Power Load Forecasting Based On Wavelet-grey Prediction Model

Power load forecasting is the directed basis of the power system investment and planning.It’s all depending on the results of power load forecasting that just like the construction necessity,the construction scale,the mode of system access,and the construction sequence arrangement of the new generator’s addition and grid expansion project.With the development of various power system technology,the application of load forecasting is more and more extensive,and the demand of the precision to prediction result is higher than before..Especially under the strict investment control of the network planning,as well as the power demand side management is increasingly strict,the load forecasting accuracy becomes the absolute prerequisite for other work.The main content of this paper is how to improve the accuracy of load forecasting with no increasing the amount of computation,and how to suitable for the load forecasting both in short-term and medium and long-term.Firstly,this paper simply introduces the characteristics of the power load and its forecasting,their influence factors as well,and briefly summarizes the classic methods of power load forecasting and the main modern methods.On the basis of summarizing the wavelet analysis theory and grey system theory,a simple analysis of the application of the two theories is carried out.Secondly,based on the previous analysis,this paper puts forward the modeling idea of wavelet grey model.Taking the advantages of wavelet transform and the easy implementation of the algorithm as the main points,this paper completed deriving of the wavelet transform,then get the wavelet decomposition function.Based on the data generated significance and specific algorithm,GM(1,1)grey model is iterative formula,got the modeling sequence,and test the feasibility of the gray model,then got the formation of the basic conditions of load forecasting.Finally,in this paper,the power load forecasting of Licheng district is used as an example to verify and analyze the wavelet grey prediction model.By introducing the electricity demand and the status quo of Licheng area grid,grasp the basic characteristics of the power load in Licheng district.By using "11th Five-Year" and "12th Five-Year" planning report data of Licheng District as the research object,carried out simulation test for prediction algorithm using MATLAB tools.The prediction error is obtained by simulation results analysis and the results to real comparison.The experimental results show that the prediction error can be stabilized under 5%,and the results are not affected by the original data when the quality of the original sequence is equal to that of the nodes near the original sequence.