软计算与硬计算融合的中国石油需求预测  

The Forecasting of China's Oil Demand Based on Fusion of Soft Computing and Hard Computing

作　　者：郭海湘[1] 诸克军[1] 李玥[1] 王得运[1] 

机构地区：[1]中国地质大学管理学院,湖北武汉430074

出　　处：《中国地质大学学报（社会科学版）》2007年第6期24-28,共5页Journal of China University of Geosciences(Social Sciences Edition)

基　　金：国家自然科学基金资助项目(70573101);中国地质大学(武汉)资源环境经济研究中心开放基金资助;中国地质大学(武汉)优秀青年教师资助计划资助项目(CUGQNW0702)

摘　　要：近年来,随着我国经济的高速增长,石油作为重要的战略资源,在经济发展中发挥着越来越重要的作用。运用软计算和硬计算相融合的方法对我国2004—2015年的石油需求进行了预测。首先介绍了软计算和硬计算融合的串联模式,然后构建了影响我国石油需求(石油消费量)的主要指标,即国内生产总值(GDP)、人口总量、能源消耗量和居民消费水平。通过分析这些指标的数据特征,得出它们随时间的推移呈现增长的趋势。根据这一结论,运用软计算和硬计算融合中的串联模式,用硬计算中的龚珀资(Gompertz)增长曲线预测法来预测影响石油需求的4个指标以及石油需求量2004—2015年的值。在这些数据的基础上,再用软计算中的BP人工神经网络,对我国2004-2015年的石油需求量进行了预测。最后通过对HC和SC预测的石油需求量进行加权平均,得到了最终的预测结果。Nowadays, with the rapid development of China＇s economy, as an important strategic resource, oil plays a more and more important role in the economic development. This paper applies the fusion of soft computing and hard computing to the forecasting of China＇s oil demand from 2004 to 2015 year. Firstly, it presents the cascaded pattern of fusion of soft computing and hard computing, and then, sets up the main indexes affecting oil demand, viz. GDP, population, energy sources wastage, standard of living and oil consumption. By analyzing the data features of these indexes, it obtains their jack-up trend of time process. According to this conclusion, the cascaded pattern of soft computing and hard computing will be applied： using Gompertz curve of hard computing to forecast the value of these five indexes from 2004 to 2015, and then, on the basis of these data, applying BP Network of soft computing to forecasting the oil demand from 2004 to 2015 in China. Finally, the paper takes the weighted average of the oil demands from HC forecasting and SC forecasting as the final forecasting result.

关 键 词：软计算 硬计算 石油需求 龚珀资曲线 

分 类 号：F426.22[经济管理—产业经济]