Short Term Load Forecasting Of Power Grid In A Certain Area Of Shaanxi Based On MEEMD And Neural Network

Electric power industry is the basic industry of the national economy and the basic industry related to the national economy and the people’s livelihood.Power load forecasting is the basis of economic operation of power system and plays a very important role in power system planning and operation.At present,with the increasing demand for electricity in various regions,accurate short-term load forecasting has become more and more important.Power Grid in a Certain Area of Shaanxi,is an important energy guarantee to support the national "the Belt and Road" strategy and serve the economic development of a new area.In recent years,Power Grid in a Certain Area of Shaanxi have actively implemented the action plan of "carbon peak and carbon neutrality" of the State Grid Corporation of China,and strive to be a leader,promoter and pioneer in the transformation and development of clean and low-carbon energy.The short-term load forecasting in this area depends on the experience of operation and maintenance personnel.The current load value is estimated through the load level of similar days,and the error of prediction results is large.Short term load forecasting is of great significance to the safe,stable and economic operation of power grid in this region.How to improve the quality of short-term load forecasting has always been a difficult problem for power workers in this region.Taking the power load data of a regional power grid in Shaanxi as the research object,combined with the short-term load forecasting process,this thesis mainly studies the data processing,artificial neural network,load model construction and forecasting analysis,and proposes a short-term load forecasting model based on modified ensemble empirical mode decomposition(MEEMD)and neural network.The main work is as follows:By reading a lot of relevant literature at home and abroad and analyzing the current situation of load forecasting at home and abroad,this thesis analyzes the specific load characteristics of a regional power grid in Shaanxi on the preprocessing of short-term load data.Deeply study the principles of empirical mode decomposition(EMD),ensemble empirical mode decomposition(EEMD)and MEEMD,decompose and analyze the load data of the regional power grid respectively.Through the comparison of the results,it is found that MEEMD method not only solves the problem of mode aliasing,but also has better processing effect on the high-frequency items in the original load data.It is helpful to improve the accuracy of load forecasting.The basic principles and prediction algorithm flow of BP neural network,Elman neural network and wavelet neural network are compared and studied.The mean absolute percentage error(MAPE)is used as the evaluation index.The simulation results of the three neural networks are compared and analyzed.It is found that the error of BP neural network is the smallest and the data fitting is the best.EMD,EEMD and MEEMD are decomposed to obtain components.These components are used to reconstruct new signals.Combined with BP neural network,EMD-BP model,EEMD-BP model and MEEMD-BP model are established respectively for prediction.Through the analysis and comparison of the prediction results,it is found that MEEMD-BP model has the best effect on the short-term load prediction of a regional power grid in Shaanxi Province,and the results are in line with the reality,which has guiding significance for the operation of the regional power grid.