Natural Gas Load Forecasting Based on Improved Back Propagation Neural Network

Abstract:

To suit for the condition that the relative error is more popular than the absolute error, and overcome the shortcoming of the traditional Back propagation neural network, this paper proposed an improved Back propagation algorithm with additional momentum item based on the sum of relative error square. The improved algorithm was applied to the example of the natural gas load forecasting, simulations showed that the improved algorithm has faster training speed than the traditional algorithm, and has higher accuracy as while.

• Periodical: Applied Mechanics and Materials (Volume 563)
• Pages: 312-315
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.563.312
• Online since: May 2014
• Authors: Yu Lian Jiang*
• Keywords: Additional Momentum Item, Back Propagation Neural Network, Forecasting, Relative Error

References

[1] Z. Wang, X. F. Zou, Journal of Beijing Institute of Technology, Vol. 21(2012), n2, pp.103-107.

[2] W. M. Tang, Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice, Vol. 25(2005), n10, pp.95-100, in Chinese.

[3] W. Hu, F. Li and H. M. Zhang, Journal of Software, Vol. 8(2013), n8, p.1960-(1967).

[4] H. T . He, Advances in Information Sciences and Service Sciences, Vol. 4(2012), n20, pp.378-385.

[5] Y. G. Xu, B. Z. Qian, Gas Industries, Vol. 12(2013), pp.25-27, in Chinese.

[6] T. Y. Yang, S. J. Fan and Y. Q. Chen, Calculation Methods of Oil and Natural Gas Reserves, Petroleum Industry Press, Beijing, (1989).