Design And Implementation Of Visual Analysis System For Water Quality Testing Data In Jinan

With the explosive development of the Internet of Things industry,computer technology has been widely used in various industries of water,and a large amount of historical data has been collected and stored.However,how to use these long-sleeved data for effective analysis,as a reference for water industry decision makers to make decisions,and to provide a platform to understand water quality and participate in supervision,These issues have become the focus of Smart Water's entry into the era of big data.Based on the relevant requirements of Smart Water,Jinan Water Quality Testing Center has focused on promoting Jinan water quality monitoring data analysis and other related work.Over the past 20 years,we have collected a large number of historical data on water quality monitoring and real-time monitoring data of single indicators.It is the focus of this system to store these data in a timely manner,conduct reasonable data ETL and effective mining analysis,manage water affairs in a more precise and dynamic manner,and provide timely and effective decision-making suggestions for water supply departments and water quality monitoring centers.The system described in this paper is a water quality comprehensive analysis and evaluation system based on actual needs,which integrates water quality data collection,data ETL,multidimensional retrieval,multidimensional statistical analysis,water quality assessment and online analysis.The monthly data of water quality monitoring and the online real-time detection data of single indicators are sorted out,and multiple sample categories are divided.According to different detection properties,water quality data is processed,inquired and analyzed in many aspects.In the data collection module,on the one hand,the data in the relational database is transmitted to the Hive repository of the Hadoop cluster,and the user-specified relational database manual collection and the system setting MySQL database timing transmission are realized;on the other hand,all Historical data and data from enterprise real-time databases are delivered in real-time to the distributed search engine ElasticSearch(ES),providing a consistent and efficient interface for data extraction.In terms of data processing,the raw data is mostly recorded by humans,and there are a large number of invalid,non-standard and descriptive data,and the data is standardized by means of regular expressions and the like;In the aspect of data retrieval,multiple search conditions such as multi-index arbitrary combination search,fuzzy query,sorting according to specified attributes,full-text paging,etc.are provided;In terms of statistical analysis,a series of statistical values such as the maximum and minimum values,the mean value,the mode,the detection rate,the pass rate,the number of tests,and the number of times exceeding the standard of the sample type or the detection property are provided.Provide more intuitive data control;In the aspect of water quality assessment,a water network chemical stability judgment index model,a surface water single index grade analysis model,a health risk comprehensive index,a N1 index,a taste index,a sewage biodegradability analysis model,and a single factor evaluation model were arranged.The principal component analysis model and the BP neural network model comprehensively evaluate the water quality monitoring data of Jinan,and quantify the scores to obtain the evaluation results;In the real-time data analysis,according to the N1 index,the taste index,the health risk comprehensive index and the comprehensive evaluation of water quality,the water quality indicators change trend and water quality information are displayed in real time according to the location information in the data,and the water quality of the specific location can be inquired according to the demand.Happening.This system is a server-side framework established by SpringBoot technology.Based on the front-end framework built by vue-cli,the front and back ends are completely separated,which makes the logic of each part more clear and collaborative work more convenient.All functions of the system are based on B/S mode.Some algorithms are implemented by Java in R language.They rely on open source mature visualization library Echart.js and R analysis mining graphics library to realize various visualizationsThe system realized in this paper provides a water monitoring data analysis system with friendly interface and simple operation for Jinan Water Quality Testing Center.The model used in this system is an evaluation model obtained by a large number of research and analysis by the relevant industry personnel in the world.In the comprehensive evaluation of water quality,the training of model parameters is also strictly required to verify that the accuracy of the set reaches three decimal places.The system provides users with online water quality assessment,real-time understanding of drinking water quality;it also provides a good performance evaluation platform for the water department to help them make better decisions and solve the urgent need for water quality monitoring data analysis.