多元线性回归在引气系统故障诊断中的应用

梁坤; 左洪福; 孙见忠; 王容辉

doi:10.13700/j.bh.1001-5965.2014.0633

多元线性回归在引气系统故障诊断中的应用

doi: 10.13700/j.bh.1001-5965.2014.0633

梁坤^1,2,
左洪福^1, ,,
孙见忠¹,
王容辉³

1.
南京航空航天大学民航学院, 南京 210016
2. 淮阴工学院交通学院, 淮安 22300
3. 厦门航空公司, 厦门 361006

基金项目: 国家自然基金委员会与中国民用航空局联合基金(60939003); 国家自然科学基金(61403198); 江苏省青年基金(BK20140827)

详细信息

作者简介:
梁坤(1978—),男,山东济宁人,博士研究生,kunl78024@nuaa.edu.cn

通讯作者:
左洪福(1959—),男,湖南茶陵人,教授,rms@nuaa.edu.cn,主要研究方向为可靠性工程、维修理论、故障诊断与监控等.

中图分类号: V245.4
计量
- 文章访问数: 852
- HTML全文浏览量: 122
- PDF下载量: 580
- 被引次数: 0
出版历程
- 收稿日期: 2014-10-15
- 网络出版日期: 2015-09-20

Application of multiple linear regression to fault diagnosis of bleed air system

1.
College of Aerospace Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
2. College of Transportation, Huaiyin Institute of Technology, Huaian 22300
3. Xiamen Airlines, Xiamen 361006, China

摘要

摘要: 针对利用快速存取记录器(QAR)数据进行民机系统故障诊断问题,以民机引气系统为对象,提出了一种适合多飞行循环数据特点的多元线性回归模型的故障检测方法.首先建立了多飞行循环数据的引气系统性能多元线性回归模型,设计了飞行循环和飞行循环内故障检测方法;然后采用最大后验估计方法进行模型参数估计;最后设计了适合多飞行循环数据的模型参数最大后验估计算法.借助仿真数据和航空公司收集的实际飞行数据对方法进行了验证,结果表明了该方法有效且具有一定工程应用价值.
- 故障诊断 /
- 多元线性回归模型 /
- 快速存取记录器(QAR)数据 /
- 最大后验估计 /
- 引气系统
Abstract: To solve the fault diagnosis problem of civil aircraft systems by utilizing quick access record(QAR) data, with the civil aircraft bleed air system as the research object, we proposed a fault detection method of multiple linear regression model for multi-flight cycle data characteristics. Firstly, the multi-linear regression model of bleed air system performance was established for multi-flight cycles data and the fault detection method of flight cycles and flight cycle's interior was designed. Then the model parameters were estimated by maximum posteriori method. Finally, the maximum posteriori estimation algorithm of the model parameters was designed for multi-flight cycle data. With simulated data and actual flight data collected by airlines, the method was validated. The results show the method's feasibility and application value in engineering practice.
- fault diagnosis /
- multiple linear regression model /
- quick access record (QAR) data /
- maximum posteriori estimation /
- bleed air system

HTML全文

参考文献(25)

