Artificial Intelligence Technique based Power Theft Monitoring and Detection in a Power Distribution System

This paper initiated an efficient method for monitoring and identifying the power theft in the distribution framework. The efficient technique comprises of Discrete Wavelet Transform - Adaptive Neuro-Fuzzy Inference System (DWT-ANFIS) and Gravitational Search Algorithm (GSA). At first, the power signal is examined with typical condition at the specific time moment. From that point forward, to explore any illicit use of power is happened or not in the power dispersion framework with the assistance of initiated system. The DWT is used for examining signals and produce the dataset. The optimal dataset is chosen on the basis of the GSA optimization algorithm then the optimal dataset is given to the ANFIS for preparing. After that the preparation and the testing procedure would be handled. The initiated strategy is used to discover the theft load associated position. The initiated efficient procedure was actualized for the results with various input conditions using Simulink and MATLAB packages. Finally, the results were compared with the similar implementations such as DWT-ANN, DWT-ANN with GA and DWT-ANFIS strategy.