Skip to content

Semantics-guided contrastive pre-training with cluster-aware decision making for zero-shot bearing fault diagnosis

Sep 2026 · Science China Technological Sciences · Vol 69 · 1 citation · 35 references

TL;DR

Fault diagnosis contrastive language-signal pre-training (FD-CLSP) is proposed, a zero-shot cross-domain framework for bearing fault diagnosis under unseen operating conditions that leverages the noise-robustness and domain-invariance of high-level semantics to guide the extraction of elusive physical features.

View source

Similar papers

Open access Aug 2026

Hybrid loss-driven source-free domain adaptation with post-hoc explainability for bearing fault diagnosis

A novel and robust Source-Free Domain Adaptation (SFDA) framework equipped with post-hoc explainability for bearing fault diagnosis is proposed, and a post-hoc visualization mechanism is incorporated to explain the model's decision-making process, enhancing the transparency and credibility of the diagnosis for practica...

Cheng-Hao Yan, Dong-Sheng Liu, Tong Wu et al. · 0 citations
2026

Heterogeneous Mixture-of-Experts Network for Bearing Fault Diagnosis Under Strong Noise

Severe noise in industrial environments often leads to distribution shifts, posing a critical challenge for deep learning-based bearing fault diagnosis. Models trained on clean data typically degrade under such conditions due to their inability to effectively disentangle noise from fault-related features. To address th...

Jin-Ze Zhang, Lu Yang, Tian-Tian Xu et al. · 0 citations
2026

Cross-Modal Guidance Learning for Zero-Shot Industrial Anomaly Detection

Zero-shot industrial anomaly detection (ZIAD) aims to develop a unified model capable of directly identifying unseen anomaly categories in images without requiring reference samples. Recently, large-scale Vision-Language Models (VLMs) such as CLIP have shown great potential for solving this task. However, existing meth...

Tiyu Fang, Lin Zhang, Ran Song et al. · 0 citations
Open access 2026

LFA-MFNet: Local Feature-Aligned Multisource Information Fusion for Fault Diagnosis Under Limited and Imbalanced Conditions

Real-world data-driven fault diagnosis is constrained by scarce fault samples, limited and imbalanced (L&I) class distributions, and heterogeneous multisensor data under operating condition shifts, which impede stable and generalizable discriminative representation learning. We propose the local feature alignment multi...

Cheng Peng, Tian-Hong Huang, Zhao-Hui Tang et al. · 0 citations
Conference Open access Aug 2026

Lacofd: A Leakage-Aware Contrastive Few-Shot Intelligent Diagnosis Framework for Cross-Condition Rotating Machinery

Experiments show that contrastive pretraining provides an effective initialization for target adaptation, particularly in the 10 -shot and 20 -shot settings, whereas frozen linear probing remains insufficient under load shifts, indicating that leakage-aware representation learning can improve the reliability of data-dr...

Yi-He Du, Deng-Feng Zhao, Run-Di Zhang · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.