Electrical load forecasting using adaptive neuro-fuzzy inference system

Abstract

Electrical load forecasting is well-known as one of the most important challenges in the management of electrical supply and demand and has been studied extensively. Electrical load forecasting is conducted at different time scales from short-term, medium-term and long-term load forecasting. Adaptive neuro-fuzzy inference system is a model that combines fuzzy logic and adaptive neuro system and is implemented in time-series forecasting. First, ANFIS structure is decided using subtractive categorization; next, ANFIS premise and consistent parameters are identified using hybrid algorithm; finally, some factors affecting future daily electrical load such as weather and population become inputs of ANFIS to forecast daily electrical load on the following day. The membership function used is Gbell membership function. The forecasting result shows that the forecasting model is considered valid with an RMSE score of 0,0298.