Water Information Management System And Decision Making Platform Based On Data Mining

Surface water is an indispensable natural resource for the survival and development and in all kinds of production of human beings.It is an important part of the geological environment system.The content and quality of the production of human activities and daily life are closely related with the surface water.With the emphasis on surface water quality in China,water quality monitoring equipment is constantly upgrading and monitoring accuracy is improving,the ministry of environmental protection has accumulated a lot of data,which provide a basis for data mining of water quality information.The total amount of surface water resources in our country is among the highest in the world,but it has the characteristics of uneven distribution of water resources and very low per capita content.At the same time,surface water resources in most parts of China are facing the threat of pollution and gradually restrict the economic development,affecting Residents' daily life.In order to describe the water quality information more accurately,predict the future water quality according to the fluctuation of historical water quality information,prevent and control water pollution more effectively and make theoretical support for government decision-making.This paper builds a water management information management system based on data mining Decision-making platform to predict the surface water quality at the entrance of Miyun Reservoir in Beijing.This paper improves the linear water quality classification method.In view of the common water quality classification method can only reflect the degree of water quality pollution,but can not reflect the lack of content of each factor affecting water pollution indicators,a new water quality classification method is constructed,the main idea is Combining the respective applicability of K-means and SVM algorithms,these two classification models are combined to make a more accurate classification of water quality.In addition,this paper changes the current common method of using a single model for water quality prediction based on the linear confidence and nonlinear information contained in the time series of water quality information,using ARIMA model for linear prediction,the difference between the linear prediction results and the actual water quality data Value as a nonlinear part of water quality information,using SVM to predict and superpose the prediction results of the two models to improve the prediction accuracy and provide theoretical support for the government to make corresponding pollution prevention and control measures.