Continuous Evolution Learning: A Lightweight Expansion-Based Continuous Learning Method for Train Transmission Systems Fault Diagnosis

TL;DR: This paper proposes a lightweight expansion-based continual learning method for train transmission systems fault diagnosis, reducing computational costs and improving model evolution through a hash space metric mechanism and joint knowledge enhancement and compression strategy.

Abstract: The dynamic fault environment, incremental data accumulation, and specific needs in train transmission systems make continual learning essential for fault diagnosis. Recent advancements in continual learning have improved diagnostic adaptability, but current methods face challenges: 1) Complex architectures to prevent catastrophic forgetting increase training difficulty and computational costs, hindering deployment. 2) Lack of new class samples leads to delayed model evolution due to long sample accumulation periods. This article introduces a lightweight continual learning method based on model expansion. A hash space metric mechanism using lightweight convolution is developed to reduce computational costs while maintaining accuracy. Additionally, a joint strategy of knowledge enhancement and compression improves model evolution by refining knowledge subsets. Compared with the state-of-the-art method, the proposed method reduces parameters by 2.59 times, FLOPs by 2.67 times, and inference time by 2.89 times, and leads by 2.5% and 0.23% in incremental accuracy and incremental forgetting rate, respectively.