[1]	周新颖,谭朝阳,刘倩.挖潜“大数据”时代QAR如何改变飞行运营 [N].中国民航报,2013-10-25(04).Zhou X Y,Tan Z Y,Liu Q,In tapping the potential of “big data” era how exploit QAR to change the flight operations [N].CCAC Journal,2013-10-25(04)(in Chinese).
[2]	祁明亮,邵雪焱,池宏.QAR超限事件飞行操作风险诊断方法[J].北京航空航天大学学报,2011,37(10):1207-1210.Qi M L,Shao X Y,Chi H.Flight operations risk diagnosis method on quick-access-record exceedance[J].Journal of Beijing University of Aeronautics and Astronautics,2011,37(10):1207-1210(in Chinese).
[3]	孔成安,李文华,尹湛.利用QAR数据实施飞机性能监控[J].中国民用航空,2008,94(10):54-56.Kong C A,Li W H,Yin Z.Monitoring aircraft performance with QAR data[J].China Civil Aviation,2008,94(10):54-56(in Chinese).
[4]	耿宏,揭俊.基于QAR数据的飞机巡航段燃油流量回归模型[J].航空发动机,2008,34(4):46-50. Geng H,Jie J.Fuel flow regression model of aircraft cruise based on QAR data[J].Aeroengine,2008,34(4):46-50(in Chinese).
[5]	曹惠玲,贾超.基于QAR的飞机爬升阶段燃油流量回归模型研究[J].中国民航大学学报,2013,31(3):31-35.Cao H L,Jia C.Research of fuel flow regression model of aircraft climb phase based on QAR[J].Journal of Civil Aviation University of China,2013,31(3):31-35(in Chinese).
[6]	王帮峰,林剑祥,芦吉云.基于EEMD—HT的飞行数据小突变信号检测[J].振动、测试与诊断,2013,33(3):388-392.Wang B F,Lin J X,Lu J Y.Small mutation signal detection for flight data based on empirical mode decomposition and Hilbert transform[J].Journal of Vibration,Measurement & Diagnosis,2013,33(3):388-392(in Chinese).
[7]	韩旻,赵清洲.一种基于阴性选择算法的飞行数据异常值检测方法[J]航空计算技术,2010,40(4):53-55.Han M,Zhao Q Z.A detection method of abnormal flight data based on negative selection[J]Aeronautical Computing Technique,2010,40(4):53-55(in Chinese).
[8]	吕永乐,郎荣玲.基于奇异值分解的飞行数据降噪方法[J].计算机工程,2010,38(3):260-262.Lyu Y L,Lang R L.Noise reduction method for flight data based on singular value decomposition[J].Computer Engineering,2010,38(3):260-262(in Chinese).
[9]	Dimogianopoulos D G,Hios J D,Fassois S.Fault detection and isolation in aircraft systems using stochastic nonlinear modelling of flight data dependencies[C]//Proceedings of 14th Mediterranean Conference on Control and Automation.Piscataway,NJ:IEEE Press,2006:1-6.
[10]	Das S,Sarkar S,Ray A,et al.Anomaly detection in flight recorder data:A dynamic data-driven approach[C]//Proceedings of American Control Conference(ACC).Piscataway,NJ:IEEE Press,2013:2668-2673.
[11]	Chu E,Gorinevsky D,Boyd S.Detecting aircraft performance anomalies from cruise flight data[C]//AIAA Infotech Aerospace Conference.Reston:AIAA,2010:2010-0337.
[12]	Gorinevsky D,Matthews B,Martin R.Aircraft anomaly detection using performance models trained on fleet data[C]//Proceedings of Conference on Intelligent Data Understanding (CIDU).Piscataway,NJ:IEEE Press,2012:17-23.
[13]	Chu E,Gorinevsky D,Boyd S P.Scalable statistical monitoring of fleet data[C]//World Congress of the International Federation of Automatic Control (IFAC).Laxenburg:IFAC,2011,18(1):13227-13232.
[14]	Abdelrahman W G,Al-Garni A Z,Al-Wadiee W.Application of back propagation neural network algorithms on modeling failure of B-737 bleed air system valves in desert conditions[J].Applied Mechanics and Materials,2012,225:505-510.
[15]	Shang L,Liu G.Sensor and actuator fault detection and isolation for a high performance aircraft engine bleed air temperature control system[J].IEEE Transactions on Control Systems Technology,2011,19(5):1260-1268.
[16]	Lakkundi A B.Prediction of crack growth in aircraft engine bleed air ducts using finite element and crack growth analysis[D].Wichita:Wichita State University,2006.
[17]	Iverson D L.Inductive system health monitoring[C]//Proceedings of International Conference on Artificial Intelligence (IC-AI).Athens,Georgia:CSREA Press,2004:605-611.
[18]	Gross K C,Singer R M,Wegerich S W,et al.Application of a model-based fault detection system to nuclear plant signals,ANL/RA/CP-92110[R].1997.
[19]	Gorinevsky D.Bayesian fault isolation in multivariate statistical process monitoring dimitry gorinevsky[C]//Proceedings of American Control Conference (ACC).Piscataway,NJ:IEEE Press,2011:1963-1968.
[20]	Wang H,Chai T Y,Ding J L,et al.Data driven fault diagnosis and fault tolerant control:Some advances and possible new directions[J].Acta Automatica Sinica,2009,35(6):739-747.
[21]	Lee S Y,Tsang S Y.Constrained maximum likelihood estimation of two-level covariance structure model via EM-type algorithms[J].Psychometrika,1999,36(4):435-450.
[22]	刘锦萍,郁金祥.基于改进的粒子群算法的多元线性回归模型参数估计[J].计算机工程与科学,2010,(4):101-105.Liu J P,Yu J X.Parameter estimation of multiple linear regression models based on the improved particle swarm optimization algorithm[J].Computer Science & Engineering,2010,(4):101-105(in Chinese).
[23]	Boyd S P,Vandenberghe L.Convex optimization[M].Cambridge:Cambridge University Press,2004:82-93.
[24]	Nesterov Y.Efficiency of coordinate descent methods on huge-scale optimization problems[J].SIAM Journal on Optimization,2012,22(2):341-362.
[25]	Richtárik P,Taká M.Iteration complexity of randomized block-coordinate descent methods for minimizing a composite function[J].Mathematical Programming,2012,144(1-2):1-38.