基于水力模型的雨水蓄水池优化研究

发布时间：2018-05-26 19:29

本文选题：暴雨强度公式 + SWMM模型　；参考：《哈尔滨工业大学》2015年硕士论文

【摘要】：近几年,随着我国城市化进程的加快以及全球气候的变化,造成城市极端内涝灾害频繁发生,如此频繁严重的城市内涝灾害对人民群众的生命安全和社会经济的发展都造成了巨大影响。雨水蓄水池作为一种低影响开发设施可以有效地削减洪峰流量,降低城市内涝风险。目前,国内外的研究主要集中在雨水蓄水池控制非点源污染及其设计方法上,本文从排水管网水力模型的角度,基于水池造价和城市内涝风险的考量,研究雨水蓄水池的优化选址和优化设计问题,本文的主要研究内容可分为3个部分。首先,本文进行了排水管网建模的研究。本文对排水管网模型理论进行了介绍,重点对SWMM模型的基本原理进行了研究。作为排水管网模型的重要输入参数——暴雨强度公式,对其推求方法进行了相关研究,进行了基于不同适线准则的频率分布曲线调整研究,研究发现基于降雨强度残差平法和最小的适线准则可以取得很好的适线效果;基于一种改进高斯-牛顿法进行了暴雨强度公式参数求解研究,并与粒子群优化算法(PSO)和实数编码遗传优化算法(RCGA)进行了对比,研究结果表明该方法具有较高的求解精度。本文以所推求的暴雨强度公式为基础,应用SWMM软件建立了W市某排水区的排水管网水力模型。其次,本文进行了多目标进化算法NSGA-II的研究。基于主从式模型,提出了一种并行NSGA-II算法,并通过基本多目标优化问题对该算法进行了相关测试,测试结果表明该算法是一种快速、有效的算法。最后,本文进行了雨水蓄水池优化研究。基于SWMM模型和优化问题,开发了11个外部接口函数,扩展了SWMM的API函数。以水池的位置和设计参数为决策变量,水池造价、溢流总量以及溢流节点总数为目标函数建立了雨水蓄水池多目标优化模型,并利用并行NSGA-II算法对所建立的优化模型进行了求解。优化结果表明最优解的空间分布(雨水蓄水池位置)与城市内涝点的空间分布基本吻合;同时,位于不同位置的解对两个水力目标的干扰的效果,从上游到下游呈现明显的过渡规律,位置位于中上游的解对于降低溢流节点总数更有优势,而位置位于下游的解对降低溢流总量更有优势。
[Abstract]:In recent years, with the acceleration of urbanization and the change of global climate in China, extreme waterlogging disasters occur frequently in cities. The frequent and serious urban waterlogging disaster has a great impact on the safety of people's lives and the development of social economy. As a low impact development facility, Rain Water reservoir can effectively reduce the flow of Hong Feng and reduce the risk of urban waterlogging. At present, the research at home and abroad is mainly focused on the control of non-point source pollution by Rain Water reservoir and its design method. From the point of view of hydraulic model of drainage pipe network, based on the consideration of pool cost and urban waterlogging risk, In this paper, the optimal location and design of Rain Water reservoir are studied. The main contents of this paper can be divided into three parts. Firstly, the modeling of drainage pipe network is studied in this paper. In this paper, the theory of drainage pipe network model is introduced, and the basic principle of SWMM model is studied. As an important input parameter of drainage pipe network model, the torrential rain intensity formula is studied, and the frequency distribution curve adjustment based on different line fitting criteria is studied. It is found that the rain-intensity residual leveling method and the minimum line fitting criterion can obtain good alignment effect, and the parameters of rainstorm intensity formula are solved based on an improved Gao Si Newton method. Compared with particle swarm optimization (PSO) and real-coded genetic optimization (RCGA), the results show that the proposed method has high accuracy. Based on the derived formula of rainstorm intensity, the hydraulic model of drainage pipe network in a drainage area of W city is established by using SWMM software. Secondly, this paper studies the multi-objective evolutionary algorithm (NSGA-II). Based on the master-slave model, a parallel NSGA-II algorithm is proposed, and the algorithm is tested by the basic multi-objective optimization problem. The test results show that the algorithm is a fast and effective algorithm. Finally, the optimization of Rain Water reservoir is studied. Based on SWMM model and optimization problem, 11 external interface functions are developed, and the API function of SWMM is extended. Taking the position and design parameters of the pool as the decision variables, the cost of the pool, the total amount of overflow and the total number of overflow nodes as objective functions, the multi-objective optimization model of Rain Water reservoir is established. The optimization model is solved by parallel NSGA-II algorithm. The optimization results show that the spatial distribution of the optimal solution (the location of the Rain Water reservoir) is basically consistent with the spatial distribution of the urban waterlogging point, and at the same time, the interference effect of the solution at different locations on the two hydraulic targets is obtained. The solution located in the upper and upper reaches has more advantages in reducing the total number of overflow nodes, while the solution located in the downstream has more advantages in reducing the total overflow volume.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU992.1

【参考文献】