Application Of Grey Theory And Artificial Neural Network In Medium And Long Term Load Forecasting Of Power System

Power load forecasting is one of the most important tasks for power system workers.And it also plays an important role in regional power system.Accurate load forecasting can effectively predict the operation state of the power generation unit.On the basis of achieving safe and stable operation of the power grid,the waste of electric energy can be reduced and the overall cost and consumption can be effectively reduced.This paper studies several aspects of this problem which are shown as follows:Firstly,several categories of power load and the characteristics of load forecasting are introduced.The basic concepts of grey theory and artificial neural network are analyzed.Several factors that affect medium and long term load forecasting are put forward,and the most important factors are found out by the grey correlation method.It lays the foundation for the following analysis of grey prediction model and artificial neural network.Secondly,the GM(1,1)model is established according to the grey theory.On this basis,the GM(1,n)model with multiple variables can be also established.Then the medium and long term load forecasting method based on the grey prediction model is obtained.The method of some parameter assignment in the grey prediction model is improved.The improved grey prediction model is rebuilt.And this paper obtains the medium and long term load forecasting method based on improved grey prediction model.Thirdly,the medium and long term load forecasting model is established by BP neural network algorithm.So the medium and long term load forecasting method based on artificial neural network can be obtained.Based on the weight theory,the load forecasting results under the grey theory are combined with the load forecasting results under the artificial neural network method.A combined optimal medium and long term load forecasting method based on two prediction methods is proposed.Finally,taking the load data of a city in 2005 for example,this paper used four methods which are grey model prediction method,improved grey model method,artificial neural network method and combination forecasting method based on weight to predict the medium and long term load of the city respectively.Comparing the simulation results,the combined forecasting method based on the improved grey model method and artificial neural network method is most suitable for medium and long term load forecasting.