Citation

Wong, Hwee Ling and Tan, Chia Kwang and Tan, Wooi Nee and Gan, Ming Tao and Yip, Sook Chin and Abu Bakar, Ab Halim (2025) Energy Theft Identification: A State Estimation Model for Partial Meter Bypass. IEEE Access, 13. pp. 121412-121431. ISSN 2169-3536

Text Energy Theft Identification_ A State Estimation Model for Partial Meter Bypass.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

Energy theft due to illegally bypassing the meter can happen in traditional and smart grids. It is still one of the most frequent types of non-technical losses (NTL) due to the simplicity of connecting the illegal load directly to the distribution cable. Fraudulent consumers tend to report partial loads to avoid suspicions. Connection of illicit loads consisting of one or more electrical appliances is addressed at length in this paper. A novel method is proposed to locate and estimate the amount of theft using a 2-step Kalman filtering process with bias filter for optimization. The schema can estimate shifting cable impedances due to the dynamic operating conditions, irrespective of theft presence. On average, energy theft occurrences could be identified up to 99% for thefts using single appliance and 92% for thefts originating from multiple appliances. The coverages also vary due to the number of premises with energy theft, the load pattern and the type of bias filter applied. Results affirmed that incorporating an optimization stage through the inclusion of the bias filter eliminates poor estimation yielding low mean absolute errors for apparent power and power factor of 0.3 volt-ampere and 0.02, respectively. The accuracy and precision of classifying theft-points per premise is close to 100% on average. A schema is also presented to reduce false identification of honest consumer as dishonest from 22.6% to 0.5%. Additionally, the proposed framework demonstrates robustness against different cable lengths, changes in temperature, cable core material, cable core dimension and cable deterioration.

Item Type:	Article
Uncontrolled Keywords:	Power distribution network
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3001-3521 Distribution or transmission of electric power
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	29 Jul 2025 02:21
Last Modified:	31 Jul 2025 08:56
URII:	http://shdl.mmu.edu.my/id/eprint/14338

Downloads

Edit (login required)