FaultDiffusion: Few-Shot Fault Time Series Generation with Diffusion Model

2025-11-19

AI Summary

In industrial equipment monitoring, fault data are scarce due to limited samples and high annotation costs, while existing generative models struggle to capture large inter-class discrepancies and high intra-class variability. To address this, we propose a diffusion-based few-shot fault time-series generation framework. Our method introduces: (1) a positive–negative discrepancy adapter that explicitly models the deviation between normal and fault domains, leveraging a pre-trained normal distribution to guide fault synthesis; and (2) cross-sample discrepancy regularization and a diversity loss to mitigate mode collapse, thereby enhancing both fidelity and diversity of generated samples. Evaluated on multiple benchmark datasets, our approach consistently outperforms state-of-the-art generative models. It is the first to simultaneously achieve high fidelity and high diversity in few-shot fault time-series generation, establishing new state-of-the-art performance in this domain.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Computer Vision: Diffusion Models for Vision

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Web data generation and simulation
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

In industrial equipment monitoring, fault diagnosis is critical for ensuring system reliability and enabling predictive maintenance. However, the scarcity of fault data, due to the rarity of fault events and the high cost of data annotation, significantly hinders data-driven approaches. Existing time-series generation models, optimized for abundant normal data, struggle to capture fault distributions in few-shot scenarios, producing samples that lack authenticity and diversity due to the large domain gap and high intra-class variability of faults. To address this, we propose a novel few-shot fault time-series generation framework based on diffusion models. Our approach employs a positive-negative difference adapter, leveraging pre-trained normal data distributions to model the discrepancies between normal and fault domains for accurate fault synthesis. Additionally, a diversity loss is introduced to prevent mode collapse, encouraging the generation of diverse fault samples through inter-sample difference regularization. Experimental results demonstrate that our model significantly outperforms traditional methods in authenticity and diversity, achieving state-of-the-art performance on key benchmarks.

Problem

• Generating realistic fault time series with limited data
• Addressing scarcity of annotated industrial fault datasets
• Improving authenticity and diversity in few-shot fault generation

Innovation

• Diffusion model generates few-shot fault time series
• Positive-negative adapter models normal-fault domain differences
• Diversity loss prevents mode collapse in samples