Feature Recognition of Engine High-Speed Rolling Bearing Wear Fault Signals
Long term use of high-speed rolling bearings in harsh working environments can lead to wear faults,seriously affect-ing the normal operation of the engine.There are various types of faults,and each type of fault has its unique characteristics,making it difficult to identify fault signal features.To effectively solve this problem,a feature recognition method for wear fault signals of high-speed rolling bearings in engines is proposed.Through EEMD decomposition and reconstruction of the vibration signal of the high-speed rolling bearing of the engine,the natural vibration mode function IMF is obtained.Based on the ob-tained IMF,a Hankelmatrix is constructed to obtain the spliced singular value sample eigenvectors.By dividing the sample space and setting iterative thresholds,a fuzzy clustering algorithm is used to cluster fault samples,and the membership of each sample in different clusters is calculated to obtain its proximity and fault characteristics.Experimental results show that the proposed al-gorithm can better identify mechanical faults with high accuracy and low probability of false positives and false positives,ensur-ing the safe operation of equipment.
EEMD AlgorithmFCM AlgorithmEngine High-Speed Rolling BearingMechanical Failure of BearingFeature Vector Extraction
NETL
NSTL
万方数据
维普
