空域运行态势评价指标体系优化研究  

Research on optimization of airspace operational situation evaluation indicator system

作　　者：隋东[1] 李倩 SUI Dong;LI Qian(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学民航学院,南京211106

出　　处：《安全与环境学报》2023年第2期351-358,共8页Journal of Safety and Environment

摘　　要：空域运行态势评估是衡量空中交通运行状况的关键手段。为构建科学合理的评价标准,从空域结构、交通运行、飞行冲突及气象条件4个维度确定了初始评价指标体系。采用灰色关联分析方法和病态指数循环分析方法,从信息重要度和重叠度两个方面优化评价指标体系。依据交通量统计和流量控制信息定义了自由态和饱和态两种数据标签,并基于K-Means聚类方法进行了实证分析,结果表明优化后的评价指标体系对不同扇区标签样本的聚类准确率高达90%以上,可为客观衡量空域运行态势提供技术支持。Ensuring the safety of airspace operations is the core of air traffic management. As the airspace operational situation is affected by many complex factors, a scientific and comprehensive evaluation standard is an important prerequisite for accurate evaluation. This paper aims to build a scientific and reasonable evaluation indicator system and improve the deficiencies in the existing research. Firstly, based on previous studies, this paper comprehensively analyzes the influencing factors of airspace operation, and selects indicators from the four dimensions of airspace structure, traffic operation, flight conflict, and meteorological conditions, thereby establishing an initial airspace operation situation evaluation indicator system. Secondly, to reduce the mutual influence relationship between the indicators and ensure the rationality of the indicator system, the grey relational analysis method and the circle ill-condition index analysis method are adopted to optimize the evaluation indicator system from the two aspects of information importance and overlap. It aims to screen out the indicators that have a significant impact on the evaluation results and reflect a low degree of overlap of information. Finally, a label determination method based on operating data is proposed. According to traffic statistics and flow control information, the free state and the saturated state of the airspace situation are defined, and the saturation degree of the two states is used as the label of the indicator data to obtain the calibration sample set. The K-Means cluster analysis is used to compare the evaluation ability of the evaluation indicator system before and after the optimization and verify the effectiveness of the optimization method in this paper. The experimental results show that the optimized evaluation system can achieve higher clustering accuracy. In addition, the clustering accuracy of labeled samples in different sectors can reach more than 90%, which can be used as a benchmark for airspace situation assessmen

关 键 词：安全社会工程 空域运行态势 指标体系 定量筛选 K-MEANS聚类 

分 类 号：X92[环境科学与工程—安全科学]