复合材料结构健康主动监测中激励信号的优化

385200610JournalofNanjingUniversityofAeronautics&AstronauticsVol.38No.5Oct.2006王帮峰　李　迎　施益峰(,,210016):利用Lamb波频散方程及其数值求解得到的频散曲线,对Lamb波模式及激励信号中心频率范围进行预测。建立了试验测试系统,利用集成在复合材料层合板上的PZT压电陶瓷片作为驱动器和传感器,激励并接收结构中激发的Lamb波信号。对具有不同函数形式、中心频率和波峰数的信号激励,结构中传播的Lamb波模式进行试验研究,为激励信号的选择与优化提供依据。利用Hilbert-Huang变换及Hilbert谱提取传感器信号特征,提出了Lamb波信号在结构中传播时的能量衰减率和损伤敏感度两个考察指标,并据此对激励信号进行优化。实验结果表明,优化后的激励信号在结构中能激发出单一模式的Lamb波(S0模式),有效地抑制了多模式现象的出现。同时,激发出的Lamb波具有最低的能量衰减和最高的损伤灵敏度,响应信号特征明显,便于损伤识别。:振动与波;结构健康监测;超声Lamb波;Hilbert-Huang变换;复合材料:TP206.1:A:1005-2615(2006)05-0613-06:(03I52076);(10072026,90405008):2005-09-05;:2005-12-13:,,,,19701,E-mail:[e-mail]alforStructuralHealthMonitoringonCompositeStructuresWangBangfeng,LiYing,ShiYifeng(AeronauticalScienceKeyLabofSmartMaterials&Structures,