A decoupling method for compound faults of rolling bearings based on multi-modal bidirectional cross-attention fusion and contrastive learning
Authors:
- 24 October 2025
- journal article
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 36 (10) , 106131
Abstract
In compound fault diagnosis of rolling bearings, traditional single-modal analysis and existing multimodal fusion methods suffer from issues including insufficient cross-modal interaction, limited adaptive weighting capability, and poor discriminability in the feature space, which leads to ambiguous decoupling of compound fault features. Which leads to ambiguous decoupling of compound fault features. For this purpose, this paper proposes a decoupling method for compound faults of rolling bearings based on multimodal bidirectional cross-attention fusion and contrastive learning. First, an autoencoder extracts temporal features from vibration signals, while adaptive rectangular convolution dynamically captures multi-scale spatial features of time-frequency images. Secondly, a bidirectional cross-attention mechanism is constructed to achieve cross-modal semantic alignment through signal-to-image and image-to-signal dual guidance. Furthermore, a dynamic gating mechanism is introduced, which generates adaptive fusion weights based on the characteristics of the samples. It dynamically adjusts contribution ratio of each modality to avoid information masking. Finally, orthogonal subspace projection maps multimodal features into independent fault subspaces. When combined with contrastive learning, this approach enhances both intra-class compactness and inter-class separability, forming clear decoupling boundaries for precise fault decoupling. Experiments on PU and lab datasets show that the proposed method achieves average decoupling accuracies of 99.36% and 98.76% under various conditions. Ablation studies further confirm the synergistic effectiveness of bidirectional cross-attention, dynamic gating, and contrastive learning, offering an industrially viable approach for decoupling compound faults in rolling bearings.Keywords
Funding Information
- Liaoning provincial department of education (LJ242410153063)
- Science and Technology Innovation and Entrepreneurship Fund of China Coal Technology and Engineering Group (2024-ZD003)
This publication has 31 references indexed in Scilit:
- Fault diagnosis of multimodal feature fusion convolutional neural network based on differential evolution optimizationComputers and Electrical Engineering, 2025
- A novel method for untrained detection of compound fault in rolling bearing via fast Fourier Transform-Transformer modelMeasurement, 2025
- Domain reinforcement feature adaptation methodology with correlation alignment for compound fault diagnosis of rolling bearingExpert Systems with Applications, 2024
- Phase-based video vibration measurement and fault feature extraction method for compound faults of rolling bearingsAdvanced Engineering Informatics, 2024
- Domain adaptation with label-aligned sampling (DALAS) for cross-domain fault diagnosis of rotating machinery under class imbalanceExpert Systems with Applications, 2023
- A self-attention based contrastive learning method for bearing fault diagnosisExpert Systems with Applications, 2023
- Compound fault diagnosis for photovoltaic arrays based on multi-label learning considering multiple faults couplingEnergy Conversion and Management, 2023
- A novel convolutional neural network with interference suppression for the fault diagnosis of mechanical rotating componentsNeural Computing and Applications, 2022
- Rolling bearing fault diagnosis based on feature fusion with parallel convolutional neural networkThe International Journal of Advanced Manufacturing Technology, 2020
- Multi-stream convolutional neural network-based fault diagnosis for variable frequency drives in sustainable manufacturing systemsProcedia Manufacturing, 2020