基于振动测试技术的跨座式单轨交通系统铸钢支座健康监测方法研究
|
摘要
论文以重庆跨座式单轨交通系统铸钢支座为研究对象。铸钢支座是连接轨道梁和墩台的关键受力部件,其健康状况对整个单轨交通系统的安全运行起着至关重要的作用。由于加工缺陷、气候环境对金属元件的腐蚀、长期服役及复杂的受力情况等因素都会对铸钢支座的健康状况产生很大的负面影响。为了单轨交通的安全运营,必须对铸钢支座的健康状况进行监测。然而,铸钢支座一般高架在室外轨道墩台之上,位置特殊,且其体积较大、形状复杂,采用常规的射线检测、超声波检测、微波检测等无损检测方法都难以对其健康状况进行监测。
在国家科技支撑计划课题“跨座式单轨交通装备研发”(2007BAG06B06)的资助下,论文根据课题要求,并结合重庆轨道交通二号线的实际情况,提出采用振动测试技术来监测铸钢支座的健康状况。首先采用工控机、力锤、测振传感器和数据采集卡等,并编写相应的软件,构成适合工程现场实际情况的振动数据采集系统;然后通过多次实验确定测振传感器在铸钢支座上面的布置位置以及激振位置;最后对铸钢支座进行人工力锤击振,并采集铸钢支座的振动脉冲响应信号。
针对铸钢支座系统故障诊断中缺乏故障样本的问题,为了对采集到的铸钢支座的振动脉冲响应信号进行分析,论文首次提出基于一类支持向量机(One-class SVM)的故障诊断方法。该方法只需要正常数据样本就可以建立起单值分类器,把正常样本和非正常样本区分出来。试验采用核主元分析(KPCA)对采集到的振动脉冲响应信号进行分解,将提取到得主元特征输入到One-class SVM分类器进行训练和测试,分类器在适当牺牲冤检率的情况下保证了无漏检情况,诊断系统的准确率为98%。实验结果表明,One-class SVM分类器可以准确识别出故障样本,而采用KPCA分解后提取的特征能有效地浓缩故障信息,使One-class SVM分类器分类效果更好、计算效率更高。
研究和系统使用结果表明本检测系统所采用的方案和方法是可行的和正确的,各项性能和指标均达到了预期的要求。上述方法的提出为铸钢支座的健康检测提供了比较科学的参考依据,具有一定的参考价值。
As the key research object of this paper, the cast steel pedestal is the most important part to connect pier and girder of Chongqing straddle-type monorail. The health condition of which plays vital function to the straddle-type monorail’s safe running. However, due to the machining defects, rust caused by climate factors, long-term service’s influence and complicated effect of time-varying load, the health condition of those cast steel pedestals will be changed. For the safe running of the straddle-type monorail traffic, a monitoring operation must be taken to make sure that all those cast steel pedestals are in normal. The cast steel pedestals are usually fixed on the high piers and girders, whose big cubage, special location and complex shape make those common nondestructive testing ways such as radioactive ray, ultrasonic, microwave can’t detect its health condition.
With financial assistance of the national science and technology support projects: 2007BAG06B06, according to the topic requests and combined actual situations of the second line of Chongqing rail transit, the author presents the vibration testing technology to detection the health condition of the cast steel pedestal. At first, using industry control computer, force hammer, vibration sensor and data acquisition card, and programming the software to form the vibration data acquisition system. Then find the right vibration sensors’position and the hit position according to many times’trials. Then put the sensors on the cast steel pedestal, and exciting the cast steel pedestal with the force hammer, and acquiring the vibration pulse signals.
