Combined Application Of Principal Component Analysis And-BP Neural Network In Water Quality Evaluation Of Zhang He

The Zhanghe River is a tributary of the south canal of Haihe River system in north China.It is an important water system in Handan,which is related to the lifeblood of people's livelihood and economic development.But in recent years,with the growing of Handan's population and the rapid development of the industry lead to grow the pollution of water quality and reduce the river flow,it has severely affected the residents' daily life and economic development along the coast.In this paper,we select MaTian,Three provinces bridge,Hezhang,Guangtai,Liu Guzhuang as monitoring cross sections.The DO in the water,potassium permanganate index,BOD,ammonia nitrogen,total phosphorus,total nitrogen,copper,zinc,fluoride,selenium,arsenic are 11 kinds monitoring of water quality indexes.We analysis the data combined with principal component analysis and neural network,to get accurate results of water quality evaluation for specific target water quality status,to analysis the reason of water quality changing.The water environment protection was building a scientific decision-making.Combining with the landform,water conservancy engineering,hydrological conditions,etc.,the main pollutants was network building and simulation research.We analysis the main components of river pollution forecast monitoring of water quality classification,and further research on pollution.The main conclusions of the study are as follows:(1)MaTian,Three provinces bridge,Hetan,Guangtai,hetan,Liu Guzhuang are monitoring cross sections.The DO in the water,potassium permanganate index,BOD,ammonia nitrogen,total phosphorus,total nitrogen,copper,zinc,fluoride,selenium,arsenic have been analyzed with principal component.The evaluation results are in turn.According to the five sites of water quality data processing results,the main water quality are the DO,permanganate index,total nitrogen,copper,zinc,selenium.Therefore the water quality can be initially determined by organic pollutant and heavy metal pollution is more serious.(2)According to building BP neural network model with matlab software,we conduct simulation study on five monitoring sections of the Matian,Three provinces bridge,Hetan,Guantai and Liujiazhuang.Network based on the five section after the principal component analysis of six monitoring indicators were analyzed,and the DO,permanganate index,total nitrogen,copper,zinc,selenium six indicators monitoring data of water quality evaluation results in matlab network simulation,finally it is concluded that five monitoring cross sections of water quality evaluation results.It can be seen from the analysis of network results that the upstream pollution of Zhanghe River is lighter and the water quality is better.The climate drying of the zhanghe river basin and the decrease of river flow are the objective reasons for the pollution of the water quality of Zhanghe River.The large amount of water used in industry and agriculture is also the main cause of water pollution in the Zhanghe River.Through the analysis of the simulation results of the water quality of the Zhanghe River by principal component analysis and BP neural network model,the water pollution situation in the area was studied and evaluated,and the water quality categories of the five monitoring sections were predicted to provide research methods for pollution control of the Zhanghe River.And the basis for the development of Zhanghe River also provides a reference.