Gray Neural Network Model Is In Quantitative Remote Sensing Inversion Of The Weihe River Water Quality In Applied Research

At present, the problem on inland water pollution of China is very serious. It is well-known that inland water quality has a direct impact on national economic development and people's production and life. Therefore, it is significant to monitor the water quality and get to know the trend of the water quality change for water resource management and environment protection, especially, water environment protection. With the development of remote sensing technology and fine-scale of remote sensing images, the research of using remote sensing technology for water quality monitoring become more and more important. Remote sensing based on water quality monitoring has the advantages of large scale monitoring area, economical, fast and dynamic processing, and easy operation for a long monitoring period. And therefore, monitoring of inland water quality becomes one of the focuses.Weihe River is located in the Guanzhong area of Shaanxi. Xi'an, Xianyang, Weinan, Baoji and other major cities lie in this basin, which is the political, financial, cultural, economic and information center of Shaanxi Province. It has centralized industry, developed agriculture and various tourism resources. And the weihe River is the main river which provides water source. In this area, however, with the development of the economy, the explosion of the population and the growing use of the industrial and domestic water, Weihe River are polluted badly. As water pollution keeps increasing, especially the organic pollution growing, the water now becomes extremely different from that in the past. Therefore, monitoring the water quality has become a significant long-term task.The main contents of this paper:First:Introduced the basic principles and research situation of remote sensing inversing water quality parameters, and commonly used remote sensing data and method of monitoring.Second:Introduces the BP neural network and its application in water quality monitoring, also describes the basic principle and application of the gray theory.Third:Analysis the feasibility of the gray model on the Weihe River water data expansion, and building process, and precision test and residual model. Describes the construction process of BP neural network combine gray expanding model.Fourth:Combination of the BP neural network and gray model described the theoretical basis and analysis of their combined strengths, and description of the method in combination, described the mathematical process of BP neural network mapping in time response function, GM(1,1) for the secondary, constructed gray BP neural network model by BP neural network prediction as the main forecast.Fifth:This thesis, using the Weihe River in shaanxi province as a case study, and referencing on stations of the monitoring data. The paper selects CODmn,COD,NH3-N,DO to inversing. Experimental results show:Grey BP Neural Network Model used in the Weihe River water quality information inversion has the best results.