A rolling bearing fault diagnosis method based on vibro-acoustic data fusion and fast Fourier transform (FFT)

In recent years, fault diagnosis based on fusion data has become a research hotspot, but most of the existing fusion methods are based on single-mode signals, which not only has incomplete fault information, but also has the risk of noise interference. In order to overcome the limitation of single-mode signal, multi-mode signal is used for fault diagnosis. A rolling bearing fault diagnosis method based on vibro-acoustic data fusion and fast Fourier transform is proposed. First of all, the data of different modes of bearings are intersected to make vibro-acoustic data complement each other and make up for the deficiency of fault information of single-mode data. Secondly, according to the number of partitions, the fusion data is partitioned at intervals to effectively improve the problem of uneven noise distribution in training set, validation set and test set. Then, according to the sampling length and sampling step, the fusion data is periodically sampled to solve the problem of unreasonable sampling. Finally, the data are further processed by fast Fourier transform and input into the network, and the traditional convolutional neural network is adjusted to extract more detailed fault features from the sound data, so that the fusion data can obtain the best recognition effect. The experimental results show that the recognition accuracy of this method on the fusion data reaches 99.98%. Compared with single-mode data, this method has higher accuracy and better robustness.