基于余氯和THMs的管网水质服务水平模型研究

发布时间：2018-05-28 02:45

本文选题：供水管网 + 二次加氯　；参考：《浙江大学》2013年博士论文

【摘要】：为了对管网水质运行情况进行评价,优化管网水质,本研究中采用扩展余氯-I/O模型和THMs-I/O模型作为计算基础,建立管网水质服务水平-加氯总费用多目标优化模型。此外,由于水质监测点在反映实际管网水质服务水平时存在误差,故本论文对管网水质服务水平进行了修正。首先,为使余氯-I/O模型在余氯快慢反应共存时仍适用,并且提高THMs各组分的计算效率,本文推导了扩展余氯-I/O模型和THMs-I/O模型。采用管网末端的饮用水和水厂滤后水各自进行了加氯实验。结果表明当初始余氯质量浓度大于某个阈值后,快反应在0.5小时内基本完成,水质监测点的余氯质量浓度与加氯点上参与慢反应的初始余氯质量浓度符合线性比例关系。基于实验结果,本文提出了扩展余氯-I/O模型的适用条件。在中试管网中验证扩展余氯-I/O模型,结果表明扩展余氯-I/O模型优于传统的余氯-I/O模型。采用基于余氯消耗的THMs各组分生成模型拟合THMs各组分的质量浓度。考虑温度、初始余氯质量浓度和溴离子质量浓度为影响因素,采用正交实验观察三因素对模型中的THMs各组分线性比例系数的影响。实验结果表明,THMs各组分线性比例系数受溴离子质量浓度的影响最大,而受温度和初始余氯质量浓度的影响较小。在中试管网中验证THMs-I/O模型,结果显示实测值与预测值间的相对误差在9%以内。其次,为研究管网水质运行情况的评价模型,本文引入余氯服务水平概念,初步探讨了管网水质服务水平评价函数,改进了二次加氯优化模型。参考余氯服务水平和《生活饮用水卫生标准》,建立了THMs各组分的服务水平。考虑加权平均值、加权连乘值和加权标准差三类函数,用于建立关联余氯及THMs各组分服务水平的管网水质服务水平表达式。本文采用管网水质服务水平-加氯总费用多目标模型对管网水质运行情况进行评价。结合一管网算例,采用EPANET_MSX模拟余氯衰减和THMs各组分生成,利用实数编码的非支配排序遗传算法(NSGA-Ⅱ)进行优化,得到各种情况下的Pareto前沿面。计算结果表明,加权平均表达式为关联余氯和THMs服务水平的最优方式。二次加氯点布置于水塔出水口和管网末端即可,过多的二次加氯点不仅不能显著提高管网水质服务水平,还会加大加氯总费用。余氯衰减系数对pareto面的影响最大,故改善水质更换老旧管道是提高管网水质服务水平且节省加氯总费用最可行的方法。最后,本文研究了水质监测点在反映实际管网水质服务水平时存在误差的现象,探讨了管网水质服务水平的修正。模拟结果表明,随着水质监测点数量的增加,误差有减小的趋势,但不呈现单调递减。当水质监测点数量较多时,误差可忽略。论文比较了水质监测点覆盖率和水质监测点数量占管网节点总数的比率拟合误差的效果,结果显示后者较优。通过一大型管网验证,结果表明修正后的管网水质服务水平更加接近于实际管网水质服务水平。本论文得到了水体污染控制与治理重大专项：饮用水水质监控预警及应急技术研究与示范(2008ZX07420-004)、潮汐影响地区饮用水安全保障技术集成与示范(2009ZX07424-001)和山地丘陵城市饮用水安全保障共性技术研究与示范(2009ZX07424-004)以及国家自然科学基金(51208455)的资助。
[Abstract]:In order to evaluate the water quality of the pipe network and optimize the water quality of the pipe network, the extended chlorine -I/O model and the THMs-I/O model are used as the calculation basis to establish the multi-objective optimization model of the total cost of the water quality service of the pipe network. In addition, the water quality monitoring point has errors when it reflects the water quality service level of the actual pipe network. The paper revise the water quality service level of the pipe network.
First, in order to make the residual chlorine -I/O model applicable to the coexistence of the fast and slow reaction of the residual chlorine, and to improve the calculation efficiency of the THMs components, the extended residual chlorine -I/O model and the THMs-I/O model are derived. The experiment of adding chlorine at the end of the pipe network and the filtered water in the water plant is carried out respectively. The results show that the initial chlorine mass concentration is greater than a certain threshold. After 0.5 hours, the fast reaction is basically completed in 0.5 hours. The concentration of residual chlorine in the water quality monitoring point is in linear proportion with the initial chlorine mass concentration in the slow reaction on the chlorine point. Based on the experimental results, the application conditions of the extended residual chlorine model are proposed. The extended residual chlorine -I/O model is verified in the medium test tube network, and the result shows the expansion surplus. The chlorine -I/O model is superior to the traditional residual chlorine -I/O model. The mass concentration of THMs components is fitted by the THMs component generation model based on the waste chlorine consumption. Considering the temperature, the initial chlorine mass concentration and the bromine ion mass concentration as the influencing factors, the orthogonal experiment is used to observe the influence of the three factors on the proportion coefficient of THMs in the model. The experimental results show that the proportion coefficient of THMs in each group is most affected by the mass concentration of bromine ion, but less influenced by the temperature and the mass of the initial chlorine. The THMs-I/O model is verified in the tube network. The results show that the relative error between the measured value and the predicted value is less than 9%.
