多界面超声脱粘检测的方法研究及信号处理
|
摘要
目前，超声检测作为一种常用的无损检测技术被广泛的应用于航空航天、核工业、石油及其它民用领域的粘接结构检测中。对于多层钢/橡胶交替层叠而成的粘接结构，由于钢和橡胶之间的声阻抗差异较大，用常规的超声检测技术很难满足检测要求。本文以火箭发动机柔性接头部件为对象，分别采用板波和纵波两种检测技术，研究多层粘接结构脱粘超声检测技术及其信号处理。
本文首先分析了板波诱发纵波在柔性接头构件中的传播规律，对原始回波信号进行时域和频域分析。在此基础上，提出了常数分离法和自适应分离法，分别用它们分离板波回波信号中的各界面回波，利用小波多分辨率分析提取原始回波信号的特征和BP神经网络进行模式识别。
文章还尝试了纵波检测技术检测这种构件，得出了用双探头穿透法可以判定是否存在界面脱粘，用自适应滤波对直探头检测的回波进行处理，所得结果反应了各界面脱粘回波的特征。
实验证明，用以上两种检测技术都能很好的检测这种构件的界面粘接状态，综合应用它们可以对构件进行全面、有效的检测。
Ultrasonic testing, one of the common nondestructive testing technologies, is widely used in aerospace, nuclear, petroleum industry and other fields for defects in the adhesive structure. It is difficult to detect interface debond in the adhesive multi-layered steel-rubber structure through the common ultrasonic testing due to the considerable acoustic impedance differences of between the steel and the rubber. The ultrasonic testing and signal processing for bonded defects in the structure of flexible joint of the solid rocket motors are studied. And the Lamb Wave method and the Longitudinal Wave method are used in the ultrasonic testing.
The transmission rule of the lamb wave in the structure of the flexible joint is discussed, and the ultrasonic pulse echo is analyzed in the time-field and frequency-field. The Constant-separate method and the Adaptive-separate method based on the above ultrasonic principle are used to separate each interface-echo in the ultrasonic pulse echo. The feature of the each ultrasonic pulse echo is gained through the wavelet transformation, and is identified through BP neural network.
The Longitudinal Wave method for testing the structure is also discussed. It is found that the bi-sensor method can determine whether the structure has interface debond. Then the adaptive filtering is adopted in the signal processing for the echo through single-sensor. Based on the above results, the feature of each interface echo is gained.
It is shown by experiments that the Lamb Wave method and the Longitudinal Wave method are both capable of detecting the adhesive state of the structure. The detection of the structure becomes more comprehensive and effective if the above two methods ate employed in combination.
本文首先分析了板波诱发纵波在柔性接头构件中的传播规律，对原始回波信号进行时域和频域分析。在此基础上，提出了常数分离法和自适应分离法，分别用它们分离板波回波信号中的各界面回波，利用小波多分辨率分析提取原始回波信号的特征和BP神经网络进行模式识别。
文章还尝试了纵波检测技术检测这种构件，得出了用双探头穿透法可以判定是否存在界面脱粘，用自适应滤波对直探头检测的回波进行处理，所得结果反应了各界面脱粘回波的特征。
实验证明，用以上两种检测技术都能很好的检测这种构件的界面粘接状态，综合应用它们可以对构件进行全面、有效的检测。
Ultrasonic testing, one of the common nondestructive testing technologies, is widely used in aerospace, nuclear, petroleum industry and other fields for defects in the adhesive structure. It is difficult to detect interface debond in the adhesive multi-layered steel-rubber structure through the common ultrasonic testing due to the considerable acoustic impedance differences of between the steel and the rubber. The ultrasonic testing and signal processing for bonded defects in the structure of flexible joint of the solid rocket motors are studied. And the Lamb Wave method and the Longitudinal Wave method are used in the ultrasonic testing.
The transmission rule of the lamb wave in the structure of the flexible joint is discussed, and the ultrasonic pulse echo is analyzed in the time-field and frequency-field. The Constant-separate method and the Adaptive-separate method based on the above ultrasonic principle are used to separate each interface-echo in the ultrasonic pulse echo. The feature of the each ultrasonic pulse echo is gained through the wavelet transformation, and is identified through BP neural network.
