In recent years, deep learning methods, especially convolutional neural network (CNN), have received extensive attentions and applications in fault diagnosis. However, recent studies have shown that the shift-invariance of CNN is not good enough, resulting in fragile feature extraction and sharp reduction in model performance when the shift occurs in the input. To improve the shift-invariance of CNN, considering the periodic characteristics of vibration signals, a multiscale CNN with feature alignment (MSCNN-FA) is proposed for bearing fault diagnosis under different working conditions. First, by analyzing the operating principles of the convolutional layer and pooling layer, a feature alignment module including single-stride convolution layer, adaptive max-pooling layer, and global average pooling layer is designed to obtain aligned features. Next, to extract shift-invariant robust features from vibration signals, a multiscale convolution strategy is utilized, and a feature-aligned multiscale feature extractor is constructed. Finally, a classifier composed of fully connected (FC) layers is constructed for bearing fault diagnosis. The effectiveness of the method is verified by a rolling bearing experiment, which outperforms other related existing CNN-based methods in terms of diagnosis accuracy and feature robustness.

To assess health conditions of rotating machinery efficiently, multiple accelerometers are mounted on different locations to acquire a variety of possible faults signals. The statistical features are extracted from these signals to identify the running status of a machine. However, the acquired vibration signals are different due to sensor's arrangement and environmental interference, which may lead to different diagnostic results. In order to improve the fault diagnosis reliability, a new multisensor data fusion technique is proposed. First, time-domain and frequency-domain features are extracted from the different sensor signals, and then these features are input into multiple two-layer sparse autoencoder (SAE) neural networks for feature fusion. Finally, fused feature vectors can be regarded as the machine health indicators, and be used to train deep belief network (DBN) for further classification. To verify the effectiveness of the proposed SAE-DBN scheme, the bearing fault experiments were conducted on a bearing test platform, and the vibration data sets under different running speeds were collected for algorithm validation. For comparison, different feature fusion methods were also applied to multisensor fusion in the experiments. Experimental results demonstrated that the proposed approach can effectively identify the machine running conditions and significantly outperform other fusion methods.

Deep neural networks present very competitive results in mechanical fault diagnosis. However, training deep models require high computing power while the performances of deep models often suffer from the lack of sufficient training data. In this paper, a transferable convolutional neural network (TCNN) is proposed to improve the learning of target tasks. Firstly, one-dimension CNN is constructed and pre-trained based on large source task datasets. Then a transfer learning strategy is adopted to build a deep model on target tasks by reusing the pre-trained network. Thus, the proposed method not only utilizes the learning power of deep network but also leverages the prior knowledge from the source task. Four case studies are performed, and the effects of transfer layers and training sample size on classification effectiveness are investigated. Results show that the proposed method exhibits better performance compared with other algorithms.