The scarcity of fault samples often occurs in the fault diagnosis of Chongqing straddle-type monorail cast steel pedestal system. In the case of this situation, this paper first puts forward a fault diagnosis method based on One-class Support Vector Machine (One-class SVM). This method can build up one-class classifier to distinguish between normal condition and abnormal condition as long as the normal data samples are provided. In the process of test, Kernel Principal Component Analysis (KPCA) is used as data preprocessing to extract the features from vibration impulse response signal as the input of One-class SVM classifier and the accuracy rate of classifier is 98%.The test result shows that the feature extraction based on KPCA can concentrate fault information more effectively and make the the One-class SVM classifier identify the fault samples more accurately.
The result of the experiments indicates that the method above providing a scientific way for health diagnosis of the straddle-type monorail cast steel pedestal, which have some value for reference.
在国家科技支撑计划课题“跨座式单轨交通装备研发”(2007BAG06B06)的资助下,论文根据课题要求,并结合重庆轨道交通二号线的实际情况,提出采用振动测试技术来监测铸钢支座的健康状况。首先采用工控机、力锤、测振传感器和数据采集卡等,并编写相应的软件,构成适合工程现场实际情况的振动数据采集系统;然后通过多次实验确定测振传感器在铸钢支座上面的布置位置以及激振位置;最后对铸钢支座进行人工力锤击振,并采集铸钢支座的振动脉冲响应信号。
针对铸钢支座系统故障诊断中缺乏故障样本的问题,为了对采集到的铸钢支座的振动脉冲响应信号进行分析,论文首次提出基于一类支持向量机(One-class SVM)的故障诊断方法。该方法只需要正常数据样本就可以建立起单值分类器,把正常样本和非正常样本区分出来。试验采用核主元分析(KPCA)对采集到的振动脉冲响应信号进行分解,将提取到得主元特征输入到One-class SVM分类器进行训练和测试,分类器在适当牺牲冤检率的情况下保证了无漏检情况,诊断系统的准确率为98%。实验结果表明,One-class SVM分类器可以准确识别出故障样本,而采用KPCA分解后提取的特征能有效地浓缩故障信息,使One-class SVM分类器分类效果更好、计算效率更高。
研究和系统使用结果表明本检测系统所采用的方案和方法是可行的和正确的,各项性能和指标均达到了预期的要求。上述方法的提出为铸钢支座的健康检测提供了比较科学的参考依据,具有一定的参考价值。
As the key research object of this paper, the cast steel pedestal is the most important part to connect pier and girder of Chongqing straddle-type monorail. The health condition of which plays vital function to the straddle-type monorail’s safe running. However, due to the machining defects, rust caused by climate factors, long-term service’s influence and complicated effect of time-varying load, the health condition of those cast steel pedestals will be changed. For the safe running of the straddle-type monorail traffic, a monitoring operation must be taken to make sure that all those cast steel pedestals are in normal. The cast steel pedestals are usually fixed on the high piers and girders, whose big cubage, special location and complex shape make those common nondestructive testing ways such as radioactive ray, ultrasonic, microwave can’t detect its health condition.
With financial assistance of the national science and technology support projects: 2007BAG06B06, according to the topic requests and combined actual situations of the second line of Chongqing rail transit, the author presents the vibration testing technology to detection the health condition of the cast steel pedestal. At first, using industry control computer, force hammer, vibration sensor and data acquisition card, and programming the software to form the vibration data acquisition system. Then find the right vibration sensors’position and the hit position according to many times’trials. Then put the sensors on the cast steel pedestal, and exciting the cast steel pedestal with the force hammer, and acquiring the vibration pulse signals.
The scarcity of fault samples often occurs in the fault diagnosis of Chongqing straddle-type monorail cast steel pedestal system. In the case of this situation, this paper first puts forward a fault diagnosis method based on One-class Support Vector Machine (One-class SVM). This method can build up one-class classifier to distinguish between normal condition and abnormal condition as long as the normal data samples are provided. In the process of test, Kernel Principal Component Analysis (KPCA) is used as data preprocessing to extract the features from vibration impulse response signal as the input of One-class SVM classifier and the accuracy rate of classifier is 98%.The test result shows that the feature extraction based on KPCA can concentrate fault information more effectively and make the the One-class SVM classifier identify the fault samples more accurately.
The result of the experiments indicates that the method above providing a scientific way for health diagnosis of the straddle-type monorail cast steel pedestal, which have some value for reference.
引文
[1]张世洪.跨座式单轨交通系统及其锚固螺杆健康监测现状[J].重庆工学院学报(自然科学版).2007, 21(7): 41-44.
[2]刘家兵.基于超声波的轻轨在役锚固螺杆检测研究[D].重庆大学,2007.
[3]郭子义.跨座式单轨轨道梁在役锚固螺杆超声波信号自动采集方法和系统[D].重庆大学,2008.
[4]于得水.基于振动技术的在役锚固螺杆紧固力矩检测研究[D].重庆大学, 2007.
[5]赛铁兵.跨座式单轨PC轨道梁线形调整施工技术[J].铁道标准设计,2007,(1):40-42.
[6]苏明辉.重庆轻轨较新线PC轨道梁架设技术[J].桥梁, 2005(增):1-4.
[7] Kim J T , Lyon R H. Spectrum analysis as a tool for robust processing, reverberation and detection of transients. Mechanical System and Signal Processing, 1992, 6(1):1~15.
[8] Michael Kuby. Factors influencing light-rail station boar dings in the United States [J] . Transportation Research Part A , 2004 ,38 :223 - 247.
[9] Nigel G Harris. The marginal profitability of train services in London [J] . Transport Policy, 1996, 13(1/ 2):17 - 20.).
[10] Sydney C, Chu K. Crew scheduling of light rail transit in Hong Kong: From modeling to implementation [J]. Computer Ops Res, 1998, 25(11):887 - 894.
[11]马继兵,蒲黔辉,夏招广.跨座式单轨交通PC轨道梁静载及疲劳试验研究[J].铁道建筑. 2008(5):3-6.
[12]韩西,邵毅明,钟厉等.跨座式轻轨PC轨道梁活动钢支座疲劳及寿命有限元分析[J].公路交通科技. 2001, 18(6):47-49.
[13]韩西,钟厉,邵毅明等.轻轨PC梁铸钢支座有限元分析及实验[J].重庆大学学报(自然科学版). 2002, 25(5):112-115.
[14]张贵寿,邵晓蓉.建筑结构的鉴定与加固[J].绍兴文理学报. 2000. 20(5), 82.
[15]王仲生,万小朋.无损检测诊断现场实用技术(第一版)[M].北京:机械工业出版社, 2002: 5-30.
[16]闫桂荣,段忠东,欧进萍.基于结构振动信息的损伤识别研究综述[J].地震工程与工程振动. 2007.
[17]张竞爽.悬臂梁非线性振动分析与实验[D].天津大学. 2009.
[18] Norden Huang, Kang Huang.基于希尔伯特_黄变换的铁路桥梁结构健康监测[J].中国铁道科学. 2006, 27(1):1-5.
[19]杨德健,郭天德.某军用飞机振动模态分析[J].天津城市建设学院学报. 1996, 2(3):1-5.
[20]王朝晖,张来斌,樊长博.基于EMD和HT时频分析方法的滚动轴承故障诊断.石油机械[J]. 2007, 35(5):32-33.
[21]邹春平,陈端石,华宏星.船舶结构振动模态综合[J]. 2003, 37(8):1213-1217.
[22]杨和振,李华军,黄维平.基于振动测试的海洋平台结构无损检测[J]. 2003, 16(4):480-484.
[23]周杏鹏,仇国富,王寿荣等.现代检测技术[M].北京,高等教育出版社. 2004:148.
[24]李德葆,陆秋海.工程振动试验分析[M].清华大学出版社. 2004.9.
[25]方向,薛璞.振动理论及应用[M].西北工业大学出版社. 1998.5.
[26]熊波.基于超声波的PC轨道梁支座锚固螺杆健康监测研究[D].重庆大学, 2005.
