An Intelligent System for Load Forecasting on a Distribution Network

Main Article Content

Saheed Gbadamosi
Ojo Olukayode Adedayo


electric load, load forecasting, Adaptive Neuro-Fuzzy Inference system, power system


This paper presents an Adaptive Neuro-Fuzzy Inference system (ANFIS) technique for medium-term load forecasting in a distribution network. This technique is an integrated system consisting of fuzzy logic systems and Artificial Neural network (ANN). The inputs to the system include days of the week, temperature, time, current and previous hourly load on the distribution network. The data collection is within the period of two years. The formulation and mapping of the input data is done using fuzzy logic system and ANN are employed for generation of inference. The experimental results show the average load pre-diction accuracy of 87.23% and regression coefficient of 0.873. The analysis of the proposed ANFIS for load forecast is effective in planning, managing and organizing the electric load forecast with accurate prediction.

Abstract 0 |