基于到达时刻的航班滑行时间预测方法研究

发布时间：2018-01-03 02:14

本文关键词：基于到达时刻的航班滑行时间预测方法研究　出处：《中国民航大学》2017年硕士论文　论文类型：学位论文

【摘要】：对航空器滑行时间作出准确预测,是降低延误时间、减少延误波及、提升机场场面交通容量的手段之一。本文所研究的到达时刻航班的滑行时间预测,对其后续的航班保障服务时间安排、离港航班时刻计划等都有着重要的参考意义。本文通过蚁群算法(ACA:ant colony algorithm)设计到港航班的滑行路由,并对蚁群算法作出自适应性改进以改善所设计的路由,利用此算法采集到相关数据,采用非参数回归模型预测该滑行路径上后续滑行节点的时间。首先对到港航班的调度流程进行阐述,分析了影响滑行时间的三个重要因素:滑行道容量、滑行路径设计和场面冲突。并建立到港航班场面滑行的数学仿真模型,为后续的路由设计和时间预测研究奠定基础。其次,选中某大型枢纽机场的部分区域抽象为带链路与节点的网络拓扑图,关键的滑行节点与停机位用不同的符号标识。以此为依托,引用ACA初步设计到港航班的滑行轨迹。对ACA中信息素的更新规则作出自适应性改善,即为自适应蚁群算法(Adaptive ACA),重新设计抵达航班的滑行轨迹。采集智能算法得到的时间信息,与场面监控的数据一起处理后,建立非参数回归模型的数据库,按照算法模型预测到港航空器的后续滑行时间。最后,按照停机位分配计划,对比智能算法与依据经验安排的滑行轨迹仿真结果,从迭代过程、滑行距离、滑行时间与等待时间等方面分别作出分析。对比分析两种方法的平均相对误差,评估时间预测的结果。
[Abstract]:To make an accurate prediction of aircraft taxiing time is to reduce delay time, reduce the delay spread, one of the means to enhance the capacity of the airport surface traffic prediction. The arrival time of flight taxiing time, arrangements for the subsequent flight security Business Hours, outgoing class time program have important reference significance. This paper ant colony algorithm (ACA:ant colony algorithm) taxi route design flights to Hong Kong, and the ant colony algorithm for improving the adaptability to improve from the designed route, using this algorithm to collect relevant data, using the glide path prediction time subsequent sliding node nonparametric regression model. Firstly, the scheduling process of flights to Hong Kong this analysis of the three important factors that influence the sliding time: taxiway capacity, glide path design and scene conflict. And the establishment of class field surface sliding to port number Simulation model, lay the foundation for the prediction of subsequent design and routing time. Secondly, the selection of a large regional hub airport is abstracted as a network topology with link and node, node and key taxi stands with different symbol. On this basis, guiding sliding track preliminary design of flights to Hong Kong ACA the improvement of the rules. By adaptive pheromone updating ACA, namely adaptive ant colony algorithm (Adaptive ACA), sliding track re design. Flight arrival time information acquisition of intelligent algorithm, and the scene monitoring data processing together after the establishment of nonparametric regression model database, according to the algorithm model to predict the subsequent the sliding time of Hong Kong aircraft. Finally, according to the gate assignment plan, comparison of intelligent algorithm and on the basis of experience for the sliding track of the simulation results from the iterative process, sliding distance, sliding. The average relative error of the two methods is compared and analyzed, and the results of the time prediction are evaluated.

【学位授予单位】：中国民航大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：V355;TP18

【参考文献】