摘要
精准预测未来煤炭需求量对能源政策的制定和产业结构的调整具有显著的积极意义。使用一种新的混合优化算法即基于指数退火的自适应蝙蝠算法(ABA-ESA算法),该算法不仅继承了蝙蝠算法的全局搜索能力和模拟退火算法的局部搜索能力,而且加强了算法在局部的搜索能力,加快了总体收敛速度。选取经济增长、城镇化进程和能源结构作为ABA-ESA的输入因子,使用1981-2015年共35年间各因子及煤炭消耗量作为观察数据,建立二次方程形式的煤炭需求预测模型,并将建立的模型与其他模型进行比较,发现该预测模型在准确性能上有很大的优势。预测结果表明,2020年和2030年我国煤炭需求量分别为30.01亿t标准煤和44.41亿t标准煤。
Accurately forecasting coal demand in the future has significant positive effects on the formulation of energy policies and the adjustment of industrial structure.This paper used a new hybrid optimization algorithm,ABA-ESA algorithm,which is based on exponential annealing adaptive bat algorithm.It not only inherits global search ability of bat algorithm and local search ability of simulated annealing algorithm,but also enhances the algorithm search in local ability and accelerates the overall convergence rate.This paper selected economic growth,urbanization process and energy structure as the input factors of ABA-ESA,and used the factors from 1981 to 2015 and coal consumption as observation data to establish the secondary form of coal demand forecasting model.The established model was compared with other models and it was found that this model has a great advantage in accuracy.The forecast results showed that China's coal demand in 2020 and 2030 are respectively 3.001 billion tons of standard coal and 4.441 billion tons of standard coal.
作者
邹绍辉
丁治立
Zou Shaohui;Ding Zhili(School of Management, Xi'an University of Science and Technology, Xi'an, Shaanxi 710054, China;Energy Economy and Management Research Center, Xi'an University of Science and Technology, Xi'an, Shaanxi 710054, China)
出处
《中国煤炭》
2018年第7期9-14,20,共7页
China Coal
基金
国家自然科学基金(71273207)
陕西省科学技术研究发展计划项目(2011kjxx54)