The Longitudinal Wave method for testing the structure is also discussed. It is found that the bi-sensor method can determine whether the structure has interface debond. Then the adaptive filtering is adopted in the signal processing for the echo through single-sensor. Based on the above results, the feature of each interface echo is gained.
It is shown by experiments that the Lamb Wave method and the Longitudinal Wave method are both capable of detecting the adhesive state of the structure. The detection of the structure becomes more comprehensive and effective if the above two methods ate employed in combination.
引文
[1] 尹华丽，王清和.界面粘接性能的影响因素[J].固体火箭技术，1998，21(3): 40-46.
[2] 李芳.固体火箭发动机包覆层状态识别的研究[D].硕士学位论文,北京：北京航空航天大学，2001.
[3] 赵慧蓉.固体火箭发动机喷管粘接界面的超声检测[J].固体火箭技术，2000, 23 (2):74-78.
[4] 吴庆刚.多界面脱粘超声检测技术研究[D].硕士学位论文，太原：华北工学院，2002.
[5] M.X.Li. Principles of an acoustic impendence method for detection and location of non-bonds in adhesive-bonded muitilayered joints[J]. NBT International, 1982, 15:129-176.
[6] C.H.Gnyott. The ultrasonic vibration characteristics of adhesive joints[J]. J.A.S.A., 1988, 83(2):632-640.
[7] T.Pialucha. An investigation of the accuracy of oblique incidence ultrasonic reflection coeffcient measurements[J].J.A.S.A,1994,96(3):1651-1660.
[8] A.Pilarski. A transverse-wave ultrasonic oblique incidence technique for interfacial weakness detection in adhesive bonds[J].J.Appl.Phys.,1988:300-307.
[9] A.Vary. Acousto-ultrasonic characterization of fiber reinforced composite[J]. Material Evaluation，1982，40:650-662.
[10] J.C.Duck. Acousto-Ultrasonic[M]. Plenum Press, New York, 1988.
[11] S.I.Rokhlin. Lamb wave interaction with lap-shear adhesive joints: theory and experiment[J]. J.A.S.A., 1991，89(6):2758-2765.
[12] M.R.Karim. Inversion of leaky lamb wave data by simplex algorithm[J]. J.A.S.A.，1990，88(1):482-491.
[13] 毛捷.多层粘接超声检测信号处理方法研究[D].博士学位论文，北京：中国科学院，2001.
[14] V.L.Newhouse. Flaw-to-grain echo enhancement by split-spectrum processing[J].
Ultrasonics, 1982, 20(3): 59-68.
[15] Rose. Utility of split-spectrum processing in ultrasonic nondestructive evaluation[J]. Mater Eval，1988，46:114-122.
[16] L.Ericsson. Cut spectrum processing:a novel signal processing algorithm for ultrasonic flaw detection[J]. NDT&E International, 1992, 25(2): 59-64.
[17] Xing Li. Spectral Histogram Using the Minimization Algorithm-Theory and Applications to Flaw Detection[J]. IEEE Trans.UFFC, 1992, 39(2):279-284.
[18] X.Li. Wiener filter realization for target detection using group delay statistics[J]. IEEE Trans.Signal Processing, 1993, 41(6):2067-2074.
[19] Y.Zhu. Ultrasonic NDT of highly scattering materials using adaptive filtering and detection[J]. IEEE Trans.UFFC, 1994, 41(1):26.
[20] 简晓明.超声检测中信号处理研究[D].博士学位论文，北京：中国科学院，1999.
[21] 简晓明，李明轩.小波变换和自适应噪声抵消在闭合裂纹超声检测中的应用[J].声学学报，2000，25(2):97-102.
[22] J.Chao. Ultrasonic signal analysis using wavelet transform[J]. Review of progress in QNE, 1993, 12:735-741.
[23] J.P.Lefebvre. Wavelet analysis for ultrasonic crack detection and modelixation[J]. Ultrasonics Symposium, 1994:1143-1146.
