Current Volume 9
Detecting and minimizing Energy (Electricity) theft in smart grid network using artificial intelligent (AI) based scheme is work aimed to proffer solution on reducing electricity theft without the traditional or conventional method.Energy theft is a treat to the reliability and sustainability of smart grid network. A Dataset of energy consumption by 5,567 households were collected from November 2011 to February 2014. Data wrangling and model training were done on Nvidia’s Tesla T4 GPU(40 cores, 300GB/s bandwidth) (colab.research.google.com, n.d.) and using packages that provide Python bindings around CUDA framework for parallel computations, other components used are Tensor Flow and scikit-learn for machine learning as well as Matplotlib for visualization. The dataset is standardized for each household to ensure a complete and uniform dataset. Generation of anomalous sample was done to train the system and these synthetic samples are critical for training supervised machine learning models. The ANN detect and minimize energy theft through prioritizing inspections, real-time alerts and tampering detection through changes in waveform signature due to tampering.
Smart Grid, ANN, Energy Theft, Cyber-Attack. Dataset, Machine Learning and Deep Learning.
IRE Journals:
Eze Ernest Chukwuemeka, Eleje Nelson Emeka "Detecting and Minimizing Energy theft in Smart Grid Network Using Artificial Neural Network (ANN)" Iconic Research And Engineering Journals Volume 9 Issue 12 2026 Page 195-204 https://doi.org/10.64388/IREV9I12-1718574
IEEE:
Eze Ernest Chukwuemeka, Eleje Nelson Emeka
"Detecting and Minimizing Energy theft in Smart Grid Network Using Artificial Neural Network (ANN)" Iconic Research And Engineering Journals, 9(12) https://doi.org/10.64388/IREV9I12-1718574