CMformer: A condition monitoring method for industrial equipment utilizing both channel and temporal dependencies
Published in Advanced Engineering Informatics, 2027
CMformer jointly models dynamic cross-sensor interactions and long-range temporal dependencies for industrial time-series segmentation and operational state detection. Segment-Wise Channel-Enhanced Embedding captures channel dependencies within temporal segments, while Point-to-Segment Attention connects individual time points to multi-scale segment context. Experiments on real-world industrial datasets show improved state labeling and transition localization, supporting predictive maintenance and condition-based maintenance scheduling.
Recommended citation: Xiaorui Li, Zhaolin Yuan, Haoran Qiao, Taohong Zhang, Ming Li, "CMformer: A condition monitoring method for industrial equipment utilizing both channel and temporal dependencies." Advanced Engineering Informatics, 2027.
Download Paper