Short-Term Load Forecasting for Special Days Using Bayesian Neural Networks

Conventional artificial neural network (ANN) based short-term load forecasting techniques have limitations in their use on holidays. This is due to dissimilar load behaviors of holidays compared with those of ordinary weekdays during the year and to insufficiency of training patterns. The purpose of this paper is to propose a new short-term load forecasting method for special days in irregular load conditions. These days include public holidays and consecutive holidays. The proposed method uses a Bayesian neural network (BNN) to forecast the hourly loads of special days. For doing that, we used hybrid Monte Carlo method. This type of learning enables us to work with simpler architecture with respect to previous works. This method was tested with actual load data of special days for the years of 2003-2004. The test results showed very accurate forecasting with the average percentage relative error of 1.93%