Currently gear fault diagnosis is mainly based on vibration signals with a few studies on\nacoustic signal analysis. However, vibration signal acquisition is limited by its contact measuring\nwhile traditional acoustic-based gear fault diagnosis relies heavily on prior knowledge of signal\nprocessing techniques and diagnostic expertise. In this paper, a novel deep learning-based gear\nfault diagnosis method is proposed based on sound signal analysis. By establishing an end-to-end\nconvolutional neural network (CNN), the time and frequency domain signals can be fed into the\nmodel as raw signals without feature engineering. Moreover, multi-channel information from\ndifferent microphones can also be fused by CNN channels without using an extra fusion algorithm.\nOur experiment results show that our method achieved much better performance on gear fault\ndiagnosis compared with other traditional gear fault diagnosis methods involving feature engineering.\nA publicly available sound signal dataset for gear fault diagnosis is also released and can be\ndownloaded as instructed in the conclusion section.
Loading....