PulseTrendingJournal ClubResearchersJournalsExplore
Instagram
HomeTrendingJournal ClubExplore
Synapse
⌘+K
Synapse
May 7, 2026SensorsOpen Access

Physics-Enhanced Orthogonal Sensing for Self-Supervised Anomaly Detection in Rolling Mills

View Full Paper
Ask AI
Bookmark
Share

Authors

YWYue WangBZBin ZhengYFYehan Feng

Discussion

Loading...

Member takes

Overview

This approach improves anomaly detection in rolling mills, suggesting enhanced monitoring with transducer technology.

Key Points

  • The research aims to enhance anomaly detection in rolling mills with a new monitoring system.
  • Developed a cyber-physical architecture for real-time anomaly detection.
  • Introduced an embedded orthogonal sensing layout to separate drive-chain vibrations from rolling-force fluctuations.
  • Utilized a two-branch network architecture involving CSD transformer and VQ-VAE for effective data handling.
  • Achieved an AUC-ROC of 0.952 for the anomaly detection system.
  • Maintained a low false alarm rate of 0.048 at a 95% true positive rate.
  • Demonstrated real-time monitoring with approximately 8 ms processing latency and 108 ms fault response time.

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/69fbefc0164b5133a91a3ae3https://doi.org/10.3390/s26092895
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Physics-informed self-supervised diagnosis of rotating machinery using latent ODEs and transformer encoders2026 · 4 citations
  2. 2Intelligent fault diagnosis of rolling mills based on gram angular difference field and dual-attention residual network2025
  3. 3Smart Sensor-Based Monitoring Technology for Machinery Fault Detection2024 · 23 citations
  4. 4Data-Driven Fault Diagnosis for Rotating Industrial Paper-Cutting Machinery2025
  5. 5Distribution Dynamic Direct Orthogonal Decomposition Method for Quality-Related Fault Detection2025