基于多元统计分析的航空发动机故障诊断研究
|
摘要
航空发动机是飞机的重要组成部分,它的微小故障也会直接导致飞机重大事故的发生。由于航空发动机具有结构复杂和工作条件恶劣等特点,所以航空发动机的故障诊断是目前研究的重点之一。本文以涡扇发动机的气路故障为研究目标,利用多元统计分析及其改进算法得到一个综合性能参数来表征单元体的性能特征用来分析整体性能状态,对发动机状态实时监测。当故障发生时,利用独立元作为训练样本建立最小二乘支持向量机模型定位故障,为实现单元体的视情维修提供了理论依据。
主要研究内容有以下三个方面。首先,研究了基于基本多元统计分析的发动机故障检测,验证了多元统计分析方法中主元分析(PCA)和核主元分析(KPCA)的应用可行性,并提出基于独立元分析(ICA)的航空发动机故障检测方法。其次,提出了基于多尺度独立元分析(MSICA)的航空发动机故障检测方法,将小波分析和ICA结合起来应用到航空发动机故障诊断领域中。最后,提出了基于ICA-LSSVM方法的航空发动机故障定位,该方法将独立元分析与最小二乘支持向量机相结合,实现了将航空发动机故障定位到各个单元体。仿真实验结果表明:基于ICA的航空发动机故障检测方法是简单有效的;MSICA方法可以进一步提高发动机故障检测系统发现较小幅度异常的能力;基于ICA-LSSVM的故障定位是可行的,比传统方法速度快,准确率高。
The aero-engine, whose slight fault will cause fatal accident directly, is important component of airplane. Moreover, the aero-engine’s structure is complex and works in harsh environment, so the research of aero-engine fault diagnosis is a focal point. In this thesis, Multivariate statistical analysis is used to obtain an overall performance parameter in order to monitor the turbofan engine’s gas path. When fault occurs, Least Squares Support Vector Machine (LSSVM) is used to implement fault accommodation, which provides the theory to component’s condition based maintenance.
The main contents are as follows. First, Aeroengine fault diagnosis based on multivariate statistical analysis is studied. Principal Component Analysis (PCA), Kernel Principal Components Analysis (KPCA) and Independent Component Analysis (ICA) are validated. Second, Multi-scale Independent Component Analysis (MSICA), which combines wavelet transform with ICA to detect aeroengine path fault, is introduced into fault diagnosis. Third, ICA-LSSVM is introduced into aeroengine fault accommodation. ICA and Least Squares Support Vector Machine (LSSVM) is combined to identify fault component. Simulation results show that ICA is effective, MSICA methods can detect slight fault further, and Fault accommodation method base on ICA-LSSVM is feasible.
主要研究内容有以下三个方面。首先,研究了基于基本多元统计分析的发动机故障检测,验证了多元统计分析方法中主元分析(PCA)和核主元分析(KPCA)的应用可行性,并提出基于独立元分析(ICA)的航空发动机故障检测方法。其次,提出了基于多尺度独立元分析(MSICA)的航空发动机故障检测方法,将小波分析和ICA结合起来应用到航空发动机故障诊断领域中。最后,提出了基于ICA-LSSVM方法的航空发动机故障定位,该方法将独立元分析与最小二乘支持向量机相结合,实现了将航空发动机故障定位到各个单元体。仿真实验结果表明:基于ICA的航空发动机故障检测方法是简单有效的;MSICA方法可以进一步提高发动机故障检测系统发现较小幅度异常的能力;基于ICA-LSSVM的故障定位是可行的,比传统方法速度快,准确率高。
The aero-engine, whose slight fault will cause fatal accident directly, is important component of airplane. Moreover, the aero-engine’s structure is complex and works in harsh environment, so the research of aero-engine fault diagnosis is a focal point. In this thesis, Multivariate statistical analysis is used to obtain an overall performance parameter in order to monitor the turbofan engine’s gas path. When fault occurs, Least Squares Support Vector Machine (LSSVM) is used to implement fault accommodation, which provides the theory to component’s condition based maintenance.
