Flight delay prediction has attracted great interest in civil aviation community due to its significant role in airline planning,flight scheduling,airport operation,and passenger service.Flight delay is affected by nu...Flight delay prediction has attracted great interest in civil aviation community due to its significant role in airline planning,flight scheduling,airport operation,and passenger service.Flight delay is affected by numerous factors and irregularly propagates in air transportation networks owing to flight connectivity,which brings critical challenges to accurate flight delay prediction.In recent years,Graph Convolutional Networks(GCNs)have become popular in flight delay prediction due to the advantage in extracting complicated relationships.However,most of the existing GCN-based methods have failed to effectively capture the spatial-temporal information in flight delay prediction.In this paper,a Geographical and Operational Graph Convolutional Network(GOGCN)is proposed for multi-airport flight delay prediction.The GOGCN is a GCN-based spatial-temporal model that improves node feature representation ability with geographical and operational spatial-temporal interactions in a graph.Specifically,an operational aggregator is designed to extract global operational information based on the graph structure,while a geographical aggregator is developed to capture the similar nature among spatially close airports.Extensive experiments on a real-world dataset demonstrate that the proposed approach outperforms the state-of-the-art methods with a satisfying accuracy improvement.展开更多
The rock–paper–scissors （RPS） game is a nice model to study the biodiversity in an ecosystem. However, in the previous studies only the nearest-neighbor interaction among the species was considered. In this paper,...The rock–paper–scissors （RPS） game is a nice model to study the biodiversity in an ecosystem. However, in the previous studies only the nearest-neighbor interaction among the species was considered. In this paper, taking the long-range migration into account, the effects of the interplay between nearest-neighbor-interaction and long-range-interaction given by Levy flight with distance distribution lh （-3 ≤ h 〈-1） in the spatial RPS game are investigated. Taking the probability, exchange rate, and power-law exponent of Levy flight as parameters, the coexistence conditions of three species are given. The critical curves for stable coexistence of three species in the parameter space are presented. It is also found that Levy flight has interesting effects on the final spatiotemporal pattern of the system. The results reveal that the long-range-interaction given by Levy flight exhibits pronounced effects on biodiversity of the ecosystem.展开更多
A systematic methodology including a computational pilot model and a pattern recognition method is presented to identify the boundary of the flight performance margin for quantifying the human factors. The pilot model...A systematic methodology including a computational pilot model and a pattern recognition method is presented to identify the boundary of the flight performance margin for quantifying the human factors. The pilot model is proposed to correlate a set of quantitative human factors which represent the attributes and characteristics of a group of pilots. Three information processing components which are influenced by human factors are modeled： information perception, decision making, and action execution. By treating the human factors as stochastic variables that follow appropriate probability density functions, the effects of human factors on flight performance can be investigated through Monte Carlo（MC） simulation. Kernel density estimation algorithm is selected to find and rank the influential human factors. Subsequently, human factors are quantified through identifying the boundary of the flight performance margin by the k-nearest neighbor（k-NN） classifier. Simulation-based analysis shows that flight performance can be dramatically improved with the quantitative human factors.展开更多
在实际监测任务中,及时有效地识别飞行模式至关重要。然而,现有的飞行模式识别方法主观性强、模式单一,限制了在复杂情况下的飞行监控能力,在实际应用中有局限性,进而导致模式边界定位不精确、识别精度低。为此提出一种基于敏感边界和...在实际监测任务中,及时有效地识别飞行模式至关重要。然而,现有的飞行模式识别方法主观性强、模式单一,限制了在复杂情况下的飞行监控能力,在实际应用中有局限性,进而导致模式边界定位不精确、识别精度低。为此提出一种基于敏感边界和长飞行序列的飞行模式智能识别方法(Intelligent Flight Pattern Recognition Method for Sensitive Boundaries and Long Flight Sequences, IFPRM-SBLFS),以对飞行模式进行智能识别。为了更好地探索多模式飞行参数的空间关系,设计自适应图嵌入,针对不同持续时间的飞行模式提出去噪深度多尺度自动编码器,以及用于减轻模型损失的分类加权焦点损失和回归联合时空交集损失。为验证所提方法的优越性,采集多架民用航班的真实参数,涵盖11种飞行模式,通过人工标注构建飞行模式数据集。仿真计算结果表明:新模型能够在连续飞行架次中自动区分不同的飞行模式,并准确提取模式边界,识别准确率达到了99.07%,且无需任何预处理或后处理;新的智能识别方法可以有效提高精确度和敏感边界的飞行模式识别效果。展开更多
飞行机动动作识别主要用于飞行员训练质量评估、飞行作战时的辅助决策等场景。为实现基于飞行参数的飞行机机动识别,研究了模式袋(bag of patterns,BoP)算法,针对算法在多维时间序列应用中的不足进行了改进,利用改进后的算法对飞行姿态...飞行机动动作识别主要用于飞行员训练质量评估、飞行作战时的辅助决策等场景。为实现基于飞行参数的飞行机机动识别,研究了模式袋(bag of patterns,BoP)算法,针对算法在多维时间序列应用中的不足进行了改进,利用改进后的算法对飞行姿态数据进行特征提取,并进行飞行机动识别分析。识别仿真结果表明,改进后的BoP算法能提高飞行机动识别的准确率和置信度,通过该算法提取的飞行参数特征能更好地表征具体的飞行机动动作。展开更多
Trajectory clustering can identify the flight patterns of the air traffic,which in turn contributes to the airspace planning,air traffic flow management,and flight time estimation.This paper presents a semantic-based ...Trajectory clustering can identify the flight patterns of the air traffic,which in turn contributes to the airspace planning,air traffic flow management,and flight time estimation.This paper presents a semantic-based trajectory clustering method for arrival aircraft via new proposed trajectory representation.The proposed method consists of four significant steps:representing the trajectories,grouping the trajectories based on the new representation,measuring the similarities between different trajectories through dynamic time warping(DTW)in each group,and clustering the trajectories based on k-means and densitybased spatial clustering of applications with noise(DBSCAN).We take the inbound trajectories toward Shanghai Pudong International Airport(ZSPD)to carry out the case studies.The corresponding results indicate that the proposed method could not only distinguish the particular flight patterns,but also improve the performance of flight time estimation.展开更多
基金supported by the National Natural Science Foundation of China(Nos.71731001,U2133210,and U2033215,61822102)。
文摘Flight delay prediction has attracted great interest in civil aviation community due to its significant role in airline planning,flight scheduling,airport operation,and passenger service.Flight delay is affected by numerous factors and irregularly propagates in air transportation networks owing to flight connectivity,which brings critical challenges to accurate flight delay prediction.In recent years,Graph Convolutional Networks(GCNs)have become popular in flight delay prediction due to the advantage in extracting complicated relationships.However,most of the existing GCN-based methods have failed to effectively capture the spatial-temporal information in flight delay prediction.In this paper,a Geographical and Operational Graph Convolutional Network(GOGCN)is proposed for multi-airport flight delay prediction.The GOGCN is a GCN-based spatial-temporal model that improves node feature representation ability with geographical and operational spatial-temporal interactions in a graph.Specifically,an operational aggregator is designed to extract global operational information based on the graph structure,while a geographical aggregator is developed to capture the similar nature among spatially close airports.Extensive experiments on a real-world dataset demonstrate that the proposed approach outperforms the state-of-the-art methods with a satisfying accuracy improvement.
基金Project partially supported by the National Natural Science Foundation of China(Grant Nos.61174150 and 60974084)the Program for New Century Excellent Talents in University of Ministry of Education of China(Grant No.NCET-09-0228)the Fundamental Research Funds for the Central Universities of Beijing Normal University,and the High Performance Computing Center of Beijing Normal University
文摘The rock–paper–scissors （RPS） game is a nice model to study the biodiversity in an ecosystem. However, in the previous studies only the nearest-neighbor interaction among the species was considered. In this paper, taking the long-range migration into account, the effects of the interplay between nearest-neighbor-interaction and long-range-interaction given by Levy flight with distance distribution lh （-3 ≤ h 〈-1） in the spatial RPS game are investigated. Taking the probability, exchange rate, and power-law exponent of Levy flight as parameters, the coexistence conditions of three species are given. The critical curves for stable coexistence of three species in the parameter space are presented. It is also found that Levy flight has interesting effects on the final spatiotemporal pattern of the system. The results reveal that the long-range-interaction given by Levy flight exhibits pronounced effects on biodiversity of the ecosystem.
基金supported by the National Basic Research Program of China(No.2010CB734103)
文摘A systematic methodology including a computational pilot model and a pattern recognition method is presented to identify the boundary of the flight performance margin for quantifying the human factors. The pilot model is proposed to correlate a set of quantitative human factors which represent the attributes and characteristics of a group of pilots. Three information processing components which are influenced by human factors are modeled： information perception, decision making, and action execution. By treating the human factors as stochastic variables that follow appropriate probability density functions, the effects of human factors on flight performance can be investigated through Monte Carlo（MC） simulation. Kernel density estimation algorithm is selected to find and rank the influential human factors. Subsequently, human factors are quantified through identifying the boundary of the flight performance margin by the k-nearest neighbor（k-NN） classifier. Simulation-based analysis shows that flight performance can be dramatically improved with the quantitative human factors.
文摘在实际监测任务中,及时有效地识别飞行模式至关重要。然而,现有的飞行模式识别方法主观性强、模式单一,限制了在复杂情况下的飞行监控能力,在实际应用中有局限性,进而导致模式边界定位不精确、识别精度低。为此提出一种基于敏感边界和长飞行序列的飞行模式智能识别方法(Intelligent Flight Pattern Recognition Method for Sensitive Boundaries and Long Flight Sequences, IFPRM-SBLFS),以对飞行模式进行智能识别。为了更好地探索多模式飞行参数的空间关系,设计自适应图嵌入,针对不同持续时间的飞行模式提出去噪深度多尺度自动编码器,以及用于减轻模型损失的分类加权焦点损失和回归联合时空交集损失。为验证所提方法的优越性,采集多架民用航班的真实参数,涵盖11种飞行模式,通过人工标注构建飞行模式数据集。仿真计算结果表明:新模型能够在连续飞行架次中自动区分不同的飞行模式,并准确提取模式边界,识别准确率达到了99.07%,且无需任何预处理或后处理;新的智能识别方法可以有效提高精确度和敏感边界的飞行模式识别效果。
文摘飞行机动动作识别主要用于飞行员训练质量评估、飞行作战时的辅助决策等场景。为实现基于飞行参数的飞行机机动识别,研究了模式袋(bag of patterns,BoP)算法,针对算法在多维时间序列应用中的不足进行了改进,利用改进后的算法对飞行姿态数据进行特征提取,并进行飞行机动识别分析。识别仿真结果表明,改进后的BoP算法能提高飞行机动识别的准确率和置信度,通过该算法提取的飞行参数特征能更好地表征具体的飞行机动动作。
文摘采用野外定点调查的方法,研究了红火蚁Solenopsis invicta多蚁后型种群婚飞新形成蚁巢的局域空间分布规律.结果表明,短期内婚飞形成的活动蚁巢在局域平面空间上呈均匀分布,分布的基本成分为单个蚁巢,且蚁巢间相互排斥.在平面空间上不同间隔距离间该类型蚁巢半方差值呈明显规律性变化,具有空间相关性.建立了5个球状模型,其变程分别为12.6、14.1、9.7、13.3和14.5 m,平均为12.8 m.
基金supported by the Joint Fund of National Natural Science Foundation of China and Civil Aviation Administration of China(U1933117)the Open Fund for Graduate Innovation Base(Laboratory)of Nanjing University of Aeronautics and Astronautics(kfjj20190709).
文摘Trajectory clustering can identify the flight patterns of the air traffic,which in turn contributes to the airspace planning,air traffic flow management,and flight time estimation.This paper presents a semantic-based trajectory clustering method for arrival aircraft via new proposed trajectory representation.The proposed method consists of four significant steps:representing the trajectories,grouping the trajectories based on the new representation,measuring the similarities between different trajectories through dynamic time warping(DTW)in each group,and clustering the trajectories based on k-means and densitybased spatial clustering of applications with noise(DBSCAN).We take the inbound trajectories toward Shanghai Pudong International Airport(ZSPD)to carry out the case studies.The corresponding results indicate that the proposed method could not only distinguish the particular flight patterns,but also improve the performance of flight time estimation.
