Optimal Dispatching Model Of Water Supply Network Based On Water Consumption Forecast And Genetic Algorithm

With the continuous development of the city,its water supply pipeline network has gradually developed to be huge and complex.The use of traditional manual scheduling methods will not be able to accurately control the water demand of each node of the pipeline network.How to minimize the cost of water supply under the premise of ensuring the water pressure and water demand of the entire city is a hot research topic at the moment.This paper aims to save water supply costs,based on the actual data of Jiaxing’s water supply pipe network,establishes the hourly water consumption forecast model,water supply pipe network micro-hydraulic model,and two-level optimal dispatching model in Jiaxing City.It is on the premise of ensuring the water demand of the entire city.The main work of this paper is as follows:Water consumption prediction is the prerequisite for optimized dispatch.This article analyzes the relevant factors that affect the short-term water consumption of Jiaxing City based on the actual water consumption monitored by the pipeline network in 2020.Based on the results of the analysis,combined with the Long Short-term Memory Neural Network(LSTM),the hourly water consumption prediction model of Jiaxing City was established.The model was used to predict the water consumption of Jiaxing City for three consecutive days and evaluate the model.The water consumption and forecast within three days Two-thirds of the deviation of the value is below 3%,and four-fifths are below 4%.The MAPE of the model is 0.031.The analysis results show that the model is used to predict the water consumption in Jiaxing City.effect.The hydraulic model is the basis for optimal dispatch.This paper collects and sorts out the static attribute data and actual monitoring data of the water supply network in Jiaxing City,and establishes the micro hydraulic model of the pipe network through the open source software EPANET developed by the Water Supply and Water Resources Development Department of the U.S.Environmental Protection Agency.And simplify the hydraulic model.The simplified model is repeatedly checked and adjusted to ensure that the accuracy of the model meets industry standards.After verification,the error between the monitoring point pressure calculated by the microhydraulic model and the real pressure is less than 1m in 84% of the time,and less than2 m in 95% of the time,basically in line with domestic standards,so the model has Good accuracy.Based on the data obtained above,this paper establishes a first-level optimal scheduling model for Jiaxing’s water supply network with the goal of minimizing water plant energy consumption,and on this basis,establishes Jiaxing’s water supply system with the goal of pumping station operating costs,that is,the lowest power consumption.The second-level optimization scheduling model of the pipe network.Next,the genetic algorithm is used to solve the two models separately.The penalty function method is used to convert the constrained problem into an unconstrained problem.The result after the solution is compared with the real situation of the pipe network.The result shows that the generated optimal scheduling plan is Compared with the actual use of the program,the average energy saving is 8.33%,which proves that the two-level optimal scheduling model of the Jiaxing water supply network can indeed achieve the effect of energy saving and reducing the power consumption of water plants.