The main contents are as follows. First, Aeroengine fault diagnosis based on multivariate statistical analysis is studied. Principal Component Analysis (PCA), Kernel Principal Components Analysis (KPCA) and Independent Component Analysis (ICA) are validated. Second, Multi-scale Independent Component Analysis (MSICA), which combines wavelet transform with ICA to detect aeroengine path fault, is introduced into fault diagnosis. Third, ICA-LSSVM is introduced into aeroengine fault accommodation. ICA and Least Squares Support Vector Machine (LSSVM) is combined to identify fault component. Simulation results show that ICA is effective, MSICA methods can detect slight fault further, and Fault accommodation method base on ICA-LSSVM is feasible.
引文
[1] Chiang L H,Russell E L,Braatz R D.Fault detection and diagnosis in industrial system [M].London:Springer-Verlag,2001
[2]李强,民航发动机健康管理技术与方法研究[D].南京航空航天大学,2008
[3] Urban L A. Gas Path Analysis App lied to Turbine Engine Condition Monitoring[R]. AIAA,1972:1072-1082
[4] Willsky A. S.A Survey of Design Methods for Fault Detection in Dynamics Systems[J]. Automatica,1976, 12(6):601-611
[5] Stamatis A, Mathioudakis K, Smith M, et al. Gas turbine component fault identification by means of daptive performance modeling[C]. ASME 1990-GT-376
[6] Zedda M. and Singh R. Gas turbine engine and sensor fault diagnosis using optimization techniques. AIAA-99-31290.
[7] Takahisa Kobayashi.Donald.L.Simon Integration of On-Line and Off-Line Diagnostic Algorithms for Aircraft Engine Health Management NASA/TM 2007-214980
[8] Dietz W E, Kiech E L, and Ali M. Jet and rocket engine fault diagnosis in real time [J]. Neural Network Compt, 1989, 1(1):5-18
[9] E. Kopytov, V. Labendik, A. Osis, A. Tarasov,Aircraft engine on-board diagnostics based on neural network[J],2006,3 (60):24-27
[10] Dong-Hyuck Seo, Tae-Seong Roh and Dong-Whan Choi,Defect diagnostics of gas turbine engine using hybrid SVM-ANN with module system in off-design condition[J],Journal of Mechanical Science and Technology,2009, 23 :677-685
[11]鲁峰,黄金泉,陈煜.航空发动机部件性能故障融合诊断方法研究[J].航空动力学报, 2009, 24(7):1649-1653
[12]程嗣怡,索中英,吴华,张官荣,钟秋.基于协调近似表示空间的航空发动机故障诊断[J] ,航空动力学报,2009,24(7):1644-1648
[13]徐可君,江龙平.基于Lyapunov指数谱的航空发动机故障诊断研究[J].应用力学学报,2006,23(3):488-492
[14]杨建平,黄洪钟,苗强,范显峰,王贵宝.基于证据理论的航空发动机早期故障诊断方法[J],航空动力学报,2008, 23(12) :2327-2331
[15]胡金海,骆广琦,李应红,汪诚,尉询凯.一种基于指数损失函数的多类分类AdaBoost算法及其应用[J].航空学报,2008,29(4):811-816
[16]陈果,宋兰琪,陈立波,张占纲.基于粗糙集理论的航空发动机滑油光谱诊断专家系统知识获取方法研究[J].机械科学与技术, 2007. 26 (7):897-501
[17]郝红勋.人工神经网络在航空发动机故障诊断中的应用[D].中国民用航空学院,2006
[18]陈果,李成刚,王德发.利用神经网络规则提取方法获取转静碰摩故障诊断知识[J].航空学报。2008,29(5):1319-1325
