The Study On The Grey Neural Network Model And Its Application In The Prediction Of Air Pollution

This paper applies grey neural network into the field of prediction of air pollution, in order to integrate the grey prediction method with the neural networks method by a valid path to constitute the model of the prediction of air pollution based on the grey neural networks. Lastly, according to the《The Research on the Control and Layout of Air Pollution of the Shenzhen City》studied by Chinese research Academy of Environmental Sciences, we do some analysis and validation about the results above by using the monitor data of contamination concentration of the Shenzhen city.The tenet of establishing the grey system model: the anomalous primal sequence is changed into the regular sequence which according to we set up grey model that be used to forecast the primal sequence. The foundation of grey model need fewer samples that without better distributing order, but also do less calculating quantity. BP network adopts the back propagation when it learns samples, so it has ability of parallel calculation,distributed information memory,admitting-error, and so on. In this paper, we combine grey prediction method with the neural networks method to found the grey neural network model shortly named GNNM. Expressly speaking, we map the function responding to the time on the grey differential equation into the BP network which includes the grey parameters. Then, we train the BP network with the simples that distilled from the known data in the grey system to ascertain the grey parameters. In this way, the parameters are being adjusted by learning, which improves the precision of forecast, but also make use of the parameters in the grey differential equation to find the original weight value and threshold value that are very closed to the exact value, so we can overcome the randomicity about the original value in the BP network, reduce the times of network training.In this article, both the improvement to grey neural network and its application on the prediction of air pollution have been done.1,Based on the theory of both grey prediction model and neural network, GNNM(1, 1) that is the grey neural network model of first order is improved and perfected.(1)Make use of the parameters attained by least squares method in the grey differential equation and correct the original value of grey model, and then find the...