Secondly, in order to study the evaluation model of the water quality of the pipe network, this paper introduces the concept of residual chlorine service level, preliminarily discusses the evaluation function of the water quality service level of the pipe network, and improves the two chlorine optimization model. The service level of the THMs components is established with reference to the level of residual chlorine service and the hygienic standard of drinking water, and the weighted average value is considered. Three functions of weight and weighted standard deviation are used to establish the expression of water quality service level for the network of associated chlorine and THMs components. In this paper, the water quality operation of the pipe network is evaluated by the multi objective model of the total cost of the water quality service of the pipe network and the total cost of the total cost of chlorine. A network calculation example is used to simulate the decay of residual chlorine and T. HMs components are generated and optimized using the real coded non dominated sorting genetic algorithm (NSGA- II) to get the Pareto frontiers in various cases. The calculation results show that the weighted average expression is the best way of the associated residual chlorine and the THMs service level. The two addition of chlorine points is arranged at the outlet of the water tower and the end of the pipe network, and the excess of two times is added. Chlorine point not only can not significantly improve the water quality service level of the pipe network, but also increase the total chlorine cost. The effect of residual chlorine attenuation coefficient on Pareto surface is the most. Therefore, it is the most feasible way to improve the quality of the pipe network and to save the total chlorine cost by improving the quality of the old pipes.
At last, this paper studies the phenomenon that the water quality monitoring point has errors when it reflects the water quality service level of the actual pipe network, and discusses the correction of the water quality service level of the pipe network. The simulation results show that the error has a decreasing trend with the increase of the quantity of water quality monitoring points, but it does not decrease monotonically. The paper compares the effect of the ratio of water quality monitoring point coverage and the quantity of water quality monitoring points to the proportion of the total number of nodes in the pipe network. The results show that the latter is better. The results show that the water quality service level of the revised pipe network is closer to the water quality service level of the actual pipe network.
This paper has been given a major project for water pollution control and control: the research and demonstration of drinking water quality monitoring, early warning and emergency technology (2008ZX07420-004), integration and demonstration of drinking water safety protection technology in tidal areas (2009ZX07424-001) and common technology research and demonstration of drinking water security in Hilly and hilly cities (2009ZX07424-004) And funding from the National Natural Science Foundation (51208455).
【学位授予单位】：浙江大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：TU991.21

【参考文献】