基于小波网络的冲激信号检测方法研究
|
摘要
本文研究了利用小波网络方法在强烈振动信号背景下检测瞬态冲激的问题,并用实际振动信号检验了这种方法的效果。
在大型结构及桥梁状态监测研究领域,桁架结构的个别杆件松动或板壳出现裂纹是导致系统失效的主要故障之一。结构出现此类故障的初期,当外载荷发生较大变化时,系统内部的局部损伤部位会出现失稳现象,而局部失稳后会导致载荷在瞬间重新分配。结构内部的载荷重新分配过程在结构外部也会有所表现:测试结构的振动加速度,可以观察到随着载荷的变化,系统响应中会出现间歇性的冲激信号。因此,监测系统响应中的冲激现象可以发现系统内部的损伤情况。但是,实际状态监测系统所采集到的信号是结构在随机工作载荷作用下的响应,杆件失稳或板壳裂纹所导致的异常冲激响应淹没在工作载荷产生的正常随机信号之中,在时域波形上很难进行直观识别。而冲激信号和随机响应都有很宽的频带并且相互重叠,因而也很难在频域直观分辨。
大型结构及桥梁等力学系统在外力作用下产生响应的过程中,结构本身是一个多自由度二阶线性系统。相对于杆件松动、板壳裂纹所产生的冲激响应,结构承受的交通、风力、温度变化、大地脉动等等载荷所产生的响应的性质是连续且变化相对缓慢,因此,可以利用小波变换从大背景噪声中将瞬间冲激分离出来,从而达到监测结构损伤的目的。为此,本文利用第二代小波的多尺度和良好的时频特性,以及用信息熵来反应信号的统计分布特性,处理含有冲激的振动信号,结果表明:用第二代小波熵的分析方法能够准确的从强背景噪声中确定冲激信号出现的时刻。
论文第一章是绪论,介绍了研究工作的背景和意义;第二章介绍了小波分析基本概念和理论,最后介绍小波网络基本概念;第三章探讨小波网络构造、学习算法以及小波网络中的小波基的选择;最后将小波网络应用在振动信号消噪中;第四章探讨了第二代小波熵,并将弱振动冲击信号定位。
This paper studies the issue that how to detect transient impulse using wavelet network in the context of strong vibration signals, and this method is tested to be effective by using actual vibration signals.
Individual member loosing or plate and shell crack in truss structure is one of main fault that led to system failure, in the state monitoring of large-scale structures and bridges. In early stage of structural failure, the area of partial damaging within the system will appear unsteady, and it will lead to reallocate of load in an instant, when external load large change. The process of the load reallocate in the internal structure will be performance in external structure: can be observed impulse intermittent signal in the system response as the load changing, by using test vibration acceleration of structure. Therefore, monitoring the impulse phenomenon in system response can be found the damage of system. However, the signal that is collected in the actual condition monitoring system is the response that stochastic work load produce, the abnormal impulse response that is caused by member instability or plate and shell crack immersed normal stochastic signals which is caused by work load, this phenomenon is very difficult direct-viewing recognition in the time-domain waveform. But the impulse signal and the stochastic response all has the very wide frequency band, and the frequency overlap mutually, it is also very difficult direct-viewing resolution in the frequency range.
The structure itself is a multi-degree of freedom second order linear system in the response processing that large-scale structures and bridges and so on mechanics systems under external force. Comparing with the impulse response produced by member loose or plate and shell crack, the response nature which the structure bear because of transportation, wind power, temperature change and earth pulsation and so on is continuously and relatively slow. Therefore, this article uses the second generation of wavelet the multi-scale and the good time frequency characteristic, as well as with the information entropy responded the signal the statistical distribution characteristic, processing includes the impulse vibration signal, finally indicated: With the second generation of wavelet entropy analysis method can accurate from the strong background noise determine the impulse signal appears time.
The paper first chapter is an introduction, introduced the research work background and the significance; Second chapter introduced the wavelet analysis basic concept and the theory, finally introduce the wavelet network basic concept; In third chapter discussion wavelet network structure, study algorithm as well as wavelet network wavelet base choice; Finally wavelet network application in vibration signal denoising; Fourth chapter has discussed the second generation of wavelet entropy, and will be weak vibrates the impact signal localization.
