基于组合模型的我国能源需求预测

China's Energy Demand Forecast Based on Combined Model

在信息不完备条件下如何有效预测能源消费总量以把握宏观经济发展趋势,是制定能源规划的基础内容.伴随我国成为全球经济增长重要引擎,能源消费量越来越大,能源安全问题也备受关注.选用灰色系统模型、三次指数平滑模型和BP神经网络模型三种预测模型,应用Shapley值权重分配法确定各预测模型的权重,从而构建组合预测模型并对我国能源消费进行组合预测.在保持过去发展规律基本不变的条件下,2014年能源需求总量为383,718.16万吨标准煤,到2020年将达462,089.33万吨标准煤.2013-2020年能源需求总量年均增长率为3.38%.

How to effectively predict the total energy consumption in order to grasp the macro-economic trends under the condition of incomplete information,which is the energy planning＇s basic content.Along with our country become an important engine of global economic growth,energy consumption is growing,the energy security problem also in focus.In this paper,we use the grey system model,the three exponential smoothing model and BP neural network model for prediction,and application of Shapley value weight distribution method to determine the weight of each model,so as to build the combination forecast model and combined forecasting energy consumption in our country.Under the condition of keeping the rule of development of the past,the total energy demand in 2014 is 383,718 million tons of standard coal,by the year 2020 will reach 462,089 million tons of standard coal,the annual energy demand growth rate of 3.38% in 2013-2020.