关键词: 灰色理论, 灰色预测模型, 能源消费, 马尔科夫链符号, 后验检验

Abstract: The rapid economic growth in China continues to result in the fast energy consumption and GHGs emission. Gray theory is used to develop the dynamic GM(1,h) models .for forecasting the final energy consumption in China. In order to improve the forecast accuracy, the original GM(1,h) models are improved by using two methodologies of equal dimension replenishment and residual modification with Markov-chain sign estimation. The results of posteriori checks show that the improved models are more reliable and have higher forecast accuracy. The forecast results of two economic growth scenarios indicate that China's final energy consumption will increase rapidly, and the average annual growth rates will reach 3.06% and 2.18% respectively in the period of 2000 to 2030.The CO2 e-missions for two scenarios in 2030 will increase to 2.15 and 1.60 times as against 2000(3898 Tg) respectively. Theenergy consumption of industrial and residential sectors will increase faster than that of other sectors. This reflects that China will accelerate the processes of industrialization and urbanization. The coal ratio will decrease and the e-electricity ratio will increase gradually in the final energy consumption.

Key words: gray theory, gray forecast model, energy consumption, Markov-chain sign, posteriori check