Research On Forecasting Model Of Electricity Sales In Baiyin City

Electricity sales is an important economic indicator for th e operation and management of Power Grid Enterprises,it is also the basic data of a series of index calculation such as average price of selling electricity,financial income,investment budget of power grid,etc..Electricity sales are affected by many complicated factors,it is hard to find a "one-size-fits-all" approach to forecasting,which is suitable for the load characteristics of different regions,scholars at home and abroad have paid much attention to it.At present,There are many researches on the forecasting of electricity sales,the commonly forecasting methods are Grey prediction method and time series method.However,these methods have some limitations in different areas,different distribution structures and different load characteristics.The purpose of this essay is to develop a method suitable for forecasting the monthly electricity sales in Baiyin to improve the accuracy of forecasting and meet the needs of enterprises.Combined with the characteristics of electric load in Baiyin,in this essay,the Baiyin Electricity load is classified according to the nature of electricity consumption,the implementation price and the national economy industry.On this basis,five mathematical models are constructed.By analyzing and comparing the emulat ion prediction results,the combination of grey prediction and artificial neural network is selected as the basic prediction model.The parameters of grey neural network prediction model are optimized by using differential evolution algorithm,the forecast model is modified continuously by taking meteorological factors as influencing factors.In order to improve the accuracy of the prediction model,this essay combines the characteristics of electric load in Baiyin.The k-means clustering algorithm is applied to the load clustering in Baiyin area,seven types of Power load are grouped into three types.Using the clustering electricity sales as the training data,a monthly electricity sales prediction model in Baiyin is established on the MATLAB platform.By comparing with the prediction results in the previous chapters,this differential grey neural network algorithm based on data clustering can further improve the accuracy of monthly electricity sales prediction.