Research on the working state prediction method of electromechanical equipment based on integrated learning

In order to predict the future working state of electromechanical equipment, this paper proposes a state prediction method based on the improved random forest regression algorithm (NGO-RFR), selects the data that intuitively reflects the working state as the fault diagnosis parameters, uses the Spearman correlation coefficient to filter out the parameters with higher correlation with the amplitude, and inputs them into the NGO-RFR model to fit the amplitude of vibration, and predicts the future working state of electromechanical equipment for a period of time. Finally, the method is verified by using the working data of the bearings to verify the validity and feasibility of the method.