Research On Optimal Dispatching Of Urban Water Supply System

The water supply system is an important component of urban infrastructure construction,it was not only directly affects the quality of life of urban residents,but also had a great influence for the development of urban economy.Therefore,under the dual requirements of water supply stability and energy conservation,it is necessary to study the optimal dispatch of water supply systemThe research on optimal scheduling of water supply system is divided into three parts:water consumption forecast,water supply pipe network pressure analysis and optimal dispatching decision.Among which,the water consumption forecast is the basis of optimal dispatch,the pipe network model pressure analysis is the decision variable of optimal dispatch,the water supply optimization dispatch is purpose.The prediction of water consumption is the premise and foundation of optimal dispatch.Combined with the characteristics of water supply in city,based on the analyzsis and comparison the current domestic and foreign water consumption forecasting methods,a model group based on GA-BP for water consumption forecasting algorithm is established.Predicting the water consumption of residents by predictive algorithm,the prediction results show that the GA-BP water consumption prediction algorithm has better prediction results,which improves the water consumption prediction accuracy,and provides a basis for subsequent optimized dispatch.The water supply network working condition model directly reflects the operation status of the whole water supply system,and the pressure of pressure measuring point in the water supply pipe network parameter is the hydraulic constraint condition of the water supply optimal scheduling model.Due to the complex diversity of pipeline networks,it is difficult to establish their mechanism models,The macroscopic pressure model of pressure measuring point based on BP neural network is established.Four pressure measuring points that reflect the operating conditions of the entire water supply system were selected for simulation analysis and compared with the measured values.The calculation results show that the pressure simulation value calculated by the neural network model has high accuracy and consistent with the operation of the water supply system.According to the operating characteristics of the water supply system,the total cost of the operation of the water supply system is set as the objective function,and takes the balance of water supply and demand in the pipeline network and the pressure of pressure measuring point as the constraints to optimize the effluent flow of water plant to achieve the purpose of energy saving and consumption reduction.This paper analyses the shortcomings of traditional particle swarm optimization algorithm,which is easy to fall into local optimum and has low convergence accuracy.An improved dynamic weight adaptive particle swarm optimization algorithm is adopted to improve the convergence of the algorithm.The improved convergence rate and convergence accuracy of the improved particle swarm optimization algorithm are improved by experimental analysis..Finally,taking Xi'an water supply system as an example to optimized dispatching analysis.The optimized water plant flow is allocated reasonably,and water plant total energy consumption reduced by 15%,which verifies the practical application value of the model.