The Research Of Rotating Machine Fault Diagnosis Based On Time-domain Analysis

In the modern industry, rotating machine (motor, bearing and shaft) is widely used for power source. The rotating machine works under the hostile environment, works a long time and start frequently, and it occur faults easily. If it can inspect the state of the rotating machine effectively, find the fault and repair in time, it should be able to avoid the chain fault, guarantee the safety and make good use of the machine. So, the rotating machine fault diagnosis has the practical meaning.In the fault diagnosis field, frequency-domain analysis has been mature. It analyses the characteristic frequency of the current and vibration signal to do fault diagnosis; The frequency-domain analysis is very effective when the rotating speed is invariable, because the vibration frequency and the rotating speed have the direct relation. In general, the rotating speed is variable and the vibration signals obtained from a rotating machine are time-variant since they are strongly related to the rotating speed that is not constant even in the macro steady state. Since the mostly used signal processing method, the Fourier analysis is only suitable for stationary signals, the development of the joint time-frequency analysis is demanded. In this paper, it research several fault diagnosis method in time-domain: Multivariate Statistical Process Control Method(MSPC), Parzen Window Probability Density Estimation Method, Parks' Vector Method; And the recent Blind Signal Processing Method is used in the frequency-domain analysis.Validating the fault diagnosis algorithm by the data, the data is obtained from a rotating machine model. The experiment improve that the MSPC and the Parzen Window Method can inspect the faint fault signals effectively, it can not be done by frequency-domain analysis. By doing image matching between the current Parks' vector images, it can separate the faults. The Blind Signal Processing Method can separate the two fault signals which occur synchronously, it improve the veracity of the fault diagnosis.