基于等维新息GM(1,1)模型群的煤炭产量预测研究

Prediction of Coal Production Based on Equal-Dimension Metabolism GM(1,1) Model Group

以煤炭产量预测研究为目的,建立了GM(1,1)模型、GM(1,1)残差模型和等维新息GM(1,1)模型,并对我国1999年至2010年的煤炭产量进行了数据拟合和预测。应用后验差检验,3种模型均是一级合格模型;根据相对误差检验,GM(1,1)模型和GM(1,1)残差模型的数据拟合精度属于四级,是不合格模型;而等维新息GM(1,1)模型群的最大相对误差仅为0.46%,平均相对误差为0.40%,是适用于煤炭产量预测分析的高精度模型。应用等维新息GM(1,1)模型群对我国近5年煤炭产量进行预测,结果表明:煤炭产量将以平均每年超过3亿t的速度增长,到2015年将超过47亿t。

To predict the coal production in China, the GM （1,1） model, GM （1,1） Residual model and equal- dimension metabolism GM （1,1） model group was established. Based on the three models, the coal production from 1999 to 2010 was fitted and predicted. Through the posted- or-variance-test, all the models were the first grade qualified model. While tested by the relative-error-test, the GM （1,1） model and the GM （1,1） residuals model was the fourth grade model and was failure model, and then, the equal-dimension metabolism GM （1,1） model group was the high precision model, its maximum relative error is only 0.46% and the average relative error is only 0.40%. The coal production of recent 5-year was predicted by equal-dimension metabolism GM （1,1） model group, the result shows the coal production is growing more than 300 million tons annual and it will be more than 470 million tons in 2015.