Deep Domain Adaptation With Channel Attention for Open-Circuit Fault Diagnosis of Charging Piles Operating Under Dynamic Working Conditions

View at publisher Summarise with VIRAT
Authors: Quanxue Guan, Xue Hu, Yu Han, Jiabei Hu, Xiaojun Tan, Qin Wang
Cited by: 0
References: 47

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Open-circuit faults (OCFs) in charging converter modules can lead to performance degradation and system failures. Traditional deep-learning-based fault diagnosis methods typically require large amounts of labeled data, which is costly and time-consuming for two-stage power converters in practice. To this end, this paper proposes a deep domain-adaptation network with a channel attention mechanism to diagnose the OCFs of charging piles operating under dynamic operating conditions. Multi-channel features are first extracted from the three-phase input currents, the output voltage, and the resonant capacitor voltage. To overcome the lack of labeled data in varying working conditions, multi-kernel maximum mean discrepancy (MK-MMD) is employed to align feature distributions between the labeled source domains and the unlabeled target domains. This enables the proposed classifier model to generalize across different operating conditions in the target domain. Experimental results demonstrate that the proposed method achieves over 99% diagnostic accuracy in the source domain and above 90% in the target domain within 11 ms, outperforming the existing baseline methods. It effectively identifies OCFs in both the Vienna rectifier and LLC converter stages. The proposed method also exhibits strong robustness to noise, load changes, and unbalanced grid conditions. By eliminating the need for extensive labeled datasets and adapting to unknown operating states, the proposed approach offers a practical and efficient solution for real-time fault diagnosis in charging piles, including multiple simultaneous faults that undermine the reliability of charging systems.

Fields of Study & Subjects

Advanced DC-DC Converters Multilevel Inverters and Converters Electrical Fault Detection and Protection

Publication details

Journal / source
IEEE Transactions on Power Electronics
Volume
41
Issue
9
Pages
16115–16128
Publication date
23 March 2026
Language
EN