A Fault Diagnosis Method of Rolling Bearing Based on Improved Recurrence Plot and Convolutional Neur

For the vibration signal of rolling bearing with non-stationary and nonlinear features, the traditional timefrequency analysis method is unable to well mine the nonlinear information of the bearing signal, and traditional machine learning requires complex feature engineering. This paper proposes a fault diagnosis method for rolling bearings based on recurrence plot and convolutional neural network (CNN). A recurrence plot can better mine the nonlinear information inside the bearing signal, and CNN can self-learn the nonlinear information in the recurrent plots and complete the classification task. The experimental show that the method can accurately diagnose bearing faults, which is improved compared to existing fault diagnosis methods. This research provides a new idea and way of bearing fault diagnosis and has specific practical value.