基于改进的Shapley值法对我国清洁能源消费的组合预测

Combined Prediction of Clean Energy Consumption in China based on Optimized Shapley Value Method

选用GM(1,1)模型、指数平滑模型和ARIMA模型三种预测模型,通过非线性规划模型改进Shapley值算法确定各预测模型的权重,综合预测我国清洁能源消费趋势.预测结果显示,组合预测模型在拟合和预测我国清洁能源消费需求时相对误差均低于其他三种单一预测方法.根据预测结果,未来5年我国清洁能源的消费需求将持续增长,增速大致在10%左右,到2024年,预计我国清洁能源的消费需求量达到19亿吨标准煤左右.清洁能源将成为未来能源的发展趋势.

In this paper,we use the GM(1,1) model,the exponential smoothing model and the Arima model for prediction,and apply of optimized Shapley value based on nonlinear programming model,weight distribution method to determine the weight of each model,so as to comprehensively forecast the trend of China clean energy consumption.The results shows that comparing to other three methods,the combined forecasting model has lower relative error when fitting and predicting the consumption demand of clean energy.It can be known from prediction results,the clean energy consumption would have a rapid growth tendency,and the growth rate would be about 10% in the next five years.The consumption in the year of 2024 would be about 1.9 billion tons standard coal.Clean energy will be development trend of future energy.