Study On Optimal Operation And Maintenance Of Water Supply Network

With the continuous progress of urban development,the range of water supply and water supply has gradually increased,and the requirements for economic benefit and safety and stability of water supply system operation have also been improved.At present,most water plants in China still adopt the empirical method to regulate and control the water supply,resulting in high water supply pressure,energy waste and other problems.Therefore,the study on optimal operation and maintenance of water supply pipe network is of great significance to improve the operation status of pipe network.In this paper,BIM technology is used to establish a three-dimensional visual model of the water plant pump station and integrate the parameter information and position information of each component.The Internet of things technology is used to realize the online monitoring of pipe network operation data,which is displayed on the water supply monitoring platform in real time,making the water supply dispatching more intuitive.Two methods of water consumption prediction were compared: moving average method and BP(Back Propagation)neural network method.The moving average method selects the water consumption data of 12 days as the data sample,calculates the data of the 12 th day through the data of the first 11 days,and compares it with the measured data.BP neural network to choose 14 days water consumption data as the data sample,data as the training sample,7 days before the day 7 to 13 days data as test samples,14 days data as the forecast validation sample,to build a three layers BP neural network,the test result proved that the BP neural network has higher precision and satisfies the requirement of water supply network optimization operational accuracy.This paper analyzes the practicability of the network micro model and the network macro model respectively,and establishes the network segment macro model based on BP neural network according to the actual demand of urban water supply dispatching.A 24-hour day is divided into 6 optimized scheduling periods,each of which follows the same optimized operation scheme.The Z city's North and Yaozhanbu waterworks model input values for the flow of water,the pressure value,the output value of eight pressure point pressure value,with 10 days of data as the training sample,11 th day data as the test sample and forecast validation sample,macroscopic model analysis shows that the established network has high precision,can accurately simulate the network running the actual condition.The two-stage optimal scheduling model of water supply system is established,and the first-stage optimal scheduling model is established with the lowest water supply cost as the objective function.The two-stage optimal scheduling model is established with the minimum operating power of water pump as the objective function,and the same method is used to transform the model into a non-constrained problem and solve it with the help of genetic algorithm.Taking the water supply system of Z city as an example,the effectiveness of the two-stage optimal dispatching model is verified.The results show that the operation power consumption of the optimized water supply system is reduced and the outlet pressure is reduced during the partial optimal dispatching period.The optimal dispatching model has a good energy-saving optimization effect and water supply security stability.