Neural Network Based Power Meter Wiring Fault Recognition of Smart Grids Under Abnormal Reactive Power Compensation Scenarios

Zheng, Huizhe; Lin, Zhongshuo; Lin, Huan; Huang, Chaokai; Huang, Xiaoqi; Ji, Suna; Zhang, Xiaoshun

doi:10.3390/en18030545

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Neural Network Based Power Meter Wiring Fault Recognition of Smart Grids Under Abnormal Reactive Power Compensation Scenarios

by

Huizhe Zheng

¹,

Zhongshuo Lin

¹,

Huan Lin

¹,

Chaokai Huang

¹,

Xiaoqi Huang

¹,

Suna Ji

¹ and

Xiaoshun Zhang

^2,*

¹

Shantou Power Supply Bureau of Guangdong Power Grid Co., Ltd., Shantou 515041, China

²

Foshan Graduate School of Innovation, Northeastern University, Foshan 528311, China

^*

Author to whom correspondence should be addressed.

Energies 2025, 18(3), 545; https://doi.org/10.3390/en18030545

Submission received: 25 December 2024 / Revised: 21 January 2025 / Accepted: 22 January 2025 / Published: 24 January 2025

(This article belongs to the Topic Intelligent, Flexible, and Effective Operation of Smart Grids with Novel Energy Technologies and Equipment)

Download Versions Notes

Abstract

This paper explores the challenges of detecting wiring anomalies in three-phase, four-wire energy metering devices, especially when large amounts of reactive power compensation are involved. Traditional methods, such as the hexagon phasor diagram technique, perform well under standard loads, but struggle to adapt to new situations, such as over- or under-compensation. To overcome these limitations, this paper proposes a hybrid approach that combines mechanism-based knowledge with data-driven technologies, including backpropagation neural networks (BPNNs). This method improves the accuracy and efficiency of anomaly detection and can better adapt to a dynamic power environment. The result is improved universality of anomaly detection, which helps to achieve safer, more accurate, and more efficient smart grid operation in complex situations.

Keywords: smart grids; abnormal wiring; reactive power compensation; network models; data-driven anomaly detection

Share and Cite

MDPI and ACS Style

Zheng, H.; Lin, Z.; Lin, H.; Huang, C.; Huang, X.; Ji, S.; Zhang, X. Neural Network Based Power Meter Wiring Fault Recognition of Smart Grids Under Abnormal Reactive Power Compensation Scenarios. Energies 2025, 18, 545. https://doi.org/10.3390/en18030545

AMA Style

Zheng H, Lin Z, Lin H, Huang C, Huang X, Ji S, Zhang X. Neural Network Based Power Meter Wiring Fault Recognition of Smart Grids Under Abnormal Reactive Power Compensation Scenarios. Energies. 2025; 18(3):545. https://doi.org/10.3390/en18030545

Chicago/Turabian Style

Zheng, Huizhe, Zhongshuo Lin, Huan Lin, Chaokai Huang, Xiaoqi Huang, Suna Ji, and Xiaoshun Zhang. 2025. "Neural Network Based Power Meter Wiring Fault Recognition of Smart Grids Under Abnormal Reactive Power Compensation Scenarios" Energies 18, no. 3: 545. https://doi.org/10.3390/en18030545

APA Style

Zheng, H., Lin, Z., Lin, H., Huang, C., Huang, X., Ji, S., & Zhang, X. (2025). Neural Network Based Power Meter Wiring Fault Recognition of Smart Grids Under Abnormal Reactive Power Compensation Scenarios. Energies, 18(3), 545. https://doi.org/10.3390/en18030545

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Neural Network Based Power Meter Wiring Fault Recognition of Smart Grids Under Abnormal Reactive Power Compensation Scenarios

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI