基于神经网络的飞机着陆速度预测模型

Prediction Model of Aircraft Landing Speed Based on Neural Network

空中交通需求的增长迫切需要在繁忙的终端区准确地建立连续降落飞机间的间隔,因此需要准确地预测着陆速度,以提高机场吞吐量,同时避免飞机间的间隔过小带来的碰撞危险。论文利用获取的机载飞行信息和机场监视信息,利用神经网络建立飞机组之间后机的进近着陆速度预测模型,模拟后机进近到着陆跑道的最终进近速度剖面,使用确定R^(2)系数和均方根误差的拟合优度进行评估,在成都双流机场的数据条件下,模型准确预测了着陆速度,96%的情况下,弱阵风和强阵风条件下的误差幅度分别为12.2%和11.9%。通过结果得出,这两种阵风条件下的着陆速度预测的不确定性至少降低了9.35%。

With the growth of air traffic demand,it is urgent to accurately establish the spacing between aircraft in busy termi⁃nal area,so it is necessary to accurately predict the landing speed,so as to improve the airport throughput and avoid the collision risk caused by too small spacing between aircraft.In this paper,using the acquired airborne flight information and airport monitor⁃ing information,the neural network is used to establish the prediction model of the approach and landing speed of the rear aircraft between aircraft groups.The final approach speed profile of the rear aircraft approaching to the landing runway is simulated and de⁃termined by using the neural network.Under the data condition of Chengdu Shuangliu airport,the model accurately predictes the landing speed.Under the condition of 96%,the error ranges of weak gust and gust are 12.2%and 11.9%respectively.The results show that the uncertainty of landing speed prediction under these two gust conditions is reduced by at least 9.35%.