SHORT-TERM LOAD FORECASTING USING A CBR-ANN MODEL

摘要

This paper presents an approach based on rough set. The approach improves case-based reasoning to reduce the initial information and to find similar historical daily information. The result of case-based reasoning will be put into an artificial neural network to process and then get the forecasting result. The paper provides a new method to selecting a relevant feature subset and feature weights. The experiment results on Hangzhou area show that the proposed method is feasible and promising for short-term load forecasting.