Artificial Neural Networks and Adaptive Neuro-Fuzzy Inference Systems Approaches to Forecast the Electricity Data for Load Demand, an Analysis of Dinar District Case

Short-term load forecasting is an important issue for the electric power system in efficiently managing the network and reducing operating costs. In addition, with the recent improvements in distributed generation and storage systems, this has become even more important. Access to the high-resolution dataset derived from smart counters allows new forecasting strategies to evolve to match distributed load on the demand side. In this study, short-term load forecasting (STF) of a small region was performed using Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) methods. For load forecasting, the electricity consumption and temperature data for the year 2017 were used as input to the network and next hour demand was predicted. The smallest forecasting error is investigated with the Mean Absolute Percent Error (MAPE) and Root Mean Square Error (RMSE) performance criteria. It showed that RMSE is better than ANFIS with 616.2753 and MAPE 8.8688 prediction error.