[24] Jianqiang Xin. Detection and resolution of multiple targets using time-frequency techniques[J].Ultrasonics Symposium, 1994:1133-1137.
[25] Agostino Abbate. Signal detection and noise suppression using a wavelet transform signal processor: application to ultrasonic flaw detection[J]. IEEE Trans.UFFC，1997，44(l):14-25.
[26] 刘镇清.增强颗粒散射中缺陷回波信号的相关加权分离谱方法[J].声学学报，1995，121(4增刊):714-725.
[27] 刘镇清.超声探伤信号的延时神经网络处理[J].声学学报，1997，22(4):297-302.
[28] 刚铁.超声回波的特征值提取与统计分析[J].无损检测，1995, 17(9):248-251.
[29] M.Lorenz. Ultrasonic characterization of defects in steel using multi-SAFT imaging and
neural and networks[J]. NDT&E International, 1993, 26(3): 127-133.
[30] 简晓明，李明轩.用自适应系统模拟解卷积进行缺陷类型识别[J].声学学报，1999，24(6):637-644.
[31] 吴淼，夏金东等.超声检测缺陷分类的小波分析与神经网络方法[J].中国矿业大学学报，2000，29(3):240-243.
[32] 刘旭，夏金东等.超声检测缺陷分类的降噪及特征提取问题的研究[J].中国矿业大学学报（自然科学报），2001，30(3):248-251.
[33] 夏金东，刘旭等.超声检测中信号提取和缺陷分类方法的研究[J].中国学术期刊文摘（科技快报），2001，7(6):748-752.
[34] 刘镇清，张海燕.人工神经网络及其在超声检测中的应用[J].无损检测，2001，23(25):221-225.
[35] A.R.Baker. The classification of defects from ultrasonic data using neural networks:The Hopfield method[J]. NDT international, 1989, 22(2):97-105.
[36] 简晓明,李明轩.超声检测中人工神经网络对缺陷定量评价[J].声学学报，2000，25(11):71-77.
[37] 杨风暴.金属与非金属粘接检测信息的融合处理.博士学位论文，太原：华北工学院，2003.
[38] Luca Goglio. Ultrasonic testing of adhesive bonds of thin metal sheets[J]. NDT&E international, 1999, 32:323-331.
[39] Charles P.D.Todd. Quantitative classification of adhesive bondline dimensions using lamb waves and artificial neural networks[J]. IEEE Trans.UFFC, 1999, 46(1):655-659.
[40] Y. J. Chen. Detection of weak bonding in friction welds by ultrasound[J]. Ultrasonics, 1998, 36:141-146.
[41] 简晓明，李明轩.层状介质界面超声检测的理论分析和自适应噪声抵消原理[J].声学学报，2000，25(4):351-356.
[42] 张建生，李明轩.层状粘接结构超声检测信号的同态卷积脱粘界面顶征[J].应用声学，2001，20(1):23-29.
[43] 张建生，李明轩.多层粘接结构中脱粘界面的人工神经网络余璇变换谱特征识别[J].
声学学报，2001，26(4):349-354.
[44] 张建生，李明轩.脱粘界面超声检测信号的小波多分辨率分析与重构[J].声学学报，2001，26(3):231-238.
[45] 李明轩.粘接质量超声检测研究[J].应用声学，2002，21(1):7-12.
[46] 王召巴，杨风暴等.火箭发动机装药包覆质量诊断的超声新技术[J].华北工学院测试技术学报，2001，15(3):9-13.
[47] 超声波探伤编写组.超声波探伤[M].北京：电力工业出版社，1980.
[48] 胡建恺,张谦琳.超声检测原理和方法[M].合肥：中国科技大学出版社，1993.
[49] 应崇福，张守玉等.超声在固体中的散射[M].北京：国防工业出版社，1994.
[50] 刘镇清.自适应滤波在超声无损检测中的应用[J].无损检测，2001，23(9):399-401.
[51] Stephane Mallat著，扬力华译.信号处理的小波导引（第二版）[M].北京：机械工业出版社，2002.
[52] 秦前清，杨宗凯.实用小波分析[M].西安：西安电子科技大学出版社，1995.
