Defect size estimation of rolling bearing life cycle based on envelope analysis and BPNN

A life cycle defect size estimation method for rolling bearings based on envelope analysis and a BPNN is presented to accurately evaluate the degree of bearing damage. The method begins with pre-whitening, band-pass filtering, and Hilbert transform applied to the original vibration signal, which allows for precise identification of the rolling element's entry and exit from the defect area, followed by defect size estimation. Subsequently, the RMS value of the bearing vibration signal is extracted and used as the input for training the BP neural network, with the estimated defect size as the output. This process establishes a mapping relationship between the RMS value and the defect size. Finally, the trained neural network predicts the defect size for the enti re lifecycle of the bearing. This approach not only improves the accuracy of defect size estimation but also offers a new perspective for analyzing defect evolution and predicting the lifespan of bearings.