面向多跑道机场航班延误恢复的机场噪声预测研究

发布时间：2018-06-22 02:54

本文选题：航班延误恢复 + 多跑道机场　；参考：《中国民航大学》2016年硕士论文

【摘要】：近年来,随着民航客货运输量的迅猛增长,国内许多机场通过增加跑道数量、扩建机场规模来增加机场容量,在提高了吞吐量的同时也造成了航班使用跑道的不确定因素增加和机场噪声影响问题的加剧。多跑道机场数量与日俱增,航班、跑道、飞行程序之间的组合更加多样化,噪声影响分布不同,导致多跑道机场的噪声预测问题与以往的预测方式不一样且更加复杂。目前的预测方法主要面向单跑道单次飞行事件的噪声预测问题,而机场噪声具有声级高,影响范围广,不稳定的特点,对于航班延误恢复时段大量航班密集起飞产生的“短时高噪”现象,使用已有的预测方法将会产生较大误差,难以适用。因此,进行航班延误恢复时段的多跑道机场噪声预测研究很有必要。现有的预测方法主要面向单跑道的噪声预测问题,对于目前因机场新建、扩建、跑道日益增加和航班延误恢复时段航班密集起飞而导致噪声影响加剧的问题,本文提出了“噪声等效”的概念,基于一种带约束条件的混合聚类算法,构建了一种基于机场噪声等效的航班聚类模型,实现了航班跑道的等效匹配,减小了多跑道机场航班延误恢复时段航班使用跑道不确定的因素。在此基础上,通过BP(Back Propagation)神经网络和NPD(Noise-Power-Distance)曲线插值的方法分别构造了机场噪声等效预测模型,预测出一定时段的噪声平均能量,为行业主管部门提供依据和参考。最后,通过理论及实验证明,该预测模型把噪声预测的不确定性因素抵消后大幅度提高了预测的准确率,实现了对多跑道机场在航班延误恢复时段的噪声等效预测。机场周围的噪声影响分布主要受飞行航迹控制,噪声大小主要受机型、飞行参数影响。为了更方便快捷的从全局上预测和评估航班延误恢复时段的多跑道机场的噪声影响情况,基于不同跑道进行航迹聚类和机型聚类,把聚类结果的每簇中心航迹和代表机型数据组合导入INM(Integrated Noise Models)计算噪声值构成噪声数据库,通过贝叶斯分类算法构建了一个基于贝叶斯分类的多跑道机场噪声优化预测模型。输入航班号、机型、航迹簇类别、目的地、出港点等基础数据即可快速得到噪声预测结果。实验结果表明,该模型不仅能够预测航班放行正常时段的噪声影响,还可以在一定误差范围内方便快捷地预测出航班延误恢复时段的机场周围敏感点的噪声。
[Abstract]:In recent years, with the rapid growth of passenger and cargo traffic in civil aviation, many domestic airports increase their capacity by increasing the number of runways and expanding the size of airports. It not only improves the throughput, but also results in the increase of the uncertainty of the runway usage and the aggravation of the airport noise problem. With the increasing number of multi-runway airports, the combination of flights, runways and flight procedures is more diversified, and the noise influence distribution is different, resulting in the noise prediction problem of multi-runway airports is different and more complex than the previous prediction methods. The current prediction methods are mainly used to predict the noise of single runway single flight event, and the airport noise is characterized by high sound level, wide range of influence and instability. For the phenomenon of "short time high noise" caused by a large number of flight densities during the flight delay recovery period, using the existing prediction methods will produce a large error, which is difficult to apply. Therefore, it is necessary to study the noise prediction of multi-runway airport in flight delay recovery period. The existing prediction methods are mainly aimed at the noise prediction of single runway. At present, due to the construction and expansion of the airport, the increasing number of runways and the heavy takeoff of flights during the flight delay recovery period, the noise impact is aggravated. In this paper, the concept of "noise equivalence" is proposed. Based on a hybrid clustering algorithm with constraints, a flight clustering model based on airport noise equivalence is constructed, which realizes the equivalent matching of flight runways. Reduces the multiple runway airport flight delay recovery time flight use runway uncertainty factor. On this basis, the equivalent prediction model of airport noise is constructed by BP (back Propagation) neural network and NPD (Noise-Power-Distance) curve interpolation method, and the average noise energy in a certain period of time is predicted, which provides the basis and reference for the competent department of the industry. Finally, the theoretical and experimental results show that the prediction model can significantly improve the accuracy of the prediction by canceling the uncertainty factors of the noise prediction, and realizes the noise equivalent prediction of multi-runway airports in the period of flight delay recovery. The influence distribution of the noise around the airport is mainly controlled by the flight path, and the noise is mainly affected by the model and the flight parameters. In order to predict and evaluate the noise impact of multi-runway airports in the time of flight delay recovery more conveniently and quickly, track clustering and model clustering are carried out based on different runways. The data of each cluster center track and representative model of clustering results are imported into INM (Integrated noise models) to calculate the noise value to form noise database, and a Bayesian classification algorithm is used to construct an optimal prediction model of multi-runway airport noise based on Bayesian classification. The noise prediction results can be obtained quickly by inputting basic data such as flight number, aircraft type, track cluster category, destination, departure point and so on. The experimental results show that the model can not only predict the noise effect during the normal period of flight release, but also predict the noise of the sensitive points around the airport during the flight delay recovery period conveniently and quickly within a certain range of errors.
【学位授予单位】：中国民航大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：V351;TP183

【相似文献】