期刊文献+

基于因子分析和K-means聚类的空中交通复杂性评价 被引量:9

Evaluation of Air Traffic Complexity Based on Factor Analysis and K-means Clustering
下载PDF
导出
摘要 针对航空器服务架次不能精确刻画空中交通复杂程度的现状,提出一种通过多指标度量空中交通复杂性的方法。首先通过实测雷达数据计算得出9个空中交通复杂性评价指标值,并对复杂性指标进行了相关性分析及因子分析的适用性检验;采用因子分析方法最大程度地消除了指标间的相关性,并从多个复杂性评价指标中提取了空中交通总量和空中交通密度2个因子;基于所提取的因子,建立了空中交通复杂性综合评价函数,并利用K-means聚类方法将空中交通复杂程度归为5类,最后通过时段流量和实测陆空通话数据进行了验证。结果表明,当空中交通复杂程度分别为低、高时,15min内的航空器数量分别为10,24架,陆空通话时长分别为315s,636s,对应的通话饱和度分别为35%,70%.随着空中交通复杂性等级的提高,时段流量和通话饱和度增加。 Aiming at the current situation that the aircraft quantities served by the air traffic controller can not accurately depict air traffic complexity,a multi index measure method was proposed.First,9typical complexity evaluation indexes were calculated on the basis of the radar data.The relationships among various air traffic evaluation indexes were studied and the applicability test of factor analysis was done.Then by using factor analysis method the correlations among the indexes were eliminated.The evaluation factors of total air traffic quantity and air traffic density were extracted from multiple ones.On the basis of the extracted factors a comprehensive evaluation function of air traffic complexity was established.By using the K-means clustering method,air traffic complexity was divided into five types.Finally these five types of air traffic complexity was verified by the time flow and the measured air-ground data.When the air traffic complexity was low or high,the number of aircraft served by controller was 10 or 24,the time length of communication was 315 or 636seconds,and the call saturation was 35% or70%,respectively,within 15 minutes.The results show that with the increase of air traffic complexity,the time flow and call saturation increase.
出处 《太原理工大学学报》 CAS 北大核心 2016年第3期384-388 404,共6页 Journal of Taiyuan University of Technology
基金 国家自然科学基金委员会与中国民用航空局联合资助项目:基于复杂网络的空中交通复杂性演化机理与控制策略研究(U1333108) 天津市应用基础与前沿技术研究计划:空中交通冲突风险传播机理研究(14JCQNJC04500)
关键词 空中交通 交通复杂性 因子分析 K-MEANS聚类 相关性 air traffic traffic complexity factor analysis K-means clustering correlation
  • 相关文献

参考文献9

二级参考文献107

共引文献481

同被引文献56

引证文献9

二级引证文献52

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部