Electric Short-Term Load Forecasting Using Artificial Neural Networks and Fuzzy Expert System

摘要

Utilize the Radical Basis Function (RBF) network and Ordinary Least Square (OLS) to define the RBF function centers. The primary load is forecasted by the trained RBF networks, and then utilize the fuzzy expert systems to correct the primary load premeditating the possibility of load variation due to changes in temperature and the load style of holiday. Some of day types are differentiated with five types in this paper. Test results shows that the hybrid model can forecast load with a higher accuracy and a faster speed. Supporting a hybrid model for short-term load forecasting which integrates artificial neural networks (ANN) and fuzzy expert system.