Rolling bearing fault acoustic diagnosis method based on time domain equivalent source method

The invention discloses a rolling bearing fault acoustic diagnosis method based on a time domain equivalent source method. The diagnosis method comprises the following steps of: firstly, respectively acquiring extremely near field sound pressure signals radiated by healthy and faulty sample bearings during operation by using a microphone, and carrying out feature extraction on the sound pressure signals of the sample bearings to obtain a training set; learning the training set through a machine learning algorithm to obtain a classification model; then, using a microphone array for collecting sound pressure signals radiated by taking the target bearing to be detected as a sound source in a near field, and reconstructing the sound source intensity of the target bearing by using a time domain equivalent source method; and finally, extracting features based on the reconstructed sound source intensity of the target bearing, obtaining a sample with an unknown label, and performing fault diagnosis by using a classif