Analysis And Forecast Of Electricity Consumption In The Whole Society Of Jilin Province Based On Grey System And Cluster Markoy Chain

With the development of society,the problem of energy shortage has gradually emerged.We need to pay more attention to the management and optimization of power resources to propose power usage scheduling that is more suitable for production planning.For Jilin province,the electricity consumption of the whole society is an important indicator to measure the national economic and social development.To enhance the benefit of electricity consumption,we are researching the electricity consumption of the whole society in Jilin province from various aspects.In this thesis,we use grey system theory and the cluster Markov chain as the main research methods to study the electricity consumption of the whole society in Jilin province from three dimensions:regional electricity,industrial and residential electricity,and sub-industry output value.The analysis shows that the electricity consumption in each region shows a steady development trend,so the policy adjustment of electricity consumption should be coordinated by all regions in the province.From the results of the grey relational degree of electricity in the whole society,it is concluded that for the short-term power gap problem,adjusting the electricity consumption of various industries in the secondary industry can control the growth rate of electricity more quickly.The results of the grey relational degree between the electricity consumption of the whole society and the output value of the sub-industries are inconsistent with the results of the grey relational degree of the sub-industries electricity consumption.By calculating the intensity of electricity consumption,it is concluded that the secondary industry’s electricity consumption intensity is much higher than the primary industry and the tertiary industry.Aiming at the problem of electricity consumption forecast in the whole society of Jilin province,we propose a grey-cluster Markov chain model based on the grey forecast model(GM(1,1)model).Firstly,a grey forecast model is established and the relative error data is obtained by fitting the original data;Secondly,the number of Markov states is determined by the systematic cluster method based on the relative error data sequence,and the number of Markov states is determined according to the K-means cluster center.The state interval is divided according to the value,and the fitted data is corrected by the cluster center value;Finally,the cluster Markov chain is used to calculate the forecast year’s relative error data state,and the forecast result is corrected based on the cluster center value corresponding to the state.The model is verified by the electricity consumption data of the whole society in Jilin province from 2000 to 2021.The results show that compared with the commonly used GM(1,1)model,the grey-cluster Markov chain model fits the original data better.The fitting accuracy is improved by 4.63%.Considering that the grey-cluster Markov chain model forecasts new data continuously,we update the data in the model,and the absolute relative error value of the updated model forecast doesn’t exceed 0.6%.Therefore,we verify a timely update of the forecast model dataset,which can improve the forecast accuracy.Using this model to forecast the electricity of the whole society in Jilin province during the "14th Five-Year Plan" period,the results show that the electricity consumption will increase gradually and the increase is relatively stable,and it’s estimated that the electricity consumption of the whole society in Jilin province will exceed 90 billion kWh in 2023,During the "14th Five-Year Plan" period,the annual electricity consumption of the whole society in Jilin province will not exceed 100 billion kWh.