Research On Optimization Of Water Supply Network Based On Multi-objective Optimization Algorithm

As an important part of modern urban infrastructure construction,water supply network plays an important role in urban development and protection of people’s lives.Nowadays,the scale of water supply network is growing rapidly with the expansion and development of the city.It is one of the most urgent challenges for the government and water supply providers to build or repair the existing water supply networkIn the field of water supply network optimization,scholars tend to minimize the investment cost as a single objective to optimize the design of water supply network.Due to the lack of water supply reliability,it can not be applied to the actual network design.In recent years,the multi-objective evolutionary algorithm obtains different compromise solutions through Pareto domination,which is favored by more and more scholars in the optimization of water supply network.In this paper,the mathematical model of water supply network optimization is designed,and an improved multi-objective evolutionary algorithm based on decomposition is proposed to solve the optimization problem of water supply network.On the basis of this algorithm,combined with the shortcomings of the existing water supply network system,the visualization system of water supply network is designed and implemented.The main work is as follows.(1)A constrained MOEA / D algorithm is proposed to solve the design problem of water supply network.First,combined with the basic theory of water supply pipe network hydraulics,the characteristics of the water supply pipe network design problem are analyzed,and the economic goals,reliability goals and solution constraints of the water supply pipe network are designed.Secondly,in the face of the complicated hydraulic calculation problems in the water supply pipe network,the EPANET tool is introduced,and the EPANET tool is combined with the MOEA/D algorithm to collaboratively solve the pipe network design plan.In the solution process,the defects generated by the calculation of hydraulic constraints are improved.The experimental results show that the benchmark test problem of Belerma network and ZJ network is compared with the classic NSGA-II algorithm.The algorithm proposed in this paper is better than the NSGA-II algorithm in Pareto frontier distribution,diversity maintenance and convergence.Finally,the design solutions generated by the algorithm in the Belerma network are analyzed,and relevant suggestions are given to the decision-makers’ choice,which verifies the effectiveness of the algorithm proposed in this paper in the optimization design of water supply pipe network.(2)A water supply network visualization system is implemented to solve the problem of sensor data islands and customized design of pipe networks.This paper divides the water supply network visualization system into user management module,sensor data analysis module,pipe network optimization design module and pipe network visual management module.On the one hand,using the sensor data analysis module,open up the process of collecting,analyzing,verifying,and storing sensor data in the water supply pipe network,and store multi-source data in a unified way to realize data interconnection and interoperability,and solve the independence of the previous multi-source data in the pipe network.The formation of data islands.On the other hand,using the pipe network optimization design module and the pipe network visual management module,the constrained MOEA/D algorithm for solving water supply pipe network design problems realized in this paper is combined with GIS and visualization technology to construct a full link of water supply pipe network data.The visual service ecology is convenient for designers to visually design and managers to visually manage the water supply network.