An adaptive neuro-fuzzy inference system-based approach for daily load curve prediction

Abstract

Forecasting future electricity demand is one of the most important areas in electrical engineering, due to its vital role for secure and profitable operations in power systems. In recent years, the advent of new concepts and technologies such as deregulation of electricity market, smart grids, electric cars and renewable energy integration have introduced great challenges for power system managers and consequently, the field of electric load forecasting becomes more and more important. This paper describes the use of an adaptive neuro-fuzzy inference system approach for daily load curve prediction. The methodology we propose uses univariate modeling to recognize the daily and weekly patterns of the electric load time series as a basis for the forecast. Results from real-world case study based on the electricity demand data in France are presented in order to illustrate the proficiency of the proposed approach. With an average mean absolute percentage error of 2.087%, the effectiveness of the proposed model is clearly revealed.