2 July 2024 in Mechanical Systems and Signal Processing
6 November 2025 in Mechanical Systems and Signal Processing
22 July 2024 in Measurement Science and Technology
Graph neural network (GNN) has the proven ability to learn feature representations from graph data, and has been utilized for the tasks of predicting the machinery remaining useful life (RUL). However, existing methods only focus on a single graph structure and cannot integrate the correlation information contained in multi-graph structures. To address these issues, a multi-graph structure GNN prediction method with attention fusion (MGAFGNN) is proposed in this paper for GNN-based bearing RUL prediction. Specifically, a multi-channel graph attention module is designed to effectively learn the similar features of node neighbors from different graph data and capture the multi-scale latent features of nodes through the nonlinear transformation. Furthermore, a multi-graph attention fusion module (MGAFM) is proposed to extract the collaborative features from the interaction graph, thereby fusing the feature embeddings from different graph structures. The fused feature representation is sent to the long short-term memory network to further learn the temporal features and achieve RUL prediction. The experimental results on two bearing datasets demonstrate that MGAFGNN outperforms existing methods in terms of prediction performance by effectively incorporating multi-graph structural information.
8 April 2024 in IEEE Transactions on Instrumentation and Measurement
The quantitative diagnosis of rolling bearing defects still mainly relies on the manual analysis of vibration signals and is limited to a specific moment in time, which restricts the intelligent identification of life-cycle defect evolution. In this article, a novel digital twin-driven graph convolutional memory network (GCMN) is proposed for evaluating the defect evolution of rolling bearings throughout the whole life. In the proposed method, a dynamic twin model is constructed to generate the vibration responses that characterize the state of bearings. The twin model is capable of accurately simulating the operational conditions of the bearing and interacting with the actual responses, thereby enhancing the accuracy of the model. In addition, a graph network model GCMN is developed to transfer knowledge from the twin model to the physical entity through domain adaptation, thereby revealing the relationship between vibration responses and defect sizes. It extracts spatial features through nonlinear transformation of graph data and incorporates temporal features via the hidden layer state at the previous moment. The experimental results demonstrate that the proposed method accurately characterizes the local defect extension throughout the bearing entire lifespan.
3 September 2024 in IEEE Sensors Journal
Convolution neural network (CNN)-based transfer learning (TL) has been investigated for knowledge transfer in machinery fault diagnosis. However, there remain certain challenges that need to be addressed: firstly, the optimization of deep network models is hindered by the substantial number of parameters involved; and secondly, the limitation of training data impacts the diagnostic accuracy of the model. In this work, a multi-level residual CNN based on dynamic feature fusion (MRCNN-DFF) is proposed for TL in machinery fault diagnosis. In MRCNN-DFF, depth-wise separable (DS) convolution is introduced to reduce the number of trainable parameters, which offers the advantages and enhances the efficiency of model optimization. Meanwhile, to fully mine the useful information, parallel MRCNN channels are proposed to extract high-dimensional features. Then, the DFF is designed to fuse the bi-channel information according to the feature importance, which effectively incorporates the key information contained in the limited data. Finally, the parameter transfer is utilized to transfer domain-shared information. The MRCNN-DFF is validated employing two bearing datasets, and the results demonstrate that the proposed MRCNN-DFF surpasses the comparison methods in terms of performance.
12 November 2026 in Mechanical Systems and Signal Processing