[27] D Sciulli, D J N.man. Isolation design for a flexible system. Journal of Sound and Vibration, 1998, 216(2):251-267
[28] Ishida K et al. Dynamic characteristics of soil-foundation interaction detected from vibration test. Proceedings of WCEE. Vol. 111: 19.
[29]刘习军,贾启芬.工程振动与测试技术[M].天津:天津大学出版社, 1999, 9.
[30]李德葆,陆秋海.工程振动试验分析[M].北京:清华大学出版社. 2004.
[31]刘笃仁,韩保君.传感器原理及应用技术[M].西安:西安电子科技大学出版社. 2003.
[32]杨玉星.生物医学传感器与检测技术[M].北京:化学工业出版社. 2005.
[33]任兴民,秦卫阳,文立华等.工程振动基础[M].北京:机械工业出版社. 2006.
[34]孔德仁,朱蕴璞,狄长安.工程测试技术[M].北京:科学出版社. 2004.
[35]王琦.基于振动多特征的轻轨锚固螺杆松动故障诊断方法研究[D].重庆大学. 2008.
[36] Park G, Cudney H H , IN.man D J. Impedance2based health monitoring technique for civil structures[A]. Proceedings of the 2nd international workshop on structural health monitoring[C]. Stanford : Stanford University, 1999. 523 - 532.
[37] Sohn H, Farrar C R. Damage diagnosis using time series analysis of vibration signals[J] . Smart Materials and Structures, 2001,(10):446 - 451.
[38]谢峻,韩大建.一种改进的基于频率测量的结构损伤识别方法[J].工程力学, 2004, 21(1): 21-25.
[39]刘文峰,柳春图,应怀樵.通过频率改变率进行损伤定位的方法研究[J].振动与冲击. 2004, 23(2): 28-30.
[40] Cawley P ,Adams R D. The location of defects in structures from measurements of natural frequencies[J]. Journal of Strain Analysis. 1979, 4(2): 49—57
[41] Gudmundson P. Eigenfrequency changes of structures due to cracks, notches or other geometrical changes[J]. Journal of Mech. Phys. Solids, 1982, 30(5): 339 -353.
[42] Yao G C C,Chang K C.A Study of damage diagnosis from earthquake records of a steelgable frame[J].Journal of the Chinese Institute of Engineers, 1995. 18(1): 115-123.
[43]郑栋梁,李中付,华宏星.结构早期损伤识别技术的现状和发展趋势[J].振动与冲击, 2002, 21(2): 1-4.
[44] Lieven N A J . Spatial correlation of mode shapes, The Coordinate Modal Assurance Criterion (COMACO) [C], Proceedings of 6th IMAC, 1988, 690-695.
[45] Fox C H J. The location of defects in structures: A comparison of the use of natural frequency and mode shape data[C]. Proceedings of 10th IMAC, 1992, 522-528.
[46] Pandey A K et al. Damage detection from changes in curvature mode shapes[J], Journal of Sound and Vibration, 1992 ,145(2):321 -332
[47] Jiann-Shium Lew. Using transfer function parameter changes for damage detection of structures[J]. AIAA Journal, 1995, 33(11) :2189-2193.
[48] David C . ZiN.merman eta1. Structural damage detection using frequency response function[C]. Proceedings of 13 IMAC, 1995,179-184.
[49] Mark J S et a1. Detecting structural damage using transmittance function[C]. Proceedings of 15 IMAC, Florida, 1997, 638-644.
[50]王术新,姜哲.基于结构振动伤识别技术的研究现状及进展[J].振动与冲击, 2004, 23(4): 99-102.
[51] N Stubbs, R Osegueda. Global nondestructive damage evaluation in solid[J]. International Journal of Analytical and Experimental Modal Analysis, 1990,5(2):67-79.
[52]韩西,钟厉,邵毅明等.城市轻轨PC梁铸钢支座强度分析和实验研究[J].土木工程学报.2003.36(2):63-66.
[53]叶梅新,黄琼.高速铁路钢-混凝土组合梁的损伤识别[J].中南大学学报(自然科学版). 2005.36(4):704-709
[54]罗公亮.核函数方法[J].冶金自动化,2002,(4):1-3.
[55] Rtsch G,Schlkopf B,Mika S, et al. SVM and boosting: one class[M]. Technical Report 119, Berlin: GMD FIRST, Kekul’estr, 2000
[56]韦振中.基于核主成分分析的特征提取方法[J].广西工学院学报, 2006, 17(4) :27-31.
[57] L.J.Cao,W.K.Chong.Feature extraction in support vector machine: a comparison of PCA, KPCA and ICA[J]. 9th International Conference on Neural Information Processing, 2002,(2):1001-1005.
[58] John Shawe-Taylor,Nello Cristianini.模式分析的核方法[M].北京:机械工业出版社,2006.
[59]廖东平.基于1类SVM的高分辨雷达真假目标识别[J].信号处理,2010,26(5): 746-752.
[60] Tax D.M.J,Duin R.P.W,Data Domain Description Using Support Vectors[A],In:Verleysen M,Edits,Proceedings of the European Symposium on Artificial Neural Networks[C],Brussels,Belgium:Facto D Press,1999:251-256.
[61] Tax D.M.J,Duin R.P.W,Support Vector Data Domain Description[J],Pattern Recognition Letters,1999,20:1191-1199.
[62] Scholkopf B,Williamson R.C,Smola A.J,Taylor J.S,Platt J.C,Estimating the Support of a High-Dimensional Distribution[A],Technical Report MSR-TR-99-87,Microsoft Research,Redmond,WA,1999.
[63] Richard O.Duda,Peter E.Hart,David G.Stork著;李宏东,姚天湘等译.模式分类(第二版)[M].北京:机械工业版社,2003.
[64]廖东平.支持向量机方法及其在机载毫米波雷达目标识别中的应用研究[D].国防科学技术大学,2006.
[65]郭红梅,佐军.LabVIEW与Matlab混合编程的过程控制应用[J].控制工程,2007,1 4(增刊):1 3 1—1 33.
[2]刘家兵.基于超声波的轻轨在役锚固螺杆检测研究[D].重庆大学,2007.
[3]郭子义.跨座式单轨轨道梁在役锚固螺杆超声波信号自动采集方法和系统[D].重庆大学,2008.
[4]于得水.基于振动技术的在役锚固螺杆紧固力矩检测研究[D].重庆大学, 2007.
[5]赛铁兵.跨座式单轨PC轨道梁线形调整施工技术[J].铁道标准设计,2007,(1):40-42.
[6]苏明辉.重庆轻轨较新线PC轨道梁架设技术[J].桥梁, 2005(增):1-4.
[7] Kim J T , Lyon R H. Spectrum analysis as a tool for robust processing, reverberation and detection of transients. Mechanical System and Signal Processing, 1992, 6(1):1~15.
[8] Michael Kuby. Factors influencing light-rail station boar dings in the United States [J] . Transportation Research Part A , 2004 ,38 :223 - 247.
[9] Nigel G Harris. The marginal profitability of train services in London [J] . Transport Policy, 1996, 13(1/ 2):17 - 20.).
[10] Sydney C, Chu K. Crew scheduling of light rail transit in Hong Kong: From modeling to implementation [J]. Computer Ops Res, 1998, 25(11):887 - 894.
[11]马继兵,蒲黔辉,夏招广.跨座式单轨交通PC轨道梁静载及疲劳试验研究[J].铁道建筑. 2008(5):3-6.
[12]韩西,邵毅明,钟厉等.跨座式轻轨PC轨道梁活动钢支座疲劳及寿命有限元分析[J].公路交通科技. 2001, 18(6):47-49.
[13]韩西,钟厉,邵毅明等.轻轨PC梁铸钢支座有限元分析及实验[J].重庆大学学报(自然科学版). 2002, 25(5):112-115.
