-
摘要:
航空安全精确预测对预防事故意义重大。目前航空安全预测主要是确定性预测,忽略了各类不确定性对预测影响。在确定性预测基础上,考虑误差不确定性开展航空安全预测,通过非参数方法获得航空安全预测误差不确定性描述,基于最高密度域求解一定可靠程度航空安全预测值最可能落入区间,量化不确定性引起航空安全预测结果的变动,从而确定该区间包含航空安全预测值的可靠程度,更好地认识被预测量在未来变化中可能存在的不确定性和面临的风险。以某航空公司1994—2015年航空安全数据为例,采用所提方法对航空安全开展预测,结果表明,所提方法能提供航空安全预测值及其更精确的不确定性变化范围,更有利于从不确定性角度对航空安全进行分析,解释航空安全预测结果的可能性水平,能为航空安全预警和管理提供理论依据。
Abstract:Accurate aviation safety prediction is of great significance for preventing accidents. At present, aviation safety prediction is mostly deterministic prediction, which ignores the influence of various uncertainties on the prediction results. Based on the traditional deterministic prediction of aviation safety point, this paper presents the prediction of aviation safety interval considering the uncertainty of error. First, through the description method of nonparametric uncertainties, aviation safety prediction error probability density function is derived. Then, the highest density domain method is applied for the most likely future value interval under a certain reliability of aviation safety, and quantitative uncertainty factors cause changes in the aviation safety prediction results. This method aims at determining that the area contains aviation safety forecast reliability, and better understanding the uncertainty and risk of those being predicted in the future change. Taking aviation safety data of civil aviation of an airline from 1994 to 2015 as an example, we predict the aviation safety using aviation safety interval prediction. The results show that the proposed method can provide aviation safety prediction curve and more accurate variation range of uncertainty, which is more conducive to modeling uncertainty of aviation safety and explaining the possibility level of aviation safety prediction results, which can provide theoretical basis for aviation safety early warning and management.
-
Key words:
- aviation safety prediction /
- uncertainty /
- error distribution /
- kernel density /
- interval estimation
-
-
[1] 端木京顺, 马震宇, 常洪, 等.航空事故预测预警预防理论方法[M].北京:国防工业出版社, 2013.DUANMU J S, MA Z Y, CHANG H, et al.Theoretical methods of the aviation accident prediction, early-warning and prevention[M].Beijing:National Defense Industry Press, 2013(in Chinese). [2] 罗帆, 贾刚.航空公司组织管理预警指标及其因子分析[J].武汉理工大学学报(信息与管理工程版), 2006, 28(10):93-100. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whqcgydxxb200610026LUO F, JIA G.Factor analysis of early warning indexes for airlines organizational management[J].Journal of Wuhan University of Technology(Information & Management Engineering), 2006, 28(10):93-100(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whqcgydxxb200610026 [3] 李柯, 张胜.空中交通管理中心安全预警管理信息系统研究[J].武汉理工大学学报(信息与管理工程版), 2007, 29(8):106-110. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whqcgydxxb200708028LI K, ZHANG S.Management information system of safety early warning for an air traffic control center[J].Journal of Wuhan University of Technology (Information & Management Engineering), 2007, 29(8):106-110(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whqcgydxxb200708028 [4] 王本涛.机场飞行区地面服务人员人为差错预警模型研究[D].武汉: 武汉理工大学, 2014. http://cdmd.cnki.com.cn/Article/CDMD-10497-1015001118.htmWANG B T.Study on the early warning model for ground service personnel human error of airport airfield[D].Wuhan: Wuhan University of Technology, 2014(in Chinese). http://cdmd.cnki.com.cn/Article/CDMD-10497-1015001118.htm [5] UDDIN RAJIB M S, FAN L.A study on the critical factors of human error in civil aviation:An early warning management perspective in Bangladesh[J].Management Science Letters, 2015, 5:21-28. doi: 10.5267/j.msl.2014.12.011 [6] 杜毅.民用航空事故征候万时率的回归分析[J].中国民航安全学报, 2010, 21(1):42-44. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgmhfxxyxb201001008DU Y.Regression analysis of ten thousand hour rate of civil aviation accident symptoms[J].Journal of China Civil Aviation Flying College, 2010, 21(1):42-44(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgmhfxxyxb201001008 [7] 杜红兵, 秦鹏慧.美国通用航空飞行事故短期预测的时序外推分析模型[J].安全与环境学报, 2011, 11(3):208-211. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=aqyhjxb201103051DU H B, QIN P H.Time sequence extrapolative analysis model of short-term forecasting of U.S.general aviation accidents[J].Journal of Safety and Environment, 2011, 11(3):208-211(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=aqyhjxb201103051 [8] 王衍洋, 曹义华.民航安全指数结果分析与预测[J].