[53] 飞思科技产品研发中心. MATLAB6.5辅助小波分析与应用[M].北京：电子工业出版社，2003.
[54] 罗发龙，李衍达.神经网络信号处理[M].北京：电子工业出版社，1993.
[55] 丛爽编.面向MATLAB工具箱的神经网络理论与应用（第2版）[M].合肥：中国科学技术大学出版社，2003.
[2] 李芳.固体火箭发动机包覆层状态识别的研究[D].硕士学位论文,北京：北京航空航天大学，2001.
[3] 赵慧蓉.固体火箭发动机喷管粘接界面的超声检测[J].固体火箭技术，2000, 23 (2):74-78.
[4] 吴庆刚.多界面脱粘超声检测技术研究[D].硕士学位论文，太原：华北工学院，2002.
[5] M.X.Li. Principles of an acoustic impendence method for detection and location of non-bonds in adhesive-bonded muitilayered joints[J]. NBT International, 1982, 15:129-176.
[6] C.H.Gnyott. The ultrasonic vibration characteristics of adhesive joints[J]. J.A.S.A., 1988, 83(2):632-640.
[7] T.Pialucha. An investigation of the accuracy of oblique incidence ultrasonic reflection coeffcient measurements[J].J.A.S.A,1994,96(3):1651-1660.
[8] A.Pilarski. A transverse-wave ultrasonic oblique incidence technique for interfacial weakness detection in adhesive bonds[J].J.Appl.Phys.,1988:300-307.
[9] A.Vary. Acousto-ultrasonic characterization of fiber reinforced composite[J]. Material Evaluation，1982，40:650-662.
[10] J.C.Duck. Acousto-Ultrasonic[M]. Plenum Press, New York, 1988.
[11] S.I.Rokhlin. Lamb wave interaction with lap-shear adhesive joints: theory and experiment[J]. J.A.S.A., 1991，89(6):2758-2765.
[12] M.R.Karim. Inversion of leaky lamb wave data by simplex algorithm[J]. J.A.S.A.，1990，88(1):482-491.
[13] 毛捷.多层粘接超声检测信号处理方法研究[D].博士学位论文，北京：中国科学院，2001.
[14] V.L.Newhouse. Flaw-to-grain echo enhancement by split-spectrum processing[J].
Ultrasonics, 1982, 20(3): 59-68.
[15] Rose. Utility of split-spectrum processing in ultrasonic nondestructive evaluation[J]. Mater Eval，1988，46:114-122.
[16] L.Ericsson. Cut spectrum processing:a novel signal processing algorithm for ultrasonic flaw detection[J]. NDT&E International, 1992, 25(2): 59-64.
[17] Xing Li. Spectral Histogram Using the Minimization Algorithm-Theory and Applications to Flaw Detection[J]. IEEE Trans.UFFC, 1992, 39(2):279-284.
[18] X.Li. Wiener filter realization for target detection using group delay statistics[J]. IEEE Trans.Signal Processing, 1993, 41(6):2067-2074.
[19] Y.Zhu. Ultrasonic NDT of highly scattering materials using adaptive filtering and detection[J]. IEEE Trans.UFFC, 1994, 41(1):26.
[20] 简晓明.超声检测中信号处理研究[D].博士学位论文，北京：中国科学院，1999.
[21] 简晓明，李明轩.小波变换和自适应噪声抵消在闭合裂纹超声检测中的应用[J].声学学报，2000，25(2):97-102.
[22] J.Chao. Ultrasonic signal analysis using wavelet transform[J]. Review of progress in QNE, 1993, 12:735-741.
[23] J.P.Lefebvre. Wavelet analysis for ultrasonic crack detection and modelixation[J]. Ultrasonics Symposium, 1994:1143-1146.
[24] Jianqiang Xin. Detection and resolution of multiple targets using time-frequency techniques[J].Ultrasonics Symposium, 1994:1133-1137.
[25] Agostino Abbate. Signal detection and noise suppression using a wavelet transform signal processor: application to ultrasonic flaw detection[J]. IEEE Trans.UFFC，1997，44(l):14-25.
