A neural network short-term load forecasting considering human comfort index and its accumulative effect

Abstract

Short-term load forecasting is one of the most important fields of electricity demand research. Many traditional models and artificial intelligence techniques have been evaluated and tested in this task, and the Artificial Neural Network (ANN) is received much attention. In this paper a development of the artificial neural network based short-term load forecasting model considering the impact of human comfort index and its accumulative effect was proposed. The ANN structure and the training data set selection are described in the paper, and holiday load forecasting correction are adapted in this model. The implementation and results in a southeast city of China indicate that the load forecasting model developed carries out accurate forecasts.