北京航空航天大学学报, 2011, 37(10):1223-1228. https://bhxb.buaa.edu.cn/CN/CNKI:11-2625/V.20111013.1727.015WANG Y Y, CAO Y H.Results analysis and prediction on civil aviation safety indices[J].Journal of Beijing University of Aeronautics and Astronautics, 2011, 37(10):1223-1228(in Chinese). https://bhxb.buaa.edu.cn/CN/CNKI:11-2625/V.20111013.1727.015 [9] 丁松滨, 王飞.基于BP神经网络的民航安全预测方法研究[J].中国民航学院学报, 2006, 24(1):53-56. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgmhxyxb200601015DING S B, WANG F.Study on civil aviation safety forecasting method based on BP neural network[J].Journal of Civil Aviation University of China, 2006, 24(1):53-56(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgmhxyxb200601015 [10] 甘旭升, 端木京顺.航空装备事故的灰色时序组合预测模型[J].中国安全科学报, 2012, 22(4):32-38. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgaqkxxb201204006GAN X S, DUANMU J S.Grey and time series combination prediction model for aviation equipment accident[J].China Safety Science Journal, 2012, 22(4):32-38(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zgaqkxxb201204006 [11] HUAN J S.A quantitative model for aviation safety risk assessment[J].Computers and Industrial Engineering, 2008, 54(1):34-44. doi: 10.1016/j.cie.2007.06.032 [12] 吕雪梅.航空公司不安全事件的预测研究[D].天津: 中国民航大学, 2008. http://cdmd.cnki.com.cn/Article/CDMD-10059-2008159509.htmLV X M.Studies on the forecast of unsafe events for airlines[D].Tianjin: Civil Aviation University of China, 2008. http://cdmd.cnki.com.cn/Article/CDMD-10059-2008159509.htm [13] 刘斌, 周京阳, 周海明, 等.一种改进的风电功率预测误差分布模型[J].华东电力, 2012, 40(2):286-291. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=hddl201202038LIU B, ZHOU J Y, ZHOU H M, et al.An improved model for wind power forecast error distribution[J].East China Electric Power, 2012, 40(2):286-291(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=hddl201202038 [14] 南晓强, 李群湛.考虑风功率预测误差分布的储能功率与容量配置法[J].电力自动化设备, 2013, 33(11):117-122. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlzdhsb201311021NAN X Q, LI Q Z.Energy storage power and capacity allocation based on wind power forecasting error distribution[J].Electric Power Automation Equipment, 2013, 33(11):117-122(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlzdhsb201311021 [15] 吴雄, 王秀丽.新能源组合出力的置信区间估计[J].电力系统自动化, 2013, 37(16):7-12. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxtzdh201316002WU X, WANG X L.Estimation on confidence intervals of combined output of renewable energy[J].Automation of Electric Power Systems, 2013, 37(16):7-12(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxtzdh201316002 [16] 徐曼, 乔颖, 鲁宗相.短期风电功率预测误差综合评价方法[J].电力系统自动化, 2011, 35(12):20-26. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxtzdh201112004XU M, QIAO Y, LU Z X.A comprehensive error evaluation method for short-term wind power prediction[J].Automation of Electric Power Systems, 2011, 35(12):20-26(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxtzdh201112004 [17] HYNDMAN R J.Computing and graphing highest density regions[J].The American Statistician, 1994, 50(2):120-126. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=10.1080/00031305.1996.10474359 [18] 章国勇, 伍永刚, 张洋, 等.一种风电功率混沌时间序列概率区间简易预测模型[J].物理学报, 2014, 63(13):422-430. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=wlxb201413059ZHANG G Y, WU Y G, ZHANG Y, et al.A simple model for probabilistic interval forecasts of wind power chaotic time series[J].Acta Physica Sinica, 2014, 63(13):422-430(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=wlxb201413059 [19] 唐兴芸, 罗明燕.非参数核密度估计核函数的最优选择[J].黔南民族师范学院学报, 2019, 39(4):10-14. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=qnmzsfxyxb201904003TANG X Y, LUO M Y.Optimal selection of kernel function for nonparametric kernel density estimation[J].Journal of Qiannan Normal University for Nationalities, 2019, 39(4):10-14(in Chinese). http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=qnmzsfxyxb201904003 [20] BOTEV Z I.Kernel density estimation using MATLAB[EB/OL].(2019-01-20)[2019-10-06].http://www.mathworks.us/matlabcentral/fileexchange/authors/27236. [21] KIM J H, SILVAPULLE P, HYNDMAN R J.Half-life estimation based on the bias-corrected bootstrap:A highest density region approach[J].Computational Statistics and Data Analysis, 2007, 51(7):3418-3432. doi: 10.1016/j.csda.2006.07.005 [22] CONEJO A J, PLAZAS M A.Day-ahead electricity price forecasting using the wavelet transform and ARIMA models[J].IEEE Transactions on Power Systems, 2005, 20(2):1035-1042. doi: 10.1109/TPWRS.2005.846054