Few-shot Class-incremental Fault Diagnosis by Preserving Class-Agnostic Knowledge with Dual-Granularity Representations

TL;DR

This paper introduces a Dual-Granularity Guidance Network (DGGN) for few-shot class-incremental fault diagnosis, effectively addressing catastrophic forgetting and overfitting by decoupling feature learning into class-specific and class-agnostic streams with dynamic fusion.

Contribution

The paper proposes a novel DGGN framework that explicitly separates fine- and coarse-grained features and introduces a multi-semantic cross-attention mechanism for improved fault diagnosis in few-shot settings.

Findings

DGGN outperforms state-of-the-art methods on TEP and MFF datasets.

The multi-semantic cross-attention effectively fuses class-specific and class-agnostic features.

The Boundary-Aware Exemplar Prioritization reduces catastrophic forgetting.

Abstract

Few-Shot Class-Incremental Fault Diagnosis (FSC-FD), which aims to continuously learn from new fault classes with only a few samples without forgetting old ones, is critical for real-world industrial systems. However, this challenging task severely amplifies the issues of catastrophic forgetting of old knowledge and overfitting on scarce new data. To address these challenges, this paper proposes a novel framework built upon Dual-Granularity Representations, termed the Dual-Granularity Guidance Network (DGGN). Our DGGN explicitly decouples feature learning into two parallel streams: 1) a fine-grained representation stream, which utilizes a novel Multi-Order Interaction Aggregation module to capture discriminative, class-specific features from the limited new samples. 2) a coarse-grained representation stream, designed to model and preserve general, class-agnostic knowledge shared across all fault types. These two representations are dynamically fused by a multi-semantic cross-attention mechanism, where the stable coarse-grained knowledge guides the learning of fine-grained features, preventing overfitting and alleviating feature conflicts. To further mitigate catastrophic forgetting, we design a Boundary-Aware Exemplar Prioritization strategy. Moreover, a decoupled Balanced Random Forest classifier is employed to counter the decision boundary bias caused by data imbalance. Extensive experiments on the TEP benchmark and a real-world MFF dataset demonstrate that our proposed DGGN achieves superior diagnostic performance and stability compared to state-of-the-art FSC-FD approaches. Our code is publicly available at https://github.com/MentaY/DGGN