Short-term load/price forecasting in deregulated electric environment using ELMAN neural network

Abstract

Load forecasting plays a significant role in power system planning. In today's scenario of deregulated electricity market as existing in New South Wales (NSW) Australia, an extremely accurate load/ price forecasting model is required because of several economic and operational advantages. It helps in dealing with the problems of economic load dispatch, unit commitment, protection, etc. Research shows that most of the classical methods are incapable to forecast the load/ price with highest possible precision, as per the expectation of deregulated and complex electricity markets. In this paper, Artificial Neural Network (ANN)-based Short Term Load Forecasting (STLF) model, i.e., ELMAN Neural Network (ELMNN) is developed and tested on NSW Australia data. The performance of the ELMNN-based model is compared with Feed Forward Neural Network (FFNN) and Radial Basis Function Neural Network (RBFNN). It is observed that ELMNN-based load forecasting model produces superior results over other ANN-based models.