在大型结构及桥梁状态监测研究领域,桁架结构的个别杆件松动或板壳出现裂纹是导致系统失效的主要故障之一。结构出现此类故障的初期,当外载荷发生较大变化时,系统内部的局部损伤部位会出现失稳现象,而局部失稳后会导致载荷在瞬间重新分配。结构内部的载荷重新分配过程在结构外部也会有所表现:测试结构的振动加速度,可以观察到随着载荷的变化,系统响应中会出现间歇性的冲激信号。因此,监测系统响应中的冲激现象可以发现系统内部的损伤情况。但是,实际状态监测系统所采集到的信号是结构在随机工作载荷作用下的响应,杆件失稳或板壳裂纹所导致的异常冲激响应淹没在工作载荷产生的正常随机信号之中,在时域波形上很难进行直观识别。而冲激信号和随机响应都有很宽的频带并且相互重叠,因而也很难在频域直观分辨。
大型结构及桥梁等力学系统在外力作用下产生响应的过程中,结构本身是一个多自由度二阶线性系统。相对于杆件松动、板壳裂纹所产生的冲激响应,结构承受的交通、风力、温度变化、大地脉动等等载荷所产生的响应的性质是连续且变化相对缓慢,因此,可以利用小波变换从大背景噪声中将瞬间冲激分离出来,从而达到监测结构损伤的目的。为此,本文利用第二代小波的多尺度和良好的时频特性,以及用信息熵来反应信号的统计分布特性,处理含有冲激的振动信号,结果表明:用第二代小波熵的分析方法能够准确的从强背景噪声中确定冲激信号出现的时刻。
论文第一章是绪论,介绍了研究工作的背景和意义;第二章介绍了小波分析基本概念和理论,最后介绍小波网络基本概念;第三章探讨小波网络构造、学习算法以及小波网络中的小波基的选择;最后将小波网络应用在振动信号消噪中;第四章探讨了第二代小波熵,并将弱振动冲击信号定位。
This paper studies the issue that how to detect transient impulse using wavelet network in the context of strong vibration signals, and this method is tested to be effective by using actual vibration signals.
Individual member loosing or plate and shell crack in truss structure is one of main fault that led to system failure, in the state monitoring of large-scale structures and bridges. In early stage of structural failure, the area of partial damaging within the system will appear unsteady, and it will lead to reallocate of load in an instant, when external load large change. The process of the load reallocate in the internal structure will be performance in external structure: can be observed impulse intermittent signal in the system response as the load changing, by using test vibration acceleration of structure. Therefore, monitoring the impulse phenomenon in system response can be found the damage of system. However, the signal that is collected in the actual condition monitoring system is the response that stochastic work load produce, the abnormal impulse response that is caused by member instability or plate and shell crack immersed normal stochastic signals which is caused by work load, this phenomenon is very difficult direct-viewing recognition in the time-domain waveform. But the impulse signal and the stochastic response all has the very wide frequency band, and the frequency overlap mutually, it is also very difficult direct-viewing resolution in the frequency range.
The structure itself is a multi-degree of freedom second order linear system in the response processing that large-scale structures and bridges and so on mechanics systems under external force. Comparing with the impulse response produced by member loose or plate and shell crack, the response nature which the structure bear because of transportation, wind power, temperature change and earth pulsation and so on is continuously and relatively slow. Therefore, this article uses the second generation of wavelet the multi-scale and the good time frequency characteristic, as well as with the information entropy responded the signal the statistical distribution characteristic, processing includes the impulse vibration signal, finally indicated: With the second generation of wavelet entropy analysis method can accurate from the strong background noise determine the impulse signal appears time.
The paper first chapter is an introduction, introduced the research work background and the significance; Second chapter introduced the wavelet analysis basic concept and the theory, finally introduce the wavelet network basic concept; In third chapter discussion wavelet network structure, study algorithm as well as wavelet network wavelet base choice; Finally wavelet network application in vibration signal denoising; Fourth chapter has discussed the second generation of wavelet entropy, and will be weak vibrates the impact signal localization.
引文
[1]刘琚,何振亚.盲源分离和盲反卷积[J].电子学报,2002,30(4):570-576.
[2]陈进,姜鸣.高阶循环统计量理论在机械故障诊断中的应用[J].振动工学报,2001,14(2):125-134.
[3]陈茜,樊可清,邓志龙.含噪的超定问题及其在振源识别中的应用[J].微计算机信息,2006,18(1):223-225.
[4]Cao Siyuan,Chen Xiangpeng.The Second-generation wavelet transform and its application in denoising of seismic data[J].Applied Geophysics,2005,2(2):70-74.
[5]Cooley JW,Tukey JW.An algorithm for the machine calculation of complex Fourier series[J].Mathematics of Computation,1965,19(90):297-301.
[6]John C Burgess.On digital spectrum analysis of periodic signals[J].J.A coust.Soc.Am,1975,58(3):556-567.
[7]Carlo Offelli,Dario Petri.The influence of windowing on the accuracy of multi-frequency signal parameter estimation[J].IEEE Transactions on Instrumentation and Measurement,1992,41(2):256-261.
[8]朱利民,钟秉林,黄仁.离散频谱多点卷积幅值修正法的理论分析[J].振动工程学报,1999,12(1):120.
[9]丁康,江利旗.离散频谱的能量重心校正法[J].振动工程学报,2001,4(3):54-358.
[10]刘进明,应怀樵.FFT谱连续细化分析的富里叶变换法[J].振动工程学报,1995,8(2):162-166.
[11]谢明,张晓飞,丁康.频谱分析中相位和频率校正的一种新方法--相位差校正法[J].振动工程学报,1999,12(4):454-459.
[12]Ding Kang,Xie Ming.Phase difference correction method for phase and frequency in spectrual analysis[J].Mechanical Systems and Signal Processing,2000,14(5):835-843.
[13]Zhu Limin,Li Hanxiong,Ding H,et al.Noise influence on estimation of signal parameter from the phase difference of discrete Fourier tanseforms[J].Mechanical System s and Signal P rocessing,2002,16(6):991-1004.
[14]Ding Kang,Luo J iangkai,Xie Ming.Time-shifting correcting method of phase difference on discrete spectrum[J].Applied Mathematics and Mechanics,2002,23(7):819-827.
[15]Kay S.A.Fast and accurate single frequency estimator[J].IEEE Trans ASSP,1989,37(12):1987-1990.
[16]刘渝.正弦波频率快速成估计方法[J].数据采集与处理,1998,13(1):7-11.
[17]丁康,朱小勇.适用于加各种窗的一种离散频谱相位差校正法[J].电子学报,2001,29(7):987-989.
[18]丁康,江利旗.离散频谱综合相位差校正法[J].振动工程学报,2002,15(1):114-116.
[19]丁康,谢明等.基于复解析带通滤波器的复调制细化频谱分析方法研究[J].振动工程学报,2001,14(1):29-35.
[20]丁康,谢明等.复解析带通滤波器及其在解调分析中的应用[J].振动工程学报,2000,13(3):385-392.
[21]谢明,丁康.基于复解析带通滤波器的复调制细化谱分析的算法研究[J].振动工程学报,2002,15(4):479-483.
[22]Ranadall R.B.A new method of modeling gear faults[J].ASME Journal of Mechanical Design,1982,104:259-267.
[23]Mofadden P D.Detecting fatigue cracks in gears by amplitude and phase demodulation of the meshing vibration[J].J.of Vibration,Acoustics,Stress,and Reliability in Design,1986:108-165.
[24]樊可清,王玲,陈保平.基于小波分析的工程信号调制解调方法及其应用[J].电子与信息学报,2001,23(8):758-763.
[25]唐德尧.共振解调技术在机械故障诊断中的应用和发展[J].设备故障管理及维修,1991,12:22-24.
[26]王延春,谢明,丁康.包络分析方法及其在齿轮故障振动中的应用[J].重庆大学学报,1995,18(1):87-91.
[27]丁康,陈健林,苏向荣.平稳和非平稳振动信号的若干处理方法及发展[J].振动工程学报,2003,16(1):1-10.
