航空学报 > 2016, Vol. 37 Issue (11): 3328-3339   doi: 10.7527/S1000-6893.2016.0086

结冰条件下人-机-环系统的飞行风险概率

薛源, 徐浩军, 胡孟权   

  1. 空军工程大学 航空航天工程学院, 西安 710038
  • 收稿日期:2016-01-03 修回日期:2016-03-16 出版日期:2016-11-15 发布日期:2016-05-04
  • 通讯作者: 徐浩军,男,硕士,教授,博士生导师。主要研究方向:飞行安全与作战效能。Tel.:029-84787637,E-mail:xuhaojun@xjtu.edu.cn E-mail:xuhaojun@xjtu.edu.cn
  • 作者简介:薛源,男,博士,讲师。主要研究方向:飞行仿真与飞行安全。Tel.:029-84787637,E-mail:xuhaojun@xjtu.edu.cn;徐浩军,男,硕士,教授,博士生导师。主要研究方向:飞行安全与作战效能。Tel.:029-84787637,E-mail:xuhaojun@xjtu.edu.cn
  • 基金资助:

    国家“973”计划(2015CB755802);国家自然科学基金(61503406)

Flight risk probability of pilot-aircraft-environment system under icing conditions

XUE Yuan, XU Haojun, HU Mengquan   

  1. Aeronautics and Astronautics Engineering College, Air Force Engineering University, Xi'an 710038, China
  • Received:2016-01-03 Revised:2016-03-16 Online:2016-11-15 Published:2016-05-04
  • Supported by:

    National Basic Research Program of China (2015CB755802); National Natural Science Foundation of China (61503406)

摘要:

以结冰条件下的飞行风险量化概率为研究对象,基于蒙特卡罗飞行仿真实验对结冰条件下人-机-环系统的耦合特性进行了分析,并获取了飞行参数极值样本。构建了飞行风险发生的判定条件;对飞行参数极值样本进行了统计特性分析,验证了其厚尾分布特征。一维分布类型辨识结果表明广义极值分布对相对速度和迎角极值的描述精度最高。为描述二维变量对相关性的各自影响程度,提出了一种新的双参数变权重Copula模型;辨识结果表明该Copula模型能以较高的精度通过假设检验。相关性分析的结果表明相对速度和迎角同时出现极大值和极小值的概率较大。基于二维极值样本的Copula分布模型求出了不同结冰程度下的飞行风险概率值,探讨了飞行风险的非线性增长趋势。

关键词: 蒙特卡罗仿真, 结冰条件, 人-机-环系统, 飞行风险, 极值分布, Copula模型

Abstract:

The quantitative flight risk probability under icing conditions is set as the research object. Based on Monte Carlo flight simulation experiment, coupling characteristics of pilot-aircraft-environment system after icing are studied, and the extreme flight parameters are extracted. The judgement conditions for flight risk are given. And statistical properties of the extreme samples are analyzed. The fact that extreme flight parameters have heavy tail distribution characteristics is verified. Identification results indicate that generalized extreme value model can best describe the distribution characteristics of relative velocity and angle of attack. In order to describe the influence level of each two-dimensional parameter on the correlations, a new Copula model that has two changeable weights is proposed; identification result shows that this new Copula model can pass the hypothesis testing with high accuracy. Correlation analysis results reveal that relative velocity and angle of attack would both show up extreme values with greater probability. Flight risk probabilities under different icing levels are calculated based on Copula distribution model with two-dimensional extreme values. The nonlinear growth trend of flight risk is also discussed.

Key words: Monte Carlo simulation, icing conditions, pilot-aircraft-environment system, flight risk, extreme value distribution, Copula model

中图分类号: