The Research Of The Key Technology Of Embedded Environmental Monitoring System And Its Application

With the development of society, environmental protection has become the focal point that the world pays close attention to. Environmental pollution predictions give early pollution warnings, and nip the pollution in the bud, so the precaution and governing can get in progress as early as possible.Support Vector Machines (SVMs) is a machine learning method based on Statistical Learning Theory. SVMs is characterized with high generalization ability, and can solve many practical problems such as small samples, over learning, high dimension and local minima. Therefore, SVMs presents a feasible method to solve the problems in pollution prediction technique.In this dissertation, the author proposed a pollution air oriented prediction model, and based on the model, developed a set of embedded Continuous Emission Monitoring System (CEMS) with the pollution prediction function. The main work done by this thesis includes:With the in-depth study of a typical kind of online support vector regression algorithm, this thesis proposed a new prediction model, which solved the problems existed in the dealing with the critical samples. The simulation results show that proposed algorithm performs high modeling precision, and training speed is increased remarkably compared with the aforementioned algorithm. The improved online SVM regression algorithm can be more effectively applied to environmental prediction.In order to solve the problems that exit in the present CEMS such as lack of real-time performance and Stability, strict working environment demands and short of interaction capability, the author developed a set of CEMS based on Windows CE. This system provides the functions such as data collection, data processing, operation control and the communication with multiple methods, and has performed very well in the practical running.