[28]丁康,罗维,朱小勇等.提高扭转振动瞬态信号处理精度的方法研究[J].振动工程学报,2000,13(2):195-201.
[29][美]L 科恩.时频分析:理论与应用.西安[M]:西安交通大学出版社,2000.
[30]Wigner E P.On the quantum correction for thermody-namic equilibrium[J].Phys.Rew.,1932,40:748-759.
[31]Vill J.Theorie et al.applications delanotion designal analytique[J].Cables et al Transmission,1948,2A:61-74.
[32]MarkW D.Spectral analysis of the convolution and filtering of non-stationary stochastic process[J].J.Sound Vib.,1970,11(1):19-63.
[33]Classen T.C M,Mecklenbrauker W F G.The Wigner Distribution-part Ⅰ.Philip s Res.J.,1980,35:217-250.
[34]Classen T C M,Mecklenbrauker W F G.The Wigner Distribution-part Ⅱ.Philip s Res.J.,1980,35:276-300.
[35]Classen T C M,Mecklenbrauker W F G.The Wigner Distribution-part Ⅲ.Philips Res.J.,1980,35:372-389.
[36]何正嘉等.机械设备非平稳信号的故障诊断原理与应用[M].北京高等教育出版社,2001,
[37]陈端,刘树棠.基于离散GABOR变换的抑制交叉项的新方法[J].西安交通大学学报,1997,31(9):77-80.
[38]Cohen L.Generalized phase-space distribution functions[J].J.of Math Phys.,1966,(7):781-786.
[39]Daubechies I,Orthonormal base of compactly supported wavelet[J].Communication on Pure and Applied Mathematics,1988,Ⅻ:909-996.
[40]Daubechies I.The wavelet transform,time frequency localization and signal analysis[J].IEEE Tansactions on Information Theory,1990,65(5):961-1006.
[41]Mallat S G.A theory for multiresolusion signal decomposition:The wavelet representation[J].IEEE Transactions on Pat tern Analysis and Machine Intelligence,1980,11(7):674-693.
[42]唐贵基,王誉蓉,郭新富等.基于第二代小波变换的振动信号去噪与故障诊断[J].气轮机技术,2006,48(4):295-297.
[43]何岭松.小波函数性质及其对小波分析结果的影响[J].振动工程学报,2000,13(1):143-146.
[44]段晨东.基于第二代小波变换的混合小波变换降噪方法[J].中国机械工程.2007,18(14):1700-1702.
[45]段晨东,何正嘉.第二代小波降噪及其在故障诊断系统中的应用[J].2004,25(7):1341-1343.
[46]Norden E,Huang Zhengshen,Steven R Long,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proc.R.Soc.Lond.A,1998:903-996.
[47]Loh CH,et al.Application of the empirical mode decomposition Hilbert spectrum method to identify near-fault ground-motion characteristics and structural responses[J].Bullet in of the Seismo logacal,Society of America,2001,(91):1339-1357.
[48]Vasudevan K,etal.Empirical mode skeletonization of deep crustal seismic data:Theory and application[J].Journal of Geophysical Research Solid Earth,2000,10(5):7845-7856.
[49]Echeverria J C,et al.Application of empirical mode decomposition to heart rate variability analysis[J].Medical &Biological Engineering &Computing,2001,(39):471-479.
[50][美]Ingrid.Daubechies.小波十讲[M].北京:国防工业出版社,2004.
[51]姚天任,孙洪.现代数字信号处理[M].武汉:华中科技大学出版社,2003.
[52]Jaffard S.Meyer.Y etal.科学技术中的小波分析[M].北京:国防工业出版社,2006.
[53]陈哲,冯天瑾.小波神经网络研究进展及展望[J].青岛海洋大学报.1999,29(4): 663-66.
[54]Pati Y C and Krishnaprasad P S.Analysis and synthesis of feed forward neural network using discrete affine wavelet[J].IEEE trans on NN,1993,4(1):73-75.
[55]Zhang Qinghua and Benveniste A.Wavelet networks[J].IEEE trans on NN,1992,3(6):889-898.
