ABSTRACT
Coal resource plays a significant role in primary energy production and consumption in China. Thus, coal price has great influence on national economy and becomes an important issue for governments and relevant enterprises. In this article, key influencing factors of coal price are selected using a generalized regression neural network. And then, we introduce the honey bees mating optimization algorithm optimized by random Nelder–Mead to determine the kernel parameters used in least squares support vector machine for rapid and reasonable coal price prediction. Finally, price of Datong premium blend coal at Qinhuangdao port during the Twelfth Five Year Plan period is predicted. Compared with the prediction results of generalized regression neural network and honey bees mating optimization, the suitability and novelty of generalized regression neural network and random Nelder–Mead–honey bees mating optimization–least squares support vector machine method is fully demonstrated.
Funding
The work described in this paper was supported by The Energy Foundation of U.S. (G-1006-12630) and Beijing Education Commission Cooperation Project.