Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 19, 2026Scientific ReportsOpen Access

Physics-guided cross-domain adaptation: a hierarchical hybrid transformer framework with contrastive learning for robust fault diagnosis under variable working conditions

View Full Paper
Ask AI
Bookmark
Share

Authors

LLLumin LiuCHChen HaoLXLin Xiong

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates robust fault diagnosis under variable conditions, highlighting effective feature alignment.

Key Points

  • This paper aims to enhance the robustness of fault diagnosis models against domain shifts caused by variable working conditions.
  • Proposes a Hierarchical Hybrid Transformer network with Contrastive Learning.
  • Implements a physics-guided cross-domain contrastive learning strategy for feature alignment.
  • Conducts extensive experiments on two public bearing datasets to validate the framework's effectiveness.
  • Achieves average diagnostic accuracies of 94.94 ± 0.32% and 90.26 ± 0.50% on two datasets.
  • Demonstrates superior mean accuracy across 18 transfer tasks compared to state-of-the-art models.
  • Exhibits robustness under significant domain shift and noise conditions.

Cite This Study

Liu et al. (2026) studied this question.

synapsesocial.com/papers/6a34dbc465a5b0777af2c68bhttps://doi.org/10.1038/s41598-026-57900-9
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-guided adaptive spectral transformation for unsupervised cross-domain vibration-based compound fault diagnosis2026
  2. 2An interpretable cross-domain bearing fault diagnosis method based on domain-aware and informed-guided domain adaptation2026
  3. 3Interpretable domain adaptation transformer: a transfer learning method for fault diagnosis of rotating machinery2024 · 55 citations
  4. 4Partial Domain Adaptation for Intelligent Machinery Fault Diagnosis2025
  5. 5A hybrid dynamic adversarial domain adaptation network with multi-channel attention mechanism for rotating machinery unsupervised fault diagnosis under varying operating conditions2024 · 4 citations