Prediction on China's Energy Demand Based on RBF Neural Network Model

Abstract:

Choose factors which influence the energy demand by the method of path analysis, build radial basis function (RBF) neural network model to predict energy demand in China. The RBF neural network is trained with the actual data of the main factors affecting energy demand during 1989-2003 and energy demand during 1993-2007 as learning sample with a good fitting effect. After testing network with the actual data of the main factors affecting energy demand during 2004-2007 and energy demand during 2008-2011, higher prediction accuracy can be obtained. By comparison with the BP network, RBF network prediction model outperforms BP network prediction model, finally RBF network is applied to make prediction of energy consumption for the year 2013-2015.

• Periodical: Advanced Materials Research (Volumes 805-806)
• Pages: 1421-1424
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.805-806.1421
• Online since: September 2013
• Authors: Xue Feng*, Wuyunbilige Bao, Ben Ha
• Keywords: Influencing Factors of Energy Demand, Path Analysis, Prediction of Energy Demand, RBF Neural Network Model

References

[1] Box G. ．E．P．, Jenkins G. ．M. ．Time Series Analysis, Forecasting and Control[M]．San Francisco: Holden-Day, 1970.

[2] Christopher A. Sims. Macroeconomics and Reality[J]. Econometrica, 1980, 48(1): 1-48.

[3] Zhi-yong Han, Yi-ming Wei. On the Cointegration and Causality between Chinese GDP and Energy Consumption [J]. Systems engineering, 2004, 22 (2) : 17-21. (In Chinese).

[4] Yi-ming Wei, Ying Fan, etc. China's energy report (2006) strategic and policy research [M]. Beijing: science press, 2006: 68-92. (In Chinese).

[5] BP. BP Statistical Review of World Energy, (2010).

[6] Juan Gao. Principle and cases of Artificial Neural Network[M]. Beijing: Machinery Industry Press, 2003: 44-80.