摘要
针对三峡工程供水期各时段的丰枯遭遇对制定供水期调度方案的影响,以及统计方法的不足,提出了比较真实反映供水期相邻时段丰枯遭遇联系的Copula联合分布函数。以基于Copula函数建立的贝叶斯网络模型,分别分析了先验概率和后验推理对丰枯遭遇状态。Copula函数计算结果和统计方法统计的结果基本吻合,证明了所建立的联合分布函数的合理性。先验概率分析结果显示相邻时段不利调度概率较小,而后验推理分析结果表明不利调度的概率较大。综合考虑,有效利用后验推理分析的信息,能在一定程度上减少来水不确定性对供水调度方案制定带来的影响,提高三峡工程供水期的综合效益。
In view of the influence of wetness - dryness encounter on operation scheme of TGP in water supply period and shortcomings of statistic method, Copula combination function is proposed, which can reflect the wetness - dryness connection of two adjacent periods in water supply period. The calculation result of Copula combination function is basically in agreement with that of statistical method, proving its rationality. Based on Copula function, a Bayesian network model is established and used to analyze the wetness -dryness encounter state by prior probability and posterior inference respectively. The analysis result shows that the unfavorable operation probability is smaller by prior probability while unfavorable operation probability is larger by posterior inference. Through comprehensive consideration, the effective utilization of information by posterior inference can reduce the adverse influence of inflow uncertainties on water supply, and improve the comprehensive benefit of TGP during water supply period.
出处
《人民长江》
北大核心
2012年第3期5-8,共4页
Yangtze River
