Design And Research Of Intelligent Water Quality Monitoring Platform

The process of our country's industrialization and urbanization is accelerating,while the pollution of urban water environment is increasing.In order to improve the efficiency of urban water environment supervision,this paper studies and designs a water quality monitoring platform,which is based on B/S architecture.It includes three parts: on-site water quality monitoring device,remote data communication module and remote water quality monitoring client.In order to solve the problem that the BOD is difficult to be measured online by sensors.This paper also proposes a support vector machine regression(SVR)water quality prediction model,which based on genetic algorithm(GA)optimized parameters.By introducing genetic algorithm,the problem of difficult selection of key parameters in the traditional SVR model has been solved.After experiments,it is proved that the GA-SVR prediction model has high application value.In order to meet the needs of the actual environment on site,this paper uses Siemens S7-1200 PLC as the controller of the on-site water quality monitoring equipment.Data transmission between the field device and the cloud server through the 4G industrial router.The on-site monitoring equipment adopts modular design,which has the advantages of easy transportation,assembly and debugging.This paper designs a remote data communication module based on the principle of socket communication.Through this module,TCP communication between the field monitoring equipment and the server is realized.This paper also designs a remote water quality monitoring client,which is based on the Django framework.Through this system,it is possible to monitor and manage the monitoring equipment of water quality monitoring points in different geographic locations,and the staff can view the current data and the historical data at any time.The client system works well,when it is monitoring a large number of water quality monitoring stations.In this paper,an SVR water quality prediction model based on GA optimized parameters is established.After the genetic algorithm is used to automatically optimize key parameters in the support vector machine regression model,a mathematical model of easy-to-measure related parameters and BOD in urban sewage is established.After testing,the average error of the model is 0.009443,and the root mean square error is 16.88 mg / L,which proves that the prediction results of the model have high accuracy.By using this platform,users can timely and accurately grasp the water quality information and equipment status of relevant detection points.At the same time,the GA-SVR water quality prediction model also provides the possibility of introducing machine learning methods in the field of urban water environment governance.