PE-TSFM: Self-Supervised Time-Series Learning for Generalizable Power Converter Health Monitoring Under Unseen Conditions
- 1 October 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (10) , 15396-15408
Abstract
Data-driven health monitoring of power converters remains limited by poor generalization to unseen operating conditions. This work addresses this out-of-distribution (OOD) challenge by building a domain-specific time-series foundation model (PE-TSFM) that learns representations directly from large-scale unlabeled converter data. Unlike generic TSFMs trained on broad time-series datasets, the proposed PE-TSFM is pretrained entirely on domain data, enabling it to learn the physical relationships unique to power electronics. To further tailor the model to this domain, we introduce a dual-attention mechanism that captures both temporal patterns and inter-channel dependencies. While generic TSFMs primarily model temporal dependencies, the added channel attention (CA) captures inter-sensor physical relationships essential for converter degradation analysis. A dataset containing 141 million unlabeled timestamps from an operating power converter is used for pretraining. Experiments show that PE-TSFMs achieve 76%–92% accuracy under unseen operating conditions. In contrast, generic TSFMs achieve around 60% and conventional time-series models achieve around 40% accuracy. This result confirms the strong OOD generalization of the proposed PE-TSFM. Ablation studies further verify that the introduced CA mechanism significantly improves model performance. In addition, we conduct detailed studies on model scalability, hyperparameter sensitivity, and interpretability to provide a comprehensive understanding of the proposed approach.Keywords
All Related Versions
Funding Information
- PolyU Seed Fund (P0056659)
- Shenzhen Science and Technology Program (JCYJ20250604184248063)
This publication has 27 references indexed in Scilit:
- Review on Condition Monitoring Methods of IGBT Power Device2025
- Predicting DC-Link Capacitor Current Ripple in AC-DC Rectifier Circuits Using Fine-Tuned Large Language Models2024
- PE-GPT: A Physics-Informed Interactive Large Language Model for Power Converter Modulation Design2024
- A Deep Learning Network based Robust Fault Diagnosis Method for IGBT Open Circuit2024
- Improving position encoding of transformers for multivariate time series classificationData Mining and Knowledge Discovery, 2023
- Impact of Loss Model Selection on Power Semiconductor Lifetime Prediction in Electric Vehicles2022
- Fault Diagnosis of Inverter Current Sensor Using Artificial Neural Network Considering Out-of-distribution2021
- InceptionTime: Finding AlexNet for time series classificationData Mining and Knowledge Discovery, 2020
- Reliability Improvement for a High-Power IGBT in Wind Energy ApplicationsIEEE Transactions on Industrial Electronics, 2018
- Time series classification from scratch with deep neural networks: A strong baseline2017