[19]赵世荣,黄向华.应用神经网络信息融合诊断航空发动机故障[J].航空动力学报, 2008,23(1):163-168
[20]杨海龙,孙健国.粒子群优化的粗糙集-神经网络在航空发动机故障诊断中的应用[J].航空动力学报,2009,24(2):458-464
[21]费逸伟,张冬梅,姜旭峰.油液监测技术及其在航空发动机故障诊断中的应用[J].航空发动机,2004,30(3):45-48
[22]陈果.航空发动机磨损故障的智能融合诊断[J].中国机械工程,2005,16(4):299-306
[23] Tom Jackson, Jim Austin,Martyn Fletcher,Mark Jessop.The Distributed Aircraft Maintenance[EB/OL]. 2008, http://en.scientificcommons.org/jim_austin
[24] Amar Kumar, Amiya Nayak, Sensor System for Crack Initiation and Crack Growth Monitoring in Aeroengine Components[J], Electrical and Computer Engineering, 2007,80(4): 1452-1455
[25] Kumar, A, Srivastava, A, Goel, N,Narasimhan, V,Nayak, A.An approach to structural health assessment and management technology[J], ICIT 2009 , 2009,2(10):1- 6
[26]张周锁,基于支持向量机的智能诊断技术及应用研究[D].西安交通大学,2004
[27]蔡开龙,谢寿生,杨伟。基于改进LS-SVM的航空发动机传感器故障诊断与自适应重构控制[J].航空动力学报。2008, 23(6):1118-1126
[28]欧阳成丽.基于主成分分析的航空发动机单元体性能辨识[D].中国民用航空学院,2004
[29]杨帆,胡金海,陈卫,张进,蔡开龙.主元分析方法在航空发动机故障检测与诊断中的应用[J].机械科学与技术,2008,27(3):132-135
[30]胡金海,谢寿生,陈卫,侯胜利,蔡开龙.基于核函数主元分析的航空发动机故障检测方法[J].推进技术, 2008,129(1):79-83
[31]杨海龙,基于智能技术的航空发动机气路故障诊断[D].南京航空航天大学,2008
[32]夏飞,基于MATLAB/SIMULINK的航空发动机建模与仿真研究[D].南京航空航天大学.2007
[33]黄永安,马路,刘慧敏.精通MATLAB7.0/Simulink6.0建模仿真开发与高级工程应用[M].清华大学出版社.2005
[34]骆广琦,桑增产,王如根,高坤华.航空燃气涡轮发动机数值仿真[M].国防工业出版社.2007
[35] http://baike.baidu.com/view/2305174.html(OL)
[36]袁春飞,姚华.基于卡尔曼滤波器和遗传算法的航空发动机性能诊断[J].推进技术,2007,28(1):9-13
[37]肖洪,薛倩,廉筱纯,吴虎.基于部件特性未知的航空发动机故障诊断技术[J].2005, 20(5):747-750
[38]郭明.基于数据驱动的流程工业性能检测与故障诊断研究[D].浙江大学.2004, 1
[39] Sch?lkopf B, Smola A, Muller K R. Norlinear component analysis as kernel eigenvalue probiem, Neural Computation, 1998, 10(5):1299-1319
[40]杨英华,吴英华,陈晓波,秦树凯.基于独立源分析的过程监测及故障诊断方法[J].系统仿真学报,2006,18(11):3220-3223
[41] Bakshi R B. Multiscale PCA with Application to Multivariate Statistical Process Monitoring [J]. AIChE Journal, 1998, 44(7): 1596-1610
[42]王乐,顾学迈.基于小波变换和ICA的卫星测控信号盲识别算法[J],南京航空航天大学学报,2009,41(1):59-63
[43] Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano, Shoko Araki, and Shoji Makino, Multistage ICA for blind source separation of real acoustic convolutive mixture [J], 4th International Symposium on Indeoendent Component Analysis and Blind Signal Separation, 2003, 4:523-528
[44]徐涛,王祈.基于MSPCA的传感器故障诊断与数据重构[J],计算机工程与应用,2008,44(11):168-170
[45]柳守斌,朱颖,刘宗田,基于MSD-ICA-EA的轴承故障诊断方法[J],轴承,2008,1(1):33-36
[46]蔡艳宁,胡昌华,汪洪桥,张琪.基于自适应动态无偏LSSVM的故障在线监测[J].系统仿真学报,2009,21(13):4129-4134
[47] S. M. Lee, W. J. Choi, T. S. Roh and D. W. Choi,A Study on Separate Learning Algorithm Using Support Vector Machine for Defect Diagnostics of Gas Turbine Engine, Journal of Mechanical Scienceand Technology,2008, 22 (12): 2489-2497.
