Fault Diagnosis Model for Aircraft Hydraulic System Based on VPNRS-RF
In order to solve the problems of redundancy and low accuracy in fault diagnosis of aircraft hydraulic system,a feature selection algorithm based on the combination of Variable Precision Neighborhood Rough Set(VPNRS)and Random Forest(RF)was proposed,and a fault diagnosis model of aircraft hydraulic system was established based on this algorithm.The VPNRS-RF algorithm mainly uses the random forest algorithm to sort the importance of the features after reducing the rough set of variable pre-cision neighborhood and fuzzy entropy,again after the screening to determine the optimal feature subset,using the optimal feature subset for online sequential extreme learning machine(OSELM)classification model for training,to improve fault information re-trieval accuracy.Finally,the VPNRS-RF-OSELMmodel is verified by the simulation of aircraft landing gear retraction system.
万方数据
维普
