Fault Diagnosis of Pulse Solenoid Valve Based on Ensemble Classifier
Pulse solenoid valve has the advantages of simple structure,low cost,fast action and etc.It is widely used in the field of bag dust removal,so the health status of its components has great significance for the safety of system operation.This paper conducts in-depth research on the diagnostic methods of common faults of pulsed solenoid valves,uses the proximity switch to detect the opening hysteresis time and hold time of the main spool as the feature,constructs the Bagging-DT(RF)classifier model for fault diagnosis,and verifies the experimental results of a variety of simulated faults and the comparative results of various models,showing that the precision(accuracy),recall(recall),and F1 values of the Bagging-DT model constructed in this paper are better than other models.And its accuracy rate reaches 86%,which realizes more accurate prediction and judgment of five or more faults of the pulse solenoid valve,and solves the problem that the staff relies on subjective experience in actual work and the credibility of the diagnosis effect is low.