[2]李强,民航发动机健康管理技术与方法研究[D].南京航空航天大学,2008
[3] Urban L A. Gas Path Analysis App lied to Turbine Engine Condition Monitoring[R]. AIAA,1972:1072-1082
[4] Willsky A. S.A Survey of Design Methods for Fault Detection in Dynamics Systems[J]. Automatica,1976, 12(6):601-611
[5] Stamatis A, Mathioudakis K, Smith M, et al. Gas turbine component fault identification by means of daptive performance modeling[C]. ASME 1990-GT-376
[6] Zedda M. and Singh R. Gas turbine engine and sensor fault diagnosis using optimization techniques. AIAA-99-31290.
[7] Takahisa Kobayashi.Donald.L.Simon Integration of On-Line and Off-Line Diagnostic Algorithms for Aircraft Engine Health Management NASA/TM 2007-214980
[8] Dietz W E, Kiech E L, and Ali M. Jet and rocket engine fault diagnosis in real time [J]. Neural Network Compt, 1989, 1(1):5-18
[9] E. Kopytov, V. Labendik, A. Osis, A. Tarasov,Aircraft engine on-board diagnostics based on neural network[J],2006,3 (60):24-27
[10] Dong-Hyuck Seo, Tae-Seong Roh and Dong-Whan Choi,Defect diagnostics of gas turbine engine using hybrid SVM-ANN with module system in off-design condition[J],Journal of Mechanical Science and Technology,2009, 23 :677-685
[11]鲁峰,黄金泉,陈煜.航空发动机部件性能故障融合诊断方法研究[J].航空动力学报, 2009, 24(7):1649-1653
[12]程嗣怡,索中英,吴华,张官荣,钟秋.基于协调近似表示空间的航空发动机故障诊断[J] ,航空动力学报,2009,24(7):1644-1648
[13]徐可君,江龙平.基于Lyapunov指数谱的航空发动机故障诊断研究[J].应用力学学报,2006,23(3):488-492
[14]杨建平,黄洪钟,苗强,范显峰,王贵宝.基于证据理论的航空发动机早期故障诊断方法[J],航空动力学报,2008, 23(12) :2327-2331
[15]胡金海,骆广琦,李应红,汪诚,尉询凯.一种基于指数损失函数的多类分类AdaBoost算法及其应用[J].航空学报,2008,29(4):811-816
[16]陈果,宋兰琪,陈立波,张占纲.基于粗糙集理论的航空发动机滑油光谱诊断专家系统知识获取方法研究[J].机械科学与技术, 2007. 26 (7):897-501
[17]郝红勋.人工神经网络在航空发动机故障诊断中的应用[D].中国民用航空学院,2006
[18]陈果,李成刚,王德发.利用神经网络规则提取方法获取转静碰摩故障诊断知识[J].航空学报。2008,29(5):1319-1325
[19]赵世荣,黄向华.应用神经网络信息融合诊断航空发动机故障[J].航空动力学报, 2008,23(1):163-168
[20]杨海龙,孙健国.粒子群优化的粗糙集-神经网络在航空发动机故障诊断中的应用[J].航空动力学报,2009,24(2):458-464
[21]费逸伟,张冬梅,姜旭峰.油液监测技术及其在航空发动机故障诊断中的应用[J].航空发动机,2004,30(3):45-48
[22]陈果.航空发动机磨损故障的智能融合诊断[J].中国机械工程,2005,16(4):299-306
[23] Tom Jackson, Jim Austin,Martyn Fletcher,Mark Jessop.The Distributed Aircraft Maintenance[EB/OL]. 2008, http://en.scientificcommons.org/jim_austin
