UHV Converter Transformer Equipment Fault Diagnosis via Cross-Modal Transformer with DGA and Infrared Image Fusion.
Saved in:
| Title: | UHV Converter Transformer Equipment Fault Diagnosis via Cross-Modal Transformer with DGA and Infrared Image Fusion. |
|---|---|
| Authors: | Yang, Xin1,2 (AUTHOR) yangxin0993@163.com, Liao, Wenlong1,2 (AUTHOR), Liu, Rui1,2 (AUTHOR), Fan, Songhai1,2 (AUTHOR), Feng, Yun1,2 (AUTHOR), Zhang, Yu1,2 (AUTHOR), Yang, Yueping1,2 (AUTHOR), Wang, Zhenyu1,2 (AUTHOR), Mu, Zhou1,2 (AUTHOR) |
| Source: | Energies (19961073). Jun2026, Vol. 19 Issue 12, p2747. 17p. |
| Subject Terms: | *Multisensor data fusion, *Transformer models, *Electric transformers, *Gas analysis, *Thermography, *Fault diagnosis |
| Abstract: | Ultra-high-voltage (UHV) converter transformer equipment is critical for UHVDC transmission systems. This paper proposes a Cross-modal Transformer framework for fault diagnosis by fusing dissolved gas analysis (DGA) and infrared (IR) thermography data. The framework encodes DGA measurements into temporal tokens and processes IR images through a ResNet-18 backbone to generate spatial tokens. A Cross-modal Transformer module enables deep semantic interaction via bidirectional cross-attention, allowing DGA tokens to attend to relevant IR regions and vice versa. A modality-gating mechanism adaptively reweights the two modalities under measurement degradation, including partial and fully missing-modality scenarios. The novelty lies in adapting these components into a leakage-controlled DGA-IR diagnostic framework for UHV converter transformers, with explicit interaction between gas-evolution tokens and spatial thermal tokens. Evaluation is performed under a leakage-controlled grouped chronological split that isolates equipment units, converter stations, and fault episodes across train, validation, and test partitions. Labels are drawn exclusively from maintenance inspection and operational records, independent of the IEC 60599 ratio features seen by the model. Under this protocol, the proposed framework consistently improves accuracy and macro-F1 over encoder-matched simple-fusion baselines (Transformer-DGA + ResNet-18 with concatenation, late fusion, and gated averaging). Additional missing-modality, noise, and ablation experiments indicate that the gains come from bidirectional cross-attention and adaptive gating rather than from stronger unimodal encoders alone. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Ultra-high-voltage (UHV) converter transformer equipment is critical for UHVDC transmission systems. This paper proposes a Cross-modal Transformer framework for fault diagnosis by fusing dissolved gas analysis (DGA) and infrared (IR) thermography data. The framework encodes DGA measurements into temporal tokens and processes IR images through a ResNet-18 backbone to generate spatial tokens. A Cross-modal Transformer module enables deep semantic interaction via bidirectional cross-attention, allowing DGA tokens to attend to relevant IR regions and vice versa. A modality-gating mechanism adaptively reweights the two modalities under measurement degradation, including partial and fully missing-modality scenarios. The novelty lies in adapting these components into a leakage-controlled DGA-IR diagnostic framework for UHV converter transformers, with explicit interaction between gas-evolution tokens and spatial thermal tokens. Evaluation is performed under a leakage-controlled grouped chronological split that isolates equipment units, converter stations, and fault episodes across train, validation, and test partitions. Labels are drawn exclusively from maintenance inspection and operational records, independent of the IEC 60599 ratio features seen by the model. Under this protocol, the proposed framework consistently improves accuracy and macro-F1 over encoder-matched simple-fusion baselines (Transformer-DGA + ResNet-18 with concatenation, late fusion, and gated averaging). Additional missing-modality, noise, and ablation experiments indicate that the gains come from bidirectional cross-attention and adaptive gating rather than from stronger unimodal encoders alone. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 19961073 |
| DOI: | 10.3390/en19122747 |
