Multi-Channel Swin Transformer Framework for Bearing Remaining Useful Life Prediction
Ali Mohajerzarrinkelk, Maryam Ahang, Mehran Zoravar, Mostafa Abbasi, Homayoun Najjaran
TL;DR
The paper tackles bearing Remaining Useful Life prediction under noisy, multi-condition operation by introducing MCSFormer, a framework that fuses wavelet-based denoising, Wavelet Packet Decomposition, and a multi-channel Swin Transformer with a hierarchical attention-based fusion. The approach includes a dual-channel convolutional front-end, time-frequency 64×64 representations, and a regression head guided by a custom loss that penalizes late predictions, achieving improved MAE and scoring in intra- and cross-condition tests on the PRONOSTIA dataset. Key contributions include the integration of WPD-derived features with a scalable transformer backbone, and a loss function that prioritizes timely maintenance decisions, reducing operational risks. Practically, MCSFormer demonstrates strong noise resilience, generalization across conditions, and a safety-focused performance profile for predictive maintenance deployments.
Abstract
Precise estimation of the Remaining Useful Life (RUL) of rolling bearings is an important consideration to avoid unexpected failures, reduce downtime, and promote safety and efficiency in industrial systems. Complications in degradation trends, noise presence, and the necessity to detect faults in advance make estimation of RUL a challenging task. This paper introduces a novel framework that combines wavelet-based denoising method, Wavelet Packet Decomposition (WPD), and a customized multi-channel Swin Transformer model (MCSFormer) to address these problems. With attention mechanisms incorporated for feature fusion, the model is designed to learn global and local degradation patterns utilizing hierarchical representations for enhancing predictive performance. Additionally, a customized loss function is developed as a key distinction of this work to differentiate between early and late predictions, prioritizing accurate early detection and minimizing the high operation risks of late predictions. The proposed model was evaluated with the PRONOSTIA dataset using three experiments. Intra-condition experiments demonstrated that MCSFormer outperformed state-of-the-art models, including the Adaptive Transformer, MDAN, and CNN-SRU, achieving 41%, 64%, and 69% lower MAE on average across different operating conditions, respectively. In terms of cross-condition testing, it achieved superior generalization under varying operating conditions compared to the adapted ViT and Swin Transformer. Lastly, the custom loss function effectively reduced late predictions, as evidenced in a 6.3% improvement in the scoring metric while maintaining competitive overall performance. The model's robust noise resistance, generalization capability, and focus on safety make MCSFormer a trustworthy and effective predictive maintenance tool in industrial applications.