Title: A detection method for electricity theft behaviour in low-voltage power stations: multi-source data fusion

Authors: Xiongfeng Ye; Zhiguo Zhou; Yizhi Cheng

Addresses: State Grid Zhejiang Electric Power Company, Quzhou Power Supply Company, Quzhou, 324000, China ' State Grid Zhejiang Electric Power Company, Quzhou Power Supply Company, Quzhou, 324000, China ' State Grid Zhejiang Electric Power Company, Quzhou Power Supply Company, Quzhou, 324000, China

Abstract: In order to improve the accuracy of electricity theft detection in low-voltage substations, a method of electricity theft detection based on multi-source data fusion is proposed and designed. Firstly, a structure design of power load data collection is designed to obtain multi-source power stealing behaviour data in low-voltage power station area. Then, K-means algorithm is used to extract the features of multi-source power theft data, and feature superposition method is used to complete the feature fusion of multi-source power theft data in low-voltage power station area. Finally, the integrated characteristic vector of electricity theft behaviour is used as input to design the electricity theft detection based on improved support vector machine (SVM) algorithm. The experimental results show that the method proposed in this paper can greatly improve the detection accuracy, and is better than the comparison method.

Keywords: multi-source data fusion; low voltage power substation area; stealing electricity; behavioural detection.

DOI: 10.1504/IJETP.2025.144303

International Journal of Energy Technology and Policy, 2025 Vol.20 No.1/2, pp.36 - 50

Received: 23 Jan 2024
Accepted: 08 Jul 2024
Published online: 05 Feb 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article