Bearings, being essential to modern industry, demand reliable cross-domain diagnostic methods. We propose a transfer learning method for bearing fault diagnosis based on a multi-channel Transformer model with shift windows. The relationship between segmented patch sequences was modeled through self-attention calculation using non-overlapping shift windows. A new partitioning strategy is employed that shifts windows and alternates between two distinct methods to create cross-window connections. To capture basic signal features while preserving positional details, several convolutional layers are introduced prior to the Transformer block. We propose a multi-channel calibration module to ensure stable optimization of the model. Additionally, we introduce a joint maximum mean discrepancy method to measure the distance between the source and target domains. Experimental results demonstrate that the proposed approach achieves superior diagnostic accuracy and offers reliable support for monitoring and predicting the state of rotating machinery.

Owing to high transmission ratio and compact structure, planetary gearboxes are widely used in industrial applications. However, due to running under noisy environment and time-varying rotational speed conditions, the acquired vibration signals are nonstationary, which seriously degrade the performance of fault diagnosis methods of planetary gears. To address this problem, a fault diagnosis method of planetary gears using a stacked denoising autoencoder (SDAE) and a gated recurrent unit neural network (GRUNN) is proposed in this paper. First, a hybrid model based on SDAE and GRUNN is developed to remove noise components from input data and process pre- and post-correlation time-series data. The training samples for planetary gear fault diagnosis are regarded as the input data of the developed hybrid model to automatically extract robust fault features. Then, the training process of the developed hybrid model is presented. The Adam optimization algorithm is utilized to optimize the parameters, and the dropout technique is employed to prevent from overfitting. Finally, a softmax classifier is employed to identify planetary gear states in test samples. The effectiveness of the proposed method is validated through a fault diagnosis experiment of planetary gears. The experimental results show that the proposed method possesses strong anti-noise ability and adaptability to time-varying rotational speed.

In order to address the problem that redundant condition attribute nodes and poor reasoning ability of flow graph may lead to high computational burden and low diagnosis accuracy, a fault severity identification method of roller bearings using flow graph and non-naive Bayesian inference is put forward in this paper. First, a normalized flow graph constructed according to fault features of roller bearings extracted from training samples is used to represent and describe the causal relationship among attributes. Then, the significance degree of condition attribute node with respect to the decision attribute node set is defined to quantitatively measure the impact of the node on the decision-making abilities of the flow graph. A node reduction algorithm based on significance degree of condition attribute node is developed to delete redundant or irrelevant condition attribute nodes, which can improve clustering distribution and reduce computational complexity. Finally, non-naive Bayesian inference is utilized to extend the flow graph to make it applicable in the tasks of reasoning, and an non-naive Bayesian inference algorithm based on flow graph is presented to identify roller-bearing conditions in test samples. The effectiveness of the proposed method is validated through a fault severity identification experiment of roller bearings. Fault severity identification results show that the proposed method can intuitively and accurately recognize fault severities of roller bearings.

To accurately evaluate the remaining life (RUL) of rolling bearings under small sample conditions and strong noise interference, a RUL prediction scheme using adaptive variational mode decomposition (VMD) and double-discriminator conditional CycleGAN (DD-cCycleGAN) is put forward. Combining chimp optimization algorithm (ChOA) with VMD, an adaptive VMD algorithm based on ChOA is presented, which selects effective mode components for reconstruction and reduces interference from strong background noise. A DD-cCycleGAN is developed to generate new samples which not only retain sample information of source domain, but also resemble samples of target one. A LSTM network after training is utilized to predict the bearing RUL in test samples. The performance of this scheme was validated by using the XJTU-SY bearing test dataset. The comparison analyses demonstrate this scheme has strong noise resistance and high accuracy.