Detction And Diagnosis Of Rotating Machinery Fault

It is often hard to avoid faults when the rotating machinery is in use. The whole machine will be affected, once a component is breakdown. In order to reduce the unnecessary disasters as far as possible, it’s extremely needed to check out the faults timely. Since the rolling bearing is one of the most popular components in mechanical equipment, detection and diagnosis of rolling bearing fault are mainly studied in this paper.There are mainly three steps in diagnosis of mechanical fault signals. Firstly, the signals are captured. Secondly, features of fault signals are extracted. Finally, the faults of the signals are detected and diagnosed. It’s the same three-step analysis for the rolling bearing fault detection.First of all, for the purpose of authority, the test data in this paper is obtained from the experiments of Case Western Reserve University. Secondly, the features of the fault data are extracted by means of wavelet analysis, like wavelet decomposition and reconstruction, and Hilbert envelope spectrum analysis is also used. The process is implemented by in the Matlab software. Finally, the Naive Bayesian Classifier is introduced in this paper, and the fault diagnosis is taken as a classification process. In order to have the Naive Bayesian Classifier applied in diagnosis of the rolling bearing fault properly, the extracted features that are characterized by continuous values should be transformed into some new features that are discrete in value. The test data can also be clearly expressed by the new features. In this way, the application of Naive Bayesian Classifier will be easily realized. It’s indicated that the accuracy of this fault diagnosis system is pretty high when some fault signals are tested. In the view of innovation, it’s quite a new combination of wavelet envelope spectrum analysis and Naive Bayesian Classifier in the field of the mechanical fault diagnosis. The faults can be effectively detected by this system. The system can be expanded to the diagnosis of other faults but not just for faults of rolling bearing.