine based on the improved locust algorithm (IGOA-KELM) is proposed, and IGOA-KELM is used to train and optimize the optimal feature matrix. test. The effectiveness of the proposed method is verified by analyzing the vibration signals of the M-type under different working conditions.
正在翻译中..