[56]Baskshi B.R.,Stephanopotous.Wavelet-net:A Multiresolution,Hierarchical Neural Network with Localization Learning[J].American Institute chemical Engineering Journal,1993,39(1):57-81.
[57]Szu H H,Telfer B,Kadambe.B.Neural Network Adaptive wavelet for Signal Repersentstion and Classification[J],Optical Engineering,1992,31(9):1906-1907.
[58]Zhang Jun,Gilbert G.Walter,et al.Wavelet Neural Network for Function Learning[J].IEEE Transon signal Porcessing SP,1995,43(6):1485-1497.
[59]石卓尔,焦李成等.子波神经网络[C].西安:中国神经网络1993年学术大会论文集(上),1993,85-96.
[60]丁宇新,沈雪勤等.基于能量密度的小波神经网络[J].计算机学报,1997,20(9):832-838.
[61]何振亚等.用于信号逼近的自适应时延小波小波神经网络[J].电子科学学刊,1998,20(5):604-610.
[62]刘芳,王岭,李焦成.一种基于子波的模糊推理神经网络及其学习算法研究[J].电子学报,1998,26(4):41-45.
[63]Zhang Q.Using.Wavelet Network in Non-paranletric Estinlation[J].IEEE Transon Neural Network 1997,8(2):227-236.
[64]王岭,焦李成.区间估计的FWNN及其区间学习算法[J].电子学报,1998,26(4):41-45.
[65]姚骏等.基于离散小波变换的小波网络学习算法及其在心电信号识别中的应用[C].南京:CCNS'97(二),1997:605-608.
[66]徐晓霞等.基于正交最小二乘的小波神经网络.电子学报,1998,26(10):115-117.
[67]刘志刚,王晓茹,钱清泉.小波网络的研究进展与应用[J].电力系统自动化,.2003,27(6):73-81.
[68]谢平,刘彬.一种基于小波网络的切削刀具故障监测[J].机械工程学报,2002,38(2):108-111.
[69]李化,岳刚.基于自适应小波网络的气轮机发电机机组振动故障诊断方法的研究[J],电工技术学报,2000,15(3):57-60.
[70]吕柏权,李天铎.一种基于小波网络的故障检测方法[J].控制理论与应用,1998,15(5):802-805.
[71]成成,黄道.基于小波网络的非稳态故障诊断方法[J].华东理工大学学报,2000,26(5):481-486.
[72]何永勇,褚福磊,钟秉林.进化小波网络及其在设备状态预测中的应用[J].机械工程学报,2002,38(8):80-84.
[73]朱继梅.非稳态振动信号分析[J].振动与冲击.2000,19(1):86-87.
[74]杨建文,贾民平,许飞云等.基于经验模式分解的旋转机械振动信号降噪处理[J].中国工程学,2005,7(8):65-69.
[75]I.Daubechies,Orthonormal Bases of Compactly Supported Wavelets[J],Comm Pure and Appl Math,1988,41:909-996.
[76]李向武,韦岗.小波网络的动态系统辨识方法及应用[J]控制理论与应用,1998,15(4):494-500.
[77]吴耀军,陶宝祺,余慎芳.B样条小波网络[J].模式识别与人工智能,1996,9(3):228-233.
[78]水鹏朗,保铮,焦李成.一种基于小波神经网络的多尺度差分方程求解方法[J].电子科学学刊,1997,19(6):733-737.
[79]段晨东,张建丁.基于第二代小波变换的转子碰摩故障特征提取方法研究[J].机械科学与技术,2006,25(10):1229-1232.
[80]Sweldens W.The Lifting Scheme:A Construction Of Second Generation Wavelet Construction[J].SIAM J.Math.Aanal,1997,9(2):511-546.
[81]张荣标,胡海燕,冯友兵.基于小波熵的微弱信号检测方法研究[J].仪器仪表学报,2007,28(11):2079-2084.
[82]印欣运,何永勇,彭志科等.小波熵及其在状态趋势分析中的应用[J].振动工程学报,2004,17(2):165-169.
[83]张大波,郭丽,巴巍.基于第二代小波变化的心电信号分析算法[J].数据采集与处理,2006,21(4):450-453.
[84]殷丽莉.小波网络用于水下目标识别的研究[D].哈尔滨工程大学硕士学位论文,2007.
[2]陈进,姜鸣.高阶循环统计量理论在机械故障诊断中的应用[J].振动工学报,2001,14(2):125-134.
[3]陈茜,樊可清,邓志龙.含噪的超定问题及其在振源识别中的应用[J].微计算机信息,2006,18(1):223-225.
[4]Cao Siyuan,Chen Xiangpeng.The Second-generation wavelet transform and its application in denoising of seismic data[J].Applied Geophysics,2005,2(2):70-74.
[5]Cooley JW,Tukey JW.An algorithm for the machine calculation of complex Fourier series[J].Mathematics of Computation,1965,19(90):297-301.
[6]John C Burgess.On digital spectrum analysis of periodic signals[J].J.A coust.Soc.Am,1975,58(3):556-567.
[7]Carlo Offelli,Dario Petri.The influence of windowing on the accuracy of multi-frequency signal parameter estimation[J].IEEE Transactions on Instrumentation and Measurement,1992,41(2):256-261.
[8]朱利民,钟秉林,黄仁.离散频谱多点卷积幅值修正法的理论分析[J].振动工程学报,1999,12(1):120.
[9]丁康,江利旗.离散频谱的能量重心校正法[J].振动工程学报,2001,4(3):54-358.
[10]刘进明,应怀樵.FFT谱连续细化分析的富里叶变换法[J].振动工程学报,1995,8(2):162-166.
[11]谢明,张晓飞,丁康.频谱分析中相位和频率校正的一种新方法--相位差校正法[J].振动工程学报,1999,12(4):454-459.
[12]Ding Kang,Xie Ming.Phase difference correction method for phase and frequency in spectrual analysis[J].Mechanical Systems and Signal Processing,2000,14(5):835-843.
[13]Zhu Limin,Li Hanxiong,Ding H,et al.Noise influence on estimation of signal parameter from the phase difference of discrete Fourier tanseforms[J].Mechanical System s and Signal P rocessing,2002,16(6):991-1004.
[14]Ding Kang,Luo J iangkai,Xie Ming.Time-shifting correcting method of phase difference on discrete spectrum[J].Applied Mathematics and Mechanics,2002,23(7):819-827.
[15]Kay S.A.Fast and accurate single frequency estimator[J].IEEE Trans ASSP,1989,37(12):1987-1990.
[16]刘渝.正弦波频率快速成估计方法[J].数据采集与处理,1998,13(1):7-11.
[17]丁康,朱小勇.适用于加各种窗的一种离散频谱相位差校正法[J].电子学报,2001,29(7):987-989.
[18]丁康,江利旗.离散频谱综合相位差校正法[J].振动工程学报,2002,15(1):114-116.
[19]丁康,谢明等.基于复解析带通滤波器的复调制细化频谱分析方法研究[J].振动工程学报,2001,14(1):29-35.
[20]丁康,谢明等.复解析带通滤波器及其在解调分析中的应用[J].振动工程学报,2000,13(3):385-392.
[21]谢明,丁康.基于复解析带通滤波器的复调制细化谱分析的算法研究[J].振动工程学报,2002,15(4):479-483.
[22]Ranadall R.B.A new method of modeling gear faults[J].ASME Journal of Mechanical Design,1982,104:259-267.
[23]Mofadden P D.Detecting fatigue cracks in gears by amplitude and phase demodulation of the meshing vibration[J].J.of Vibration,Acoustics,Stress,and Reliability in Design,1986:108-165.
[24]樊可清,王玲,陈保平.基于小波分析的工程信号调制解调方法及其应用[J].电子与信息学报,2001,23(8):758-763.
[25]唐德尧.共振解调技术在机械故障诊断中的应用和发展[J].设备故障管理及维修,1991,12:22-24.
[26]王延春,谢明,丁康.包络分析方法及其在齿轮故障振动中的应用[J].重庆大学学报,1995,18(1):87-91.
[27]丁康,陈健林,苏向荣.平稳和非平稳振动信号的若干处理方法及发展[J].振动工程学报,2003,16(1):1-10.
[28]丁康,罗维,朱小勇等.提高扭转振动瞬态信号处理精度的方法研究[J].振动工程学报,2000,13(2):195-201.
[29][美]L 科恩.时频分析:理论与应用.西安[M]:西安交通大学出版社,2000.
[30]Wigner E P.On the quantum correction for thermody-namic equilibrium[J].Phys.Rew.,1932,40:748-759.
[31]Vill J.Theorie et al.applications delanotion designal analytique[J].Cables et al Transmission,1948,2A:61-74.
[32]MarkW D.Spectral analysis of the convolution and filtering of non-stationary stochastic process[J].J.Sound Vib.,1970,11(1):19-63.
[33]Classen T.C M,Mecklenbrauker W F G.The Wigner Distribution-part Ⅰ.Philip s Res.J.,1980,35:217-250.
[34]Classen T C M,Mecklenbrauker W F G.The Wigner Distribution-part Ⅱ.Philip s Res.J.,1980,35:276-300.
[35]Classen T C M,Mecklenbrauker W F G.The Wigner Distribution-part Ⅲ.Philips Res.J.,1980,35:372-389.
[36]何正嘉等.机械设备非平稳信号的故障诊断原理与应用[M].北京高等教育出版社,2001,
[37]陈端,刘树棠.基于离散GABOR变换的抑制交叉项的新方法[J].西安交通大学学报,1997,31(9):77-80.
[38]Cohen L.Generalized phase-space distribution functions[J].J.of Math Phys.,1966,(7):781-786.
[39]Daubechies I,Orthonormal base of compactly supported wavelet[J].Communication on Pure and Applied Mathematics,1988,Ⅻ:909-996.
[40]Daubechies I.The wavelet transform,time frequency localization and signal analysis[J].IEEE Tansactions on Information Theory,1990,65(5):961-1006.
[41]Mallat S G.A theory for multiresolusion signal decomposition:The wavelet representation[J].IEEE Transactions on Pat tern Analysis and Machine Intelligence,1980,11(7):674-693.
[42]唐贵基,王誉蓉,郭新富等.基于第二代小波变换的振动信号去噪与故障诊断[J].气轮机技术,2006,48(4):295-297.
[43]何岭松.小波函数性质及其对小波分析结果的影响[J].振动工程学报,2000,13(1):143-146.
[44]段晨东.基于第二代小波变换的混合小波变换降噪方法[J].中国机械工程.2007,18(14):1700-1702.
[45]段晨东,何正嘉.第二代小波降噪及其在故障诊断系统中的应用[J].2004,25(7):1341-1343.
[46]Norden E,Huang Zhengshen,Steven R Long,et al.The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].Proc.R.Soc.Lond.A,1998:903-996.
[47]Loh CH,et al.Application of the empirical mode decomposition Hilbert spectrum method to identify near-fault ground-motion characteristics and structural responses[J].Bullet in of the Seismo logacal,Society of America,2001,(91):1339-1357.
[48]Vasudevan K,etal.Empirical mode skeletonization of deep crustal seismic data:Theory and application[J].Journal of Geophysical Research Solid Earth,2000,10(5):7845-7856.
[49]Echeverria J C,et al.Application of empirical mode decomposition to heart rate variability analysis[J].Medical &Biological Engineering &Computing,2001,(39):471-479.
[50][美]Ingrid.Daubechies.小波十讲[M].北京:国防工业出版社,2004.
[51]姚天任,孙洪.现代数字信号处理[M].武汉:华中科技大学出版社,2003.
[52]Jaffard S.Meyer.Y etal.科学技术中的小波分析[M].北京:国防工业出版社,2006.
[53]陈哲,冯天瑾.小波神经网络研究进展及展望[J].青岛海洋大学报.1999,29(4): 663-66.
[54]Pati Y C and Krishnaprasad P S.Analysis and synthesis of feed forward neural network using discrete affine wavelet[J].IEEE trans on NN,1993,4(1):73-75.
[55]Zhang Qinghua and Benveniste A.Wavelet networks[J].IEEE trans on NN,1992,3(6):889-898.
[56]Baskshi B.R.,Stephanopotous.Wavelet-net:A Multiresolution,Hierarchical Neural Network with Localization Learning[J].American Institute chemical Engineering Journal,1993,39(1):57-81.
[57]Szu H H,Telfer B,Kadambe.B.Neural Network Adaptive wavelet for Signal Repersentstion and Classification[J],Optical Engineering,1992,31(9):1906-1907.
[58]Zhang Jun,Gilbert G.Walter,et al.Wavelet Neural Network for Function Learning[J].IEEE Transon signal Porcessing SP,1995,43(6):1485-1497.
[59]石卓尔,焦李成等.子波神经网络[C].西安:中国神经网络1993年学术大会论文集(上),1993,85-96.
[60]丁宇新,沈雪勤等.基于能量密度的小波神经网络[J].计算机学报,1997,20(9):832-838.
[61]何振亚等.用于信号逼近的自适应时延小波小波神经网络[J].电子科学学刊,1998,20(5):604-610.
[62]刘芳,王岭,李焦成.一种基于子波的模糊推理神经网络及其学习算法研究[J].电子学报,1998,26(4):41-45.
[63]Zhang Q.Using.Wavelet Network in Non-paranletric Estinlation[J].IEEE Transon Neural Network 1997,8(2):227-236.
[64]王岭,焦李成.区间估计的FWNN及其区间学习算法[J].电子学报,1998,26(4):41-45.
[65]姚骏等.基于离散小波变换的小波网络学习算法及其在心电信号识别中的应用[C].南京:CCNS'97(二),1997:605-608.
[66]徐晓霞等.基于正交最小二乘的小波神经网络.电子学报,1998,26(10):115-117.
[67]刘志刚,王晓茹,钱清泉.小波网络的研究进展与应用[J].电力系统自动化,.2003,27(6):73-81.
[68]谢平,刘彬.一种基于小波网络的切削刀具故障监测[J].机械工程学报,2002,38(2):108-111.
[69]李化,岳刚.基于自适应小波网络的气轮机发电机机组振动故障诊断方法的研究[J],电工技术学报,2000,15(3):57-60.
[70]吕柏权,李天铎.一种基于小波网络的故障检测方法[J].控制理论与应用,1998,15(5):802-805.
[71]成成,黄道.基于小波网络的非稳态故障诊断方法[J].华东理工大学学报,2000,26(5):481-486.
[72]何永勇,褚福磊,钟秉林.进化小波网络及其在设备状态预测中的应用[J].机械工程学报,2002,38(8):80-84.
[73]朱继梅.非稳态振动信号分析[J].振动与冲击.2000,19(1):86-87.
[74]杨建文,贾民平,许飞云等.基于经验模式分解的旋转机械振动信号降噪处理[J].中国工程学,2005,7(8):65-69.
[75]I.Daubechies,Orthonormal Bases of Compactly Supported Wavelets[J],Comm Pure and Appl Math,1988,41:909-996.
[76]李向武,韦岗.小波网络的动态系统辨识方法及应用[J]控制理论与应用,1998,15(4):494-500.
[77]吴耀军,陶宝祺,余慎芳.B样条小波网络[J].模式识别与人工智能,1996,9(3):228-233.
[78]水鹏朗,保铮,焦李成.一种基于小波神经网络的多尺度差分方程求解方法[J].电子科学学刊,1997,19(6):733-737.
[79]段晨东,张建丁.基于第二代小波变换的转子碰摩故障特征提取方法研究[J].机械科学与技术,2006,25(10):1229-1232.
[80]Sweldens W.The Lifting Scheme:A Construction Of Second Generation Wavelet Construction[J].SIAM J.Math.Aanal,1997,9(2):511-546.
[81]张荣标,胡海燕,冯友兵.基于小波熵的微弱信号检测方法研究[J].仪器仪表学报,2007,28(11):2079-2084.
[82]印欣运,何永勇,彭志科等.小波熵及其在状态趋势分析中的应用[J].振动工程学报,2004,17(2):165-169.
[83]张大波,郭丽,巴巍.基于第二代小波变化的心电信号分析算法[J].数据采集与处理,2006,21(4):450-453.
[84]殷丽莉.小波网络用于水下目标识别的研究[D].哈尔滨工程大学硕士学位论文,2007.