A New Method of Machine Fault Detection Based on Machine Learning on Line

Abstract

The article first analyzes many existing machine running fault diagnosis methods, which may not be good for on-line detection in view of the complexity of the algorithm. At the same time, studies the characteristics of random noise modulation. Then, designs a new on-line self-learning evolutionary fault detection method based on the FFT algorithm. The simulation shows that the method has good detection performance for a weak fault signal.