Surface Water Quality Automatic Monitoring System Design,

Designing an advanced "surface water quality automatic monitoring system" with technology of network, environmental monitoring, automatic control and computer is demand of modern environmental information management, and it is also an important component part of the "digital Zhejiang environmental protection". This paper researched and designed "surface water quality automatic monitoring system" based on project of environmental monitoring central station of Zhejiang province.This paper starts with describing the present situation of water monitoring and importance of water automatic monitoring, and then fished the surface water quality automatic monitoring system integration and software design. This system is based on actual demand of "environmental information management system" and "surface water quality automatic monitoring system design" which is Zhejiang province's project. The paper designed the system with new features of systematism, networking, intelligentization and standardization, and this project is well applied in some important district border reaches of water automatic monitoring station such as Qiantang river station. In correctness, functionality, dependability, security and efficiency , the water automatic monitoring software we designed in this paper passed software evaluation by electronic products inspection offices of Zhejiang province.This paper systematically introduces principles of water indexes detection, design philosophy and structure of surface water quality automatic monitoring system. We describes in detail function, method and test course of local workstation software and long-range city workstation software. For the purpose of having a clear knowledge of the prospective changing conditions of water quality and modeling water quality indexes prediction, after analyzing and comparing prediction methods including time series, grey system theory and neutral network, this paper choosed BP neutral network optimized with Levenberg-Marguardt algorithm to prediction water quality. Water quality index of time series is taken as an instructive sample. Taking nine indexes such as total phosphor (TP) .total nitrogen (TN) , chemical oxygen demand (COD) as the predicted indexes, samples were modeled and optimized with BP network. The predicted results indicates that prediction of water quality is precise and fast ,and the BPnetwork can be satisfactorily applied to predict water quality index and fit for pre-alarm of water quality trend. '...