基于自组织神经网络的空中交通复杂度参数优化及预测

Air Traffic Complexity Parameter Optimization and Prediction Based on Self-organizing Neural Network

首先介绍了国内外关于空中交通复杂度的定义及研究现状。然后在现有的空中交通复杂度模型中挑选空中交通复杂度参数,再利用自组织神经网络,分析所选取的空中交通复杂度参数之间的关系,达到将高维空中交通复杂度参数进行降维分析的目的。然后对于优化了的空中交通复杂度参数使用RBF神经网络进行预测,比较预测结果,得到最优的空中交通复杂度参数。

Firstly this paper introduces domestic and international air traffic complexity about the definition and the present research status. Then air traffic complexity parameters are selected in the existing air traffic complexity models. By using the self-organizing neural network, the authors analyze the relationship between the selected air traffic complexity parameters to achieve the purpose of reducing the high-dimensional air traffic complexity parameters. For optimization of the air traffic complexity parameters, they use RBF neural network to predict and compare the prediction results, so that the optimal air traffic complexity parameters are achieved.