Multi-objective Optimal Location Of Water Quality Monitoring Stations For Conventional Water Supply Networks In Mountain City

The water supply network is an important infrastructure of the city,and it spreads in every corner as the city develops.Setting up water quality monitors in the pipe network is an effective means to monitor the water quality conditions in the pipe network,but due to economic factors,it is not possible to set monitors at all nodes.Therefore,under the limited number of monitoring points,how to select the node that can best represent the water quality characteristics in the pipeline network for monitoring is worth further research.This study is funded by "National key research and development program(2017YFC0404706)".Based on the characteristics of mountain city water supply network and demand coverage method,the goal of optimized location model is to maximize demand coverage and highest water quality score.Then based on the idea of the ideal point method,using GA genetic algorithm to solve it on the MATLAB platform,which provides scientific basis and experience for the optimization arrangement of water quality monitors in mountainous cities.(1)First of all,establishing hydraulic water quality model of pipe network based on Epanet 2.0 software.After verification and simulation,it can truly reflect the operating conditions of the water supply network and obtain the hydraulic and water quality datas of the pipe network at different times including water transfer data between pipe network nodes.The MATLAB platform is used to obtain the coverage matrix of any node through the conversion of the programming language to provide basic data for solving the optimization model.(2)Secondly,based on the characteristics of mountain city water supply network and the flaw of demand coverage method,proposed "node water quality scoring model",which including pipeline conditions,water quality conditions,and node location conditions around the node.The fuzzy analytic hierarchy process of group decision-making was used to solve the weights of each index,and the water quality score of each node was obtained.A higher score means more worthy of being monitored.(3)Thirdly,according to the node water quality scoring model,a two-objective optimization model is established with the goal of "maximize demand coverage and highest water quality score".Combining the characteristics of the mountain city pipe network with the flaws of demand coverage method,and improving on the basis of the traditional single-objective optimization model,a second objective function,"highest water quality score",was introduced to improve the representative.At the same time,the node water quality scoring model considered the weight factors of important nodes(such as hospitals,schools,etc.).Because of that,it improves the reliability of optimization results.(4)fourthly,according to the idea of the ideal point method,a solution model of multi-objective optimization model based on GA genetic algorithm is proposed,which reduces the difficulty of solution.Aiming at the improved two-objective optimization model,GA genetic algorithm was used to optimize the optimization model on the MATLAB platform and verified by a pipe network example.The results show that the two-objective optimization model is more representative.(5)Finally,based on engineering examples,a simplified model of the actual water supply network and a reasonable division of the water supply area are established.The improved multi-objective optimization model was used to solve the model on the MATLAB platform,and a reasonable water quality monitoring point location was determined.The results show that rearranging the location of water quality monitors in the area according to the optimization can improve the safety of urban water supply quality.