[14]张贵寿,邵晓蓉.建筑结构的鉴定与加固[J].绍兴文理学报. 2000. 20(5), 82.
[15]王仲生,万小朋.无损检测诊断现场实用技术(第一版)[M].北京:机械工业出版社, 2002: 5-30.
[16]闫桂荣,段忠东,欧进萍.基于结构振动信息的损伤识别研究综述[J].地震工程与工程振动. 2007.
[17]张竞爽.悬臂梁非线性振动分析与实验[D].天津大学. 2009.
[18] Norden Huang, Kang Huang.基于希尔伯特_黄变换的铁路桥梁结构健康监测[J].中国铁道科学. 2006, 27(1):1-5.
[19]杨德健,郭天德.某军用飞机振动模态分析[J].天津城市建设学院学报. 1996, 2(3):1-5.
[20]王朝晖,张来斌,樊长博.基于EMD和HT时频分析方法的滚动轴承故障诊断.石油机械[J]. 2007, 35(5):32-33.
[21]邹春平,陈端石,华宏星.船舶结构振动模态综合[J]. 2003, 37(8):1213-1217.
[22]杨和振,李华军,黄维平.基于振动测试的海洋平台结构无损检测[J]. 2003, 16(4):480-484.
[23]周杏鹏,仇国富,王寿荣等.现代检测技术[M].北京,高等教育出版社. 2004:148.
[24]李德葆,陆秋海.工程振动试验分析[M].清华大学出版社. 2004.9.
[25]方向,薛璞.振动理论及应用[M].西北工业大学出版社. 1998.5.
[26]熊波.基于超声波的PC轨道梁支座锚固螺杆健康监测研究[D].重庆大学, 2005.
[27] D Sciulli, D J N.man. Isolation design for a flexible system. Journal of Sound and Vibration, 1998, 216(2):251-267
[28] Ishida K et al. Dynamic characteristics of soil-foundation interaction detected from vibration test. Proceedings of WCEE. Vol. 111: 19.
[29]刘习军,贾启芬.工程振动与测试技术[M].天津:天津大学出版社, 1999, 9.
[30]李德葆,陆秋海.工程振动试验分析[M].北京:清华大学出版社. 2004.
[31]刘笃仁,韩保君.传感器原理及应用技术[M].西安:西安电子科技大学出版社. 2003.
[32]杨玉星.生物医学传感器与检测技术[M].北京:化学工业出版社. 2005.
[33]任兴民,秦卫阳,文立华等.工程振动基础[M].北京:机械工业出版社. 2006.
[34]孔德仁,朱蕴璞,狄长安.工程测试技术[M].北京:科学出版社. 2004.
[35]王琦.基于振动多特征的轻轨锚固螺杆松动故障诊断方法研究[D].重庆大学. 2008.
[36] Park G, Cudney H H , IN.man D J. Impedance2based health monitoring technique for civil structures[A]. Proceedings of the 2nd international workshop on structural health monitoring[C]. Stanford : Stanford University, 1999. 523 - 532.
[37] Sohn H, Farrar C R. Damage diagnosis using time series analysis of vibration signals[J] . Smart Materials and Structures, 2001,(10):446 - 451.
[38]谢峻,韩大建.一种改进的基于频率测量的结构损伤识别方法[J].工程力学, 2004, 21(1): 21-25.
[39]刘文峰,柳春图,应怀樵.通过频率改变率进行损伤定位的方法研究[J].振动与冲击. 2004, 23(2): 28-30.
[40] Cawley P ,Adams R D. The location of defects in structures from measurements of natural frequencies[J]. Journal of Strain Analysis. 1979, 4(2): 49—57
[41] Gudmundson P. Eigenfrequency changes of structures due to cracks, notches or other geometrical changes[J]. Journal of Mech. Phys. Solids, 1982, 30(5): 339 -353.
[42] Yao G C C,Chang K C.A Study of damage diagnosis from earthquake records of a steelgable frame[J].Journal of the Chinese Institute of Engineers, 1995. 18(1): 115-123.
[43]郑栋梁,李中付,华宏星.结构早期损伤识别技术的现状和发展趋势[J].振动与冲击, 2002, 21(2): 1-4.
[44] Lieven N A J . Spatial correlation of mode shapes, The Coordinate Modal Assurance Criterion (COMACO) [C], Proceedings of 6th IMAC, 1988, 690-695.
[45] Fox C H J. The location of defects in structures: A comparison of the use of natural frequency and mode shape data[C]. Proceedings of 10th IMAC, 1992, 522-528.
[46] Pandey A K et al. Damage detection from changes in curvature mode shapes[J], Journal of Sound and Vibration, 1992 ,145(2):321 -332
[47] Jiann-Shium Lew. Using transfer function parameter changes for damage detection of structures[J]. AIAA Journal, 1995, 33(11) :2189-2193.
[48] David C . ZiN.merman eta1. Structural damage detection using frequency response function[C]. Proceedings of 13 IMAC, 1995,179-184.
[49] Mark J S et a1. Detecting structural damage using transmittance function[C]. Proceedings of 15 IMAC, Florida, 1997, 638-644.
[50]王术新,姜哲.基于结构振动伤识别技术的研究现状及进展[J].振动与冲击, 2004, 23(4): 99-102.
[51] N Stubbs, R Osegueda. Global nondestructive damage evaluation in solid[J]. International Journal of Analytical and Experimental Modal Analysis, 1990,5(2):67-79.
[52]韩西,钟厉,邵毅明等.城市轻轨PC梁铸钢支座强度分析和实验研究[J].土木工程学报.2003.36(2):63-66.
[53]叶梅新,黄琼.高速铁路钢-混凝土组合梁的损伤识别[J].中南大学学报(自然科学版). 2005.36(4):704-709
[54]罗公亮.核函数方法[J].冶金自动化,2002,(4):1-3.
[55] Rtsch G,Schlkopf B,Mika S, et al. SVM and boosting: one class[M]. Technical Report 119, Berlin: GMD FIRST, Kekul’estr, 2000
[56]韦振中.基于核主成分分析的特征提取方法[J].广西工学院学报, 2006, 17(4) :27-31.
[57] L.J.Cao,W.K.Chong.Feature extraction in support vector machine: a comparison of PCA, KPCA and ICA[J]. 9th International Conference on Neural Information Processing, 2002,(2):1001-1005.
[58] John Shawe-Taylor,Nello Cristianini.模式分析的核方法[M].北京:机械工业出版社,2006.
[59]廖东平.基于1类SVM的高分辨雷达真假目标识别[J].信号处理,2010,26(5): 746-752.
[60] Tax D.M.J,Duin R.P.W,Data Domain Description Using Support Vectors[A],In:Verleysen M,Edits,Proceedings of the European Symposium on Artificial Neural Networks[C],Brussels,Belgium:Facto D Press,1999:251-256.
[61] Tax D.M.J,Duin R.P.W,Support Vector Data Domain Description[J],Pattern Recognition Letters,1999,20:1191-1199.
[62] Scholkopf B,Williamson R.C,Smola A.J,Taylor J.S,Platt J.C,Estimating the Support of a High-Dimensional Distribution[A],Technical Report MSR-TR-99-87,Microsoft Research,Redmond,WA,1999.
[63] Richard O.Duda,Peter E.Hart,David G.Stork著;李宏东,姚天湘等译.模式分类(第二版)[M].北京:机械工业版社,2003.
[64]廖东平.支持向量机方法及其在机载毫米波雷达目标识别中的应用研究[D].国防科学技术大学,2006.
[65]郭红梅,佐军.LabVIEW与Matlab混合编程的过程控制应用[J].控制工程,2007,1 4(增刊):1 3 1—1 33.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |