An Overview Of Predictive Maintenance For Industrial Machine Using Vibration Analysis

In this paper, an attempt is made to overview different vibration analysis techniques and predictive mainte-nance for industrial machines. The techniques considered in this paper involves analysis in time-domain, frequency-domain, time-frequency representation (TFR), Wavelet transforms, time synchronous averaging (TSA), and order analysis. Convolution neural networks (CNNs) for feature learning are also discussed. The vibration signal reveals more information about the fault development of industrial machine and the evolution of indus-trial machine health monitoring methods and its mostly used techniques are reviewed in this paper. The machine lifetime en-hancement can be achieved by condition-based monitoring which takes place to keep away from any impulsive failure appearing before the upcoming schedule. The Vibration, Wavelet, Voltage, Acoustic Emission, Temperature, Current, Noise, etc., are some measurements used for industrial machine health monitoring. It is proposed to use wavelet transform along with CNN for vibration-based health monitoring and predictive maintenance.