| Visualization is the theory,method and technique of converting data into intuitive images or graphics and displaying them on a computer screen through computer image processing or graphics technology.In the Internet age,the combination of visualization technology and network technology makes remote visualization services a reality.Visualization technologies mainly include scientific computing visualization,data visualization and information visualization.Geographic information system(GIS)is a visual realization method and is a hot research topic today.The water supply network is the“lifeline”of the water supply system.Due to the dispersal,diversity and complexity of the water supply network,relying solely on manual or traditional computer-aided management forms can no longer meet the actual production management needs of modern urban water supply operations.This topic introduces the visualization technology into the management of urban water supply pipe network,and studies the key technologies of pipe network information visualization,data analysis and data visualization based on WebGIS open source framework Cesium,which realizes the perfect combination of electronic map and pipe network data.The pipe network object is directly queried and analyzed on the map interface,and the water supply load time series prediction model provides reference and decision support for water consumption analysis.Finally,the design and implementation of urban water supply pipe network visualization management system was carried out in the case of Shaanxi Fugu,and the feasibility and practicability of the method were verified.The specific research work of this paper is as follows:1)The visualization technology of water supply pipe network based on WebGIS is studied,including map registration technology based on four-parameter calculation,WebGIS technology based on Cesium and its application in visualization of water supply pipe network.The map registration technology based on four-parameter calculation mainly analyzes several coordinate systems and their differences,three-dimensional four-parameter model,and BJ-54 to GCJ-02 coordinates,which realizes the drawing of the example pipe network diagram and completes the water supply network.Information visualization management.2)Researched data analysis and data visualization technologies,including web crawler technology,heterogeneous database data synchronization technology based on distributed task queue,and data visualization chart technology based on ECharts.Through the research of heterogeneous database data synchronization technology,the real-time monitoring data of water supply is synchronized,and the statistical chart of dynamic generation of multiple types and multiple dimensions is realized,and the visual analysis and management of water supply data is completed.3)The water supply load prediction model is studied,and the law and significance of water supply load are analyzed.An effective water supply load prediction model is established through theoretical analysis and experimental verification.The verification results show that the NAR network prediction model based on Bayesian regularization has higher prediction of short-term water supply load than the traditional time series method,standard BP neural network and Bayesian regularized BP neural network.The adaptability and prediction accuracy can accurately reflect the short-term changes of the water supply load.The multivariate LSTM cyclic neural network in deep learning can consider the influencing factors such as weather,temperature,holidays and long-term changes,analyze the overall logic of the sequence and the complex time correlation,and improve the medium and long-term prediction accuracy of the water supply load.4)Based on the research of the above key technologies,this paper takes the construction of the management system of the Fugu water supply pipe network in Shaanxi as an example,and integrates the WebGIS open source framework Cesium with the JavaEE framework Spring,SpringMVC,Mybatis and other frameworks,and uses Bootstrap,AngularJS,etc.The popular Web development technology realizes the responsive layout and MVC design mode.The Keras deep learning framework and Matlab are used to experiment and analyze the water supply load prediction algorithm,thus completing the design of four modules of WebGIS,water supply management,data management and water supply load prediction.The development of the visual management system for urban water supply pipeline network has been realized. |
