Application of adaptive network-based fuzzy inference system in short term load forecasting

A number of computing models based on adaptive network- based fuzzy inference system (ANFIS) are proposed in this paper to forecast peak load demands of an electric power utility. The models are capable of forecasting the peak demands both week-days and weekend-days i.e. Sundays and holidays as well. At the same time the models possess adaptability to the changing data pattern which may occur in case the load demand pattern changes or the weather parameters change. The paper involves forecasting models with zero order and first order Sugeno model of ANFIS with various types of membership functions and optimization method combinations. The proposed models are validated using load demand data of a real power utility to forecast its peak demand.