Prediction Of Civil Aviation Passenger Traffic Volume On Grey Theory And RBF Neural Network

Accurate consequence of the civil aviation passenger traffic volume prediction is one of the major basis for civil air traffic planning and management. This paper aims to researching on the characteristics of prediction, the paper finds out the way to increase accuracy of prediction.This paper firstly studies china's civil aviation passenger traffic volume trends. What's more, the article also analyzes the impact of civil aviation passenger traffic-related factors from both macroscopic and microcosmic factor. On the basis of this, this paper summarizes the characteristic of civil aviation passenger traffic volume prediction, and analyzes its forecasting methods. Then, this paper compares and analyses civil aviation passenger traffic volume prediction methods.Secondly grey theory has obvious advantages in dealing with inadequate information and uncertain problems and is suitable for civil aviation passenger traffic data characteristics. This paper forecasts civil aviation passenger traffic volume with grey theoretical model. That model can not be ignored some of the shortcomings, making prediction error relatively larger. After that, for RBF neural network can approximate any continuous function with any precision. Because of strong adapting ability and learning ability, it is very suitable for the prediction of non-linear system. The paper uses RBF neural network to civil aviation passenger traffic prediction. Due to small sample and rough original data, it makes prediction error relatively larger.Then, the paper presents grey theory and RBF neural network model's advantages and disadvantages and optimizes civil aviation passenger traffic prediction model, a grey-RBF neural network prediction model. On this basis, the paper works out the idea and steps of how the model constructed; analyzes the key problems like pretreatment of input/output data in the process when constructs model, spread and prediction for test results and the provides effective solution.Finally, the paper takes civil aviation passenger traffic volume data , for instance in the three models. The calculation proves that civil aviation passenger traffic volume prediction based on the G-RBFN win more accuracy in the fitting of historical data.