Forecasting several-hours-ahead electricity demand using neural network

This paper presents a practical method for short-term load forecasting considering the temperature as climate factor. The method is based on artificial neural network (ANN) combined similar days approach, which achieved a good performance in the very special region. Performance of the proposed methodology is verified with simulations of actual data pertaining to Okinawa Electric Power Co. in Japan. Forecasted load is obtained from ANN, which is the corrected output of similar days data. Load curve is forecasted by using information of the days being similar to weather condition of the forecast day. An Euclidean norm with weighted factors is used to evaluate the similarity between a forecast day and searched previous days. Special attention was paid to model accurately in different seasons, i.e., summer, winter, spring, and autumn. Moreover, the forecaster is robust, easy to use, and produces accurate results in the case of rapid weather changes.