arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.18138eess.SYcs.SY
收藏

基于自监督学习的电力电子变换器健康监测大模型初步探索

Preliminary Exploration of Large Models Based on Self-Supervised Learning for Health Monitoring of Power Electronic Converters

The Hong Kong Polytechnic University(香港理工大学);Wuhan University of Technology(武汉理工大学);Huazhong University of Science and Technology(华中科技大学);Three Gorges Intelligent Control Technology Co., Ltd.(三峡智能控制技术有限公司)

Xinyuan Liao, Xiangtian Deng, Yang Liu, Zhaobing Cai, Haoran Wang, Yi Zhang

AI总结:

该研究针对电力电子变换器健康监测中未标记数据利用不足的问题,提出PE-TSFM模型,其准确率达87%,显著优于传统模型,推动领域向统一可扩展基础模型发展。

摘要:

当前用于电力电子变换器的人工智能(AI)方法,对实际运行中产生的大量未标记数据利用有限,这一未充分开发的资源对提升模型性能具有重要潜力。为解决这一缺口,本文提出PE-TSFM,一种专为电力模块健康监测定制的时间序列基础模型。该模型采用PatchTST架构结合自监督学习,在超过1.43亿个时间戳上进行训练,其中超过98%的训练数据为未标记数据,直接采集自运行中的牵引变换器。在退化分类任务中,PE-TSFM的准确率达87%,显著超越5种传统时间序列AI模型,这些模型的性能介于27%至45%之间。这些结果表明该模型具备强大的少样本学习能力和跨不同运行条件的泛化能力。本研究推动电力电子领域从专注于特定任务的模型向统一、可扩展的基础模型转变,这类模型可从大规模未标记运行数据中挖掘价值。

英文摘要:

Current artificial intelligence (AI) methods for power electronics converters make limited use of the abundant unlabeled data generated during real-world operation. This underexplored resource holds significant potential for advancing model performance. To address this gap, this paper presents PE-TSFM, a time-series foundation model tailored for power module health monitoring. The model adopts the PatchTST architecture combined with self-supervised learning and is trained on more than 143 million timestamps, where over 98\% of this training data is unlabeled and collected directly from an operating traction converter. In degradation classification tasks, PE-TSFM achieves an accuracy of 87\%, considerably surpassing five conventional time-series AI models whose performance ranges from 27\% to 45\%. These results demonstrate the model's strong few-shot learning capability and generalization across different operating conditions. This work promotes a shift from narrowly focused models in power electronics toward unified, scalable foundation models that can unlock value from large-scale unlabeled operational data.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