Multibranch and Multiscale Dynamic Convolutional Network for Small Sample Fault Diagnosis of Rotating Machinery

Deep learning methods have been widely used in the field of fault diagnosis of rotating machinery. As one of the deep learning methods, a multiscale convolutional neural network (MSCNN) can achieve accurate fault diagnosis by extracting feature information of vibration signals at different scales. However, in practical applications, a small number of fault samples limit the performance of MSCNN. To overcome this problem, a small sample rotating machinery fault diagnosis method based on a multibranch and multiscale dynamic convolutional network (MBSDCN) is proposed. First, a feature splitting strategy is proposed, which splits a single input into multibranch inputs to ensure that each multiscale convolutional layer is matched to the appropriate input. Second, a channel reconstruction (CR) attention mechanism is designed, and CR attention mechanism can form a direct mapping of high-dimensional space with the fully connected layer through the reconstruction processing of channels, thus making the connection between feature channels closer. Finally, a novel multiscale feature extraction model is constructed, in which multiscale convolutional layers are applied to extract multibranch input features, and the CR attention mechanism is used to calibrate the contribution of multiscale convolution layers. The performance of MBSDCN is evaluated with Case Western Reserve University (CWRU) rolling bearing dataset and University of Connecticut (UoC) gearbox dataset and is compared with some advanced deep learning methods. The comparison results indicate that the proposed MBSDCN can accurately diagnose faults under small sample conditions. In addition, the experimental results in noisy environments indicate that the proposed MBSDCN has satisfactory anti-noise performance under small sample conditions and noisy environments.