[26] 刘镇清.增强颗粒散射中缺陷回波信号的相关加权分离谱方法[J].声学学报，1995，121(4增刊):714-725.
[27] 刘镇清.超声探伤信号的延时神经网络处理[J].声学学报，1997，22(4):297-302.
[28] 刚铁.超声回波的特征值提取与统计分析[J].无损检测，1995, 17(9):248-251.
[29] M.Lorenz. Ultrasonic characterization of defects in steel using multi-SAFT imaging and
neural and networks[J]. NDT&E International, 1993, 26(3): 127-133.
[30] 简晓明，李明轩.用自适应系统模拟解卷积进行缺陷类型识别[J].声学学报，1999，24(6):637-644.
[31] 吴淼，夏金东等.超声检测缺陷分类的小波分析与神经网络方法[J].中国矿业大学学报，2000，29(3):240-243.
[32] 刘旭，夏金东等.超声检测缺陷分类的降噪及特征提取问题的研究[J].中国矿业大学学报（自然科学报），2001，30(3):248-251.
[33] 夏金东，刘旭等.超声检测中信号提取和缺陷分类方法的研究[J].中国学术期刊文摘（科技快报），2001，7(6):748-752.
[34] 刘镇清，张海燕.人工神经网络及其在超声检测中的应用[J].无损检测，2001，23(25):221-225.
[35] A.R.Baker. The classification of defects from ultrasonic data using neural networks:The Hopfield method[J]. NDT international, 1989, 22(2):97-105.
[36] 简晓明,李明轩.超声检测中人工神经网络对缺陷定量评价[J].声学学报，2000，25(11):71-77.
[37] 杨风暴.金属与非金属粘接检测信息的融合处理.博士学位论文，太原：华北工学院，2003.
[38] Luca Goglio. Ultrasonic testing of adhesive bonds of thin metal sheets[J]. NDT&E international, 1999, 32:323-331.
[39] Charles P.D.Todd. Quantitative classification of adhesive bondline dimensions using lamb waves and artificial neural networks[J]. IEEE Trans.UFFC, 1999, 46(1):655-659.
[40] Y. J. Chen. Detection of weak bonding in friction welds by ultrasound[J]. Ultrasonics, 1998, 36:141-146.
[41] 简晓明，李明轩.层状介质界面超声检测的理论分析和自适应噪声抵消原理[J].声学学报，2000，25(4):351-356.
[42] 张建生，李明轩.层状粘接结构超声检测信号的同态卷积脱粘界面顶征[J].应用声学，2001，20(1):23-29.
[43] 张建生，李明轩.多层粘接结构中脱粘界面的人工神经网络余璇变换谱特征识别[J].
声学学报，2001，26(4):349-354.
[44] 张建生，李明轩.脱粘界面超声检测信号的小波多分辨率分析与重构[J].声学学报，2001，26(3):231-238.
[45] 李明轩.粘接质量超声检测研究[J].应用声学，2002，21(1):7-12.
[46] 王召巴，杨风暴等.火箭发动机装药包覆质量诊断的超声新技术[J].华北工学院测试技术学报，2001，15(3):9-13.
[47] 超声波探伤编写组.超声波探伤[M].北京：电力工业出版社，1980.
[48] 胡建恺,张谦琳.超声检测原理和方法[M].合肥：中国科技大学出版社，1993.
[49] 应崇福，张守玉等.超声在固体中的散射[M].北京：国防工业出版社，1994.
[50] 刘镇清.自适应滤波在超声无损检测中的应用[J].无损检测，2001，23(9):399-401.
[51] Stephane Mallat著，扬力华译.信号处理的小波导引（第二版）[M].北京：机械工业出版社，2002.
[52] 秦前清，杨宗凯.实用小波分析[M].西安：西安电子科技大学出版社，1995.
[53] 飞思科技产品研发中心. MATLAB6.5辅助小波分析与应用[M].北京：电子工业出版社，2003.
[54] 罗发龙，李衍达.神经网络信号处理[M].北京：电子工业出版社，1993.
[55] 丛爽编.面向MATLAB工具箱的神经网络理论与应用（第2版）[M].合肥：中国科学技术大学出版社，2003.
© 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 |