Anomaly sound detection of high-voltage shunt reactors based on soft-constrained latent regularized adversarial learning

Authors

Zhihong WANG, State Grid Hunan Extra High Voltage Substation Company, Changsha 410004, China; Substation Intelligent Operation and Inspection Laboratory of State Grid Hunan Electric Power Co., Ltd., Changsha 410004, China
Fuqiang XIONG, State Grid Hunan Extra High Voltage Substation Company, Changsha 410004, China; Substation Intelligent Operation and Inspection Laboratory of State Grid Hunan Electric Power Co., Ltd., Changsha 410004, China
Jiawen ZUO, College of Energy and Power Engineering, Changsha University of Science & Technology, Changsha 410114, China
Mingzhu TANG, College of Energy and Power Engineering, Changsha University of Science & Technology, Changsha 410114, ChinaFollow
Jianjun ZHANG, State Grid Hunan Extra High Voltage Substation Company, Changsha 410004, China; Substation Intelligent Operation and Inspection Laboratory of State Grid Hunan Electric Power Co., Ltd., Changsha 410004, China
Xingyu TANG, State Grid Hunan Extra High Voltage Substation Company, Changsha 410004, China; Substation Intelligent Operation and Inspection Laboratory of State Grid Hunan Electric Power Co., Ltd., Changsha 410004, China

Keywords

acoustic anomaly detection; semi-supervised learning; adversarial learning; latent variable regularizedadversarial learning

Abstract

With the widespread application of high-voltage shunt reactors in power systems, abnormal phenomena arising during their operation have attracted increasing attention. In existing latent regularization adversarial anomaly detection (LRAAD) methods, the hyperparameter M imposes a hard constraint on the upper bound of the KL divergence of generator-produced spectrograms, which hampers the model's ability to effectively distinguish normal from abnormal data in the latent space, leading to training instability and degraded anomaly detection performance. To address this issue, this paper proposes a soft-constrained latent regularization adversarial anomaly detection (Soft-LRAAD) method. The proposed method introduces a soft constraint loss to replace the hard constraint loss, and enhances the discrimination capability in the latent space and the training stability by approximating the upper bound of the KL divergence using a smooth function. Experimental results demonstrate that the proposed method effectively improves the accuracy and robustness of anomaly detection for high-voltage shunt reactors, providing a superior solution for power equipment fault diagnosis.

DOI

10.19781/j.issn.1673-9140.2026.01.029

First Page

307

Last Page

318

Recommended Citation

WANG, Zhihong; XIONG, Fuqiang; ZUO, Jiawen; TANG, Mingzhu; ZHANG, Jianjun; and TANG, Xingyu (2026) "Anomaly sound detection of high-voltage shunt reactors based on soft-constrained latent regularized adversarial learning," Journal of Electric Power Science and Technology: Vol. 41: Iss. 1, Article 29.
DOI: 10.19781/j.issn.1673-9140.2026.01.029
Available at: https://jepst.researchcommons.org/journal/vol41/iss1/29