Short-Term Load Forecasting Based On Artificial Neural Networks

Short-Term Load Forecasting(STLF) is one of the most important contents of running and dispatching power system. It is a very important aspect of power system to ensure operating safely economically and achieve scientific management in the power system. And it is one part of energy management system as well as a necessary content of the electricity marketplace operation management.There are many methods for STLF nowadays,but there hasn't a method which can obtain the satisfied forecasting results on any occasion for the reason of lack of accuracy. In this paper,we conducted a new model based on single exponent smoothness model and artificial neural networks model according to the law of power load variance. Single exponent smoothness model fully embodies the characters of trend and cycle of the continuous varying power load, but it can not think about the effect of the weather elements carefully. The powerful learning ability and nonlinear reflecting functions of the artificial neural networks can combine the weather factors, typical days and historical load data, but it is not good enough at the embodying continuous variance. Based on the analysis of power load in Zhengzhou area, we obtained the law of power load variance and the affecting factors. Also it suggested that combined forecasting method can acquire good forecasting results. we have improved the properties in the application. It can make networks converge quickly into the global minimum at the large probability. In conclusion, focusing on the real conditions of power load in Zhengzhou, we can offer short-term power load forecasting method which improve forecasting accuracy to be suitable for this area.