基于最大熵法的雨水排水系统灾害损失不确定性研究  

Maximum Entropy-based Approach to Characterize Uncertainties of Flood Losses in Urban Storm Water Drainage Systems

作　　者：夏高原 李星[2] 金菊良[1] 刘丽[1] 

机构地区：[1]合肥工业大学土木与水利工程学院,安徽合肥230009 [2]长沙有色冶金设计研究院有限公司,湖南长沙410011

出　　处：《水电能源科学》2016年第6期123-127,共5页Water Resources and Power

基　　金：安徽省自然科学基金项目(1208085ME75);国家自然科学基金项目(51479046)

摘　　要：快速城市化进程和全球气候变化加重了排水系统压力,导致城市内涝频繁发生,严重威胁社会经济发展和公众生命安全。为有效评估现有排水系统灾害损失,充分考虑各种不确定性因子对评估结果的影响,采用蒙特卡洛法随机生成致灾因子集,并基于SWMM模型计算其对应的灾害损失,结合最大熵法得到最小无偏概率估计的特征,构建灾害损失的概率密度分布函数,对安徽省合肥市某区域雨水排水系统进行灾害损失评估。结果表明,该方法可有效提高随机模拟的收敛速度,计算结果客观合理,在城市排水系统灾害损失及风险分析中具有推广应用价值。Climate change and urbanization increased environmental stresses of urban drainage system which causes the frequent flooding occurrence in urbanized regions and threatens seriously social economic development and public life safety.To assess effectively the flood loss of urban drainage systems,the Monte-Carlo method was adopted to generate disaster-induced factors by considering the influence of uncertain factors on the evaluation results fully.And then the SWMM model was used to estimate the hazard loss.Finally,the maximum entropy-based approach was developed to construct the probability density distribution function of hazard losses.Application to a sample drainage system in Hefei City demonstrates that it provides an appropriate technique to determine probability distribution of flood losses.The results reveal that the proposed approach was capable of accelerating the convergence of stochastic simulation and could be applicable to hazard loss and risk assessment of urban water systems.

关 键 词：最大熵法 随机模拟 灾害风险 雨水排水系统 不确定性 

分 类 号：S276.3[农业科学—农业水土工程]