Concurrent Fault Diagnosis Based on Bayesian Discriminating Analysis and Time Series Analysis With Dimensionless Parameters

In mechanical fault diagnosis, the improvement of diagnosis accuracy is the ultimate aim for researchers. Discriminative features selected and extracted from vibration measurements are commonly used to identify fault types. When a certain distinguishing characteristic is exhibited in a feature, numerous classification methods are applied for pattern recognition and diagnosis decisions. However, it is difficult to conduct accurate diagnosis if the distinguishing ability of the extracted or optimized feature remains poor. To this end, this paper proposes a new approach for classifying concurrent faults in rotating machinery, based on Bayesian discriminating analysis and time series analysis, which can solve the problem of there being no characteristic parameter that can provide a degree of discrimination information for concurrent fault types. In order to demonstrate the validity of the proposed methodology, several experiments on concurrent fault diagnosis are conducted on rotating machinery. The experimental results indicate that the method can effectively improve the accuracy of concurrent fault diagnosis.