Predicting Load Current of Electrical Distribution System Using Artificial Intelligence Techniques

Abstract

Artificial Intelligence (AI) is currently used by people working in different fields of science, also the need for Smart Grids in electrical networks requires the use of intelligent technologies. Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) methods are the creation of AI. For the first time, AI methods will be used to determine the Load Current (LC) of distribution network’s buses. The objective is to test and compare ANN and ANFIS methods in order to predict LC in every bus of the I3E 33 bus system and to evaluate the possibility of using them in systems where iterative methods do not converge and also to obtain the predicted values in real time. For this comparison, two discrepancies indicators have been used to test the reliability of the two methods. Generally, the obtained results from the study have been good enough to recommend the employment of AI techniques to solve difficult problems that lack data or have large and complicated systems. In addition, ANFIS method is more precise and requires less time for training and predicting compared with ANN method, indicating that combining two or more intelligent methods can be beneficial.