Research On Energy Consumption Prediction Based On Time Series Analysis In Xinjiang

Xinjiang is a magnificent resource area in China as well as a province of high energy consumption.The obvious imbalance of energy consumption space and excessive consumption intensity have become the "short board" that restricts the development of low-carbon economy.Since the 18 th National Congress of the CPC,Xinjiang has been designated as the core area of the Silk Road Economic Zone by the central government.It bears a significant responsibility for reducing the national energy consumption.Deeply analysis and prediction of Xinjiang's future energy consumption level is of great practical significance for the government to formulate a long-term development strategy for effective energy utilization,so as to achieve the goal of energy conservation and emission reduction.“Online Monitoring Platform for Xinjiang's Energy Consumption Data” serves as the basic database for aiding energy administrative departments to know the energy consumption situation in Xinjiang,and plays a role of “energy barometer”.In this thesis,the energy consumption data of the platform from 2011 to 2017 is collected in order to summarize the trends of energy consumption in Xinjiang of recent years.Then,aiming to further realize the early warning function of online monitoring energy consumption data on the platform,the published energy consumption data by the Statistics Bureau of Xinjiang is selected to establish three types of model based on the time series analysis method,i.e.the trend extrapolation model,the ARIMA model and the exponential smoothing model,by using Eviews software.Results of comparative experiment among these three models on actual values show that the exponential smoothing model was the optimal.Therefore,the exponential smoothing model was used to predict the total energy consumption in Xinjiang from 2018 to 2020 and the forecasting bias was analyzed and summarized for which provided a way to further optimize the forecasting model.Finally,the exponential smoothing model was also used to analyze and predict the consumption proportion of coal and clean energy so as to get the change of Xinjiang energy structure,summarize the current situation of energy development,and put forward energy utilization development proposals for the government to provide a basis for energy conservation and emission reduction targets.