Intelligent Fault Diagnosis of Electrical Secondary Circuits Using Multi-Feature Fusion Deep Learning
DOI:
https://doi.org/10.54097/tmz5bj36Keywords:
Electrical secondary circuits, Multi-feature fusion, Deep learning, Fault diagnosisAbstract
To address the complex fault characteristics of electrical secondary circuits and the limited diagnostic capability of single-source information, this study proposes an intelligent fault diagnosis method based on multi-feature fusion deep learning. Current and voltage waveforms, statistical parameters, protection operations, circuit-breaker positions, and alarm states are jointly modeled. A multi-scale one-dimensional convolutional neural network extracts local transient patterns, a Transformer encoder captures long-range temporal dependencies, and an attention mechanism adaptively fuses the resulting representations. On an eight-class dataset containing 9600 samples, the proposed model achieves an Accuracy of 97.43%, an F1-score of 97.35%, and an AUC of 99.41%, outperforming representative machine learning and deep learning baselines. Ablation and noise tests further demonstrate the contributions of the feature branches and the robustness of the fusion strategy. The method provides a practical basis for intelligent maintenance and online condition monitoring of electrical secondary circuits.
Downloads
References
[1] Chen, F., Luo, J., Wang, R., Zhang, H., & Liu, X. (2025). Research on fault diagnosis method of substation relay protection secondary circuit based on improved D-S evidence theory. Measurement, 256, Article 118232. https://doi.org/10.1016/j.measurement.2025.118232
[2] Mei, Y., Ni, S., & Zhang, H. (2024). Fault diagnosis of intelligent substation relay protection system based on Transformer architecture and migration training model. Energy Informatics, 7, Article 120. https://doi.org/10.1186/s42162-024-00423-4
[3] Shao, N., Chen, Q., Yu, C., Li, J., & Wang, F. (2024). Fault tracing method for relay protection system–circuit breaker based on improved Random Forest. Electronics, 13(3), Article 582. https://doi.org/10.3390/electronics13030582
[4] Wang, J., Jing, S., Yao, Y., Li, H., Zhang, W., & Chen, G. (2024). A state evaluation and fault diagnosis strategy for substation relay protection system integrating multiple intelligent algorithms. The Journal of Engineering, 2024(12), Article e70013. https://doi.org/10.1049/ije2.70013
[5] Li, Q., Zhang, Y., Zhang, Z., Yang, J., & Wang, H. (2024). Operation monitoring platform of relay protection equipment at distribution network side under the background of new power system. Energy Informatics, 7, Article 145. https://doi.org/10.1186/s42162-024-00421-6
[6] Han, D., Zhang, Y., Yu, Y., Li, W., & Wang, Z. (2024). Multi-source heterogeneous information fusion fault diagnosis method based on deep neural networks under limited datasets. Applied Soft Computing, 154, Article 111371. https://doi.org/10.1016/j.asoc.2024.111371
[7] Song, Y., Du, J., Li, S., Wang, R., & Chen, X. (2023). Multi-scale feature fusion convolutional neural networks for fault diagnosis of electromechanical actuator. Applied Sciences, 13(15), Article 8689. https://doi.org/10.3390/app13158689
[8] Qian, L., Li, B., & Chen, L. (2022). CNN-based feature fusion motor fault diagnosis. Electronics, 11(17), Article 2746. https://doi.org/10.3390/electronics11172746
[9] Li, H., Huang, J., Gao, M., Yang, S., & Zhao, L. (2022). Multi-view information fusion fault diagnosis method based on attention mechanism and convolutional neural network. Applied Sciences, 12(22), Article 11410. https://doi.org/10.3390/app122211410
[10] Kuang, H., Yi, P., Luo, Y., Chen, W., & Liu, Z. (2022). Research on fault diagnosis method of secondary equipment in intelligent substation. Journal of Physics: Conference Series, 2260, Article 012020. https://doi.org/10.1088/1742-6596/2260/1/012020
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Computing and Electronic Information Management
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
