Fault Diagnosis Of Inter-Shaft Bearing Based On Information Entropy

The inter-shaft bearing works in the high temperature,high speed and lubrication difficult working environment,prone to failure,once the inter-shaft bearing failure will cause fatal impact on the aircraft engine.Therefore,it is of great significance to realize fault diagnosis of inter-shaft bearing for aeroengine status detection and fault diagnosis.In view of the problems such as long transmission path of fault signal of inter-shaft bearing and difficulty in obtaining fault characteristics,this paper combines information entropy and quantum theory to propose a fault diagnosis method of random forest inter-shaft bearing based on optimal characteristics.Firstly,according to the working position and structural characteristics of the inter-shaft bearing of the aero-engine,the fault simulation test bench of the double-rotor inter-shaft bearing is designed,and the fault simulation test bench of the outer ring,inner ring and rolling element is used to simulate the fault,and the acoustic emission signal is collected.Then,according to the basic principles of quantum theory,the multi-qubit system of acoustic emission signals is established,and the approximate entropy,sample entropy,permutation entropy and multi-scale entropy are compared and analyzed to construct the calculation method of multi-scale quantum entropy.The effectiveness of the multi-scale quantum entropy algorithm is verified by gaussian white noise and pink noise.Finally,based on the multi-scale quantum entropy theory,the optimal feature extraction of acoustic emission signals was carried out,and the fault feature vector composed of multi-scale quantum entropy values at six scales was constructed.Then,the fault diagnosis of the inter-shaft bearing was carried out by using the random forest algorithm.The results show that the multi-scale quantum entropy proposed in this paper can effectively describe the complexity of the fault signal of the inter-shaft bearing.Multi-scale quantum entropy and random forest algorithm with optimal characteristics are used to realize fault diagnosis of inter-shaft bearing,which improves the accuracy of fault diagnosis.This research is of great significance to improve the safety and reliability of aero-engine operation.