Research On Artificial Neural Networks Used In Short Term Load Forecasting

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.This paper firstly gives a summary for present method of load forecasting; Secondly, it has made a in-depth research into ANN modeling problem, to give more applicable modeling method and principle; After studying plenty of documents and analyzing various important factors of electric power load, a three-tier BP neural networks has been constructed. To establish BP network model, implicit layers number identification, hidden nodes determination, the times and accuracy of training, learning rate option, the initial weights, the choice of training samples and normalized treatment, and other related issues are under more in-depth qualitative and quantitative analysis. After making compared experiment through examples, useful conclusions are drawn.With the analysis on BP deficiencies, on the basis of the adoption of improved BP neural network algorithm, a short-term load forecasting model is established. The improved BP network algorithm is applied in load forecasting. Consequently the different algorithms' different predicting results are compared. Finally, the methods to establish the neural network model taken into account of the meteorological parameters are discussed. Experiment research is put forward, thus a verification and analysis is obtained.