摘要
Hubbert曲线的研究方法逐渐成为能源供给问题研究的热点，Hubbert曲线的系统动力学模型建立了动态模拟平台，讨论中国石油产量高峰和高峰年份，但对模型的预测能力并未进行检验。针对经典Hubbert曲线采用的1960年之前美国48州石油产量数据，预测1961～2006年美国石油年产量，并与经典模型、实际石油产量进行比较和检验，给出线形回归和离差平方和的检验结果。结果表明Hubbert曲线的系统动力学模型对石油最终可采储量的取值比较敏感，如果最终可采储量选取适宜则可以给出比较接近实际产量的预测结果。
Research method of Hubbert curve has become a current focus in the field of energy supply. The Hubbert curve based on the System Dynamics （SD） establishes a dynamic stimulated platform to discuss the Chinese oil production peak, but its forecast ability has not been tested yet. According to the classic Hubbert curve, the American oil data （from 48 states） before 1960 is adapted in the SD model to forecast American annual oil production, and to compare the corresponding forecast results with ones from classic model and actual production. The tests of the forecast capability for SD model of Hubbert curve are carried out by calculating the linear regression and sum of squares for results in low and standard scenarios. The conclusions show that the forecast results is sensitive to the ultimate reserves, and that the SD model of Hubbert curve can provide more robust forecast results when the selected amount of ultimate reserves is proper.
出处
《系统工程》
CSCD
北大核心
2009年第2期102-108,共7页
Systems Engineering
基金
国家自然科学基金资助项目(70772097)