[24] Amar Kumar, Amiya Nayak, Sensor System for Crack Initiation and Crack Growth Monitoring in Aeroengine Components[J], Electrical and Computer Engineering, 2007,80(4): 1452-1455
[25] Kumar, A, Srivastava, A, Goel, N,Narasimhan, V,Nayak, A.An approach to structural health assessment and management technology[J], ICIT 2009 , 2009,2(10):1- 6
[26]张周锁,基于支持向量机的智能诊断技术及应用研究[D].西安交通大学,2004
[27]蔡开龙,谢寿生,杨伟。基于改进LS-SVM的航空发动机传感器故障诊断与自适应重构控制[J].航空动力学报。2008, 23(6):1118-1126
[28]欧阳成丽.基于主成分分析的航空发动机单元体性能辨识[D].中国民用航空学院,2004
[29]杨帆,胡金海,陈卫,张进,蔡开龙.主元分析方法在航空发动机故障检测与诊断中的应用[J].机械科学与技术,2008,27(3):132-135
[30]胡金海,谢寿生,陈卫,侯胜利,蔡开龙.基于核函数主元分析的航空发动机故障检测方法[J].推进技术, 2008,129(1):79-83
[31]杨海龙,基于智能技术的航空发动机气路故障诊断[D].南京航空航天大学,2008
[32]夏飞,基于MATLAB/SIMULINK的航空发动机建模与仿真研究[D].南京航空航天大学.2007
[33]黄永安,马路,刘慧敏.精通MATLAB7.0/Simulink6.0建模仿真开发与高级工程应用[M].清华大学出版社.2005
[34]骆广琦,桑增产,王如根,高坤华.航空燃气涡轮发动机数值仿真[M].国防工业出版社.2007
[35] http://baike.baidu.com/view/2305174.html(OL)
[36]袁春飞,姚华.基于卡尔曼滤波器和遗传算法的航空发动机性能诊断[J].推进技术,2007,28(1):9-13
[37]肖洪,薛倩,廉筱纯,吴虎.基于部件特性未知的航空发动机故障诊断技术[J].2005, 20(5):747-750
[38]郭明.基于数据驱动的流程工业性能检测与故障诊断研究[D].浙江大学.2004, 1
[39] Sch?lkopf B, Smola A, Muller K R. Norlinear component analysis as kernel eigenvalue probiem, Neural Computation, 1998, 10(5):1299-1319
[40]杨英华,吴英华,陈晓波,秦树凯.基于独立源分析的过程监测及故障诊断方法[J].系统仿真学报,2006,18(11):3220-3223
[41] Bakshi R B. Multiscale PCA with Application to Multivariate Statistical Process Monitoring [J]. AIChE Journal, 1998, 44(7): 1596-1610
[42]王乐,顾学迈.基于小波变换和ICA的卫星测控信号盲识别算法[J],南京航空航天大学学报,2009,41(1):59-63
[43] Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano, Shoko Araki, and Shoji Makino, Multistage ICA for blind source separation of real acoustic convolutive mixture [J], 4th International Symposium on Indeoendent Component Analysis and Blind Signal Separation, 2003, 4:523-528
[44]徐涛,王祈.基于MSPCA的传感器故障诊断与数据重构[J],计算机工程与应用,2008,44(11):168-170
[45]柳守斌,朱颖,刘宗田,基于MSD-ICA-EA的轴承故障诊断方法[J],轴承,2008,1(1):33-36
[46]蔡艳宁,胡昌华,汪洪桥,张琪.基于自适应动态无偏LSSVM的故障在线监测[J].系统仿真学报,2009,21(13):4129-4134
[47] S. M. Lee, W. J. Choi, T. S. Roh and D. W. Choi,A Study on Separate Learning Algorithm Using Support Vector Machine for Defect Diagnostics of Gas Turbine Engine, Journal of Mechanical Scienceand Technology,2008, 22 (12): 2489-2497.