基于灰信息的黄河冰凌灾害风险评估研究

发布时间：2018-05-23 22:31

本文选题：灰关联 + 灰色聚类　；参考：《华北水利水电大学》2017年硕士论文

【摘要】：灰色评估模型是评价模型中的的重要组成部分,本文对几个灰色评估模型进行了研究,并将其运用于信息具有灰色特性的黄河冰凌灾害风险评估问题中,主要成果如下:首先,对流凌期的黄河宁蒙河段进行了冰塞易发性风险评估。对于水文信息具有灰色特性的黄河宁蒙河段,考虑到评估者的心理因素,提出一种基于后悔理论的灰关联评估方法。构建了欣喜后悔值函数,求得黄河巴彦高勒河段、三湖河口河段和头道拐河段的欣喜值与后悔值;并以欣喜后悔值最大的原则建立权重优化模型,得到流量、气温、水深等指标权重;以欣喜后悔值的大小对各河段进行排序。结果表明:三湖河口河段较巴彦高勒河段、头道拐河段更易发生冰塞灾害。其次,对黄河巴彦高勒河段进行了10年凌汛易发性风险分析。依据冰凌形成机理,构建由气温、流量、水位组成的凌汛易发性风险评价标准;借助灰色聚类方法和分析技巧,建立灰色白化权函数的二阶段聚类评估模型;选取黄河巴彦高勒河段2006-2016年气温、流量、水位指标数据进行聚类分析,并根据聚类结果作出合理的风险评估。结果表明:黄河巴彦高勒河段每年凌期2/3左右的天数的凌汛易发性的风险等级分布在中等、较高级别;就每年度整体风险而言,近10年中有3年处于较高风险,7年处于中等风险;而且每年风险时间分布呈现驼峰,即为流凌期风险较低,初封期风险较高,稳封期风险中等,开河初期风险较高,开河末期风险较低。最后,对开河期的黄河宁蒙河段进行了动态风险型评估分析。选取的2004-2016年黄河石嘴山河段、巴彦高勒河段、三湖河口河段的指标数据,并划分为3个阶段,构建了动态风险型灰靶模型。利用后悔理论与数学期望相结合的方法,构建了集结欣喜-后悔值函数,把风险型评估问题转化为无风险评估问题,并以近期数据参考价值更高为基准,把动态评估问题转化为静态评估问题;引入椭形灰靶模型,求得各个方案的偏离靶心度并对3个河段进行排序。结果表明:巴彦高勒河段开河期风险较三湖河口河段、石嘴山河段更高。本文构建了多个灰色评估模型,并应用于黄河冰凌灾害评估问题中。合理的评估结果证实了模型的科学性与实用性,也为黄河冰凌灾害的防治提供了借鉴。
[Abstract]:Grey evaluation model is an important part of evaluation model. This paper studies several grey evaluation models, and applies them to the problem of risk assessment of Yellow River ice with gray characteristics. The main results are as follows: first of all, Ice slug vulnerability risk assessment was carried out in the Ningmeng reach of the Yellow River during convection transit. In view of the psychological factors of the evaluators, a grey correlation evaluation method based on regret theory is proposed for the Ningmeng reach of the Yellow River, where hydrological information has grey characteristics. The function of happy regret value is constructed to obtain the joy value and regret value of Bayangol reach, Sanhu estuary reach and Tou Daoguanhe reach of the Yellow River, and the weight optimization model is set up with the principle of maximum happy regret value, and the flow rate and temperature are obtained. The weight of indexes such as water depth and the order of river reach by the value of joy regret. The results show that the reach of Sanhu Estuary is more prone to ice slug than that of Bayangol reach. Secondly, the risk of ice flood in Bayangol reach of the Yellow River for 10 years is analyzed. According to the ice formation mechanism, the risk evaluation standard of ice flood susceptibility is established, which is composed of temperature, discharge and water level, and the two-stage clustering evaluation model of grey whitening weight function is established with the help of grey clustering method and analysis technique. The data of temperature, discharge and water level in Bayangol reach of the Yellow River from 2006 to 2016 were selected for cluster analysis, and a reasonable risk assessment was made according to the clustering results. The results show that the risk level of ice flood susceptibility of the Bayangol reach of the Yellow River is about 2 / 3 days per year, and the risk level of flood susceptibility is in the middle, higher level, and in terms of the overall risk per year, In the last 10 years, 3 of them are at higher risk, 7 of them are at medium risk, and the risk time distribution of each year is hump, that is, the risk of flow transit period is lower, the risk of initial seal period is higher, the risk of stable seal period is medium, and the risk of initial period of Kaihe River is higher. The risk at the end of the river is low. Finally, the dynamic risk assessment of the Ningmeng reach of the Yellow River in Kaihe period was carried out. The index data of Shizuishan reach, Bayangol reach and Sanhu River reach of the Yellow River from 2004 to 2016 were selected and divided into three stages. A dynamic risk grey target model was constructed. Based on the method of combining regret theory with mathematical expectation, this paper constructs the aggregate-regret value function to transform the risk-type evaluation problem into a risk-free evaluation problem, and takes the higher reference value of the recent data as the benchmark. The dynamic evaluation problem is transformed into the static evaluation problem and the elliptical grey target model is introduced to calculate the deviation from the target centroid of each scheme and to sort the three reaches. The results show that the risk of Bayangol reach is higher than that of Sanhu estuary reach and Shizuishan reach. In this paper, several grey assessment models are constructed and applied to the problem of ice disaster assessment of the Yellow River. The reasonable evaluation results confirm the scientific and practicability of the model, and also provide reference for the prevention and control of the Yellow River ice fall disaster.
【学位授予单位】：华北水利水电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TV875;TV882.1;N941.5

【参考文献】