论文部分内容阅读
为了寻找合适的空中交通流量预测方法,在综合回归预测方法和人工神经网络预测方法优点的基础上,提出采用组合预测方法的思想,并基于多元线性回归模型确定组合方法的权重系数。以北京管制区大王庄导航台流量预测为实例,分析结果表明:组合预测方法对实际流量有好的拟合度,能提高人工神经网络的泛化能力,并减小预测的误差,即总体上不仅优于传统的回归预测方法,也优于单独的人工神经网络预测方法。组合方法为空中交通流量的预测提供了一种可靠而有效的新途径。
In order to find a suitable air traffic flow forecasting method, based on the combination of the regression forecasting method and the artificial neural network forecasting method, the idea of the combined forecasting method is put forward and the weighting coefficient of the combined method is determined based on the multiple linear regression model. The case of Da Wangzhuang navigation station in Beijing controlled area is taken as an example. The results show that the combined forecasting method has a good fitting degree to the actual flow, which can improve the generalization ability of artificial neural network and reduce the prediction error, Not only better than the traditional regression prediction method, but also superior to the single artificial neural network prediction method. The combined approach provides a reliable and effective new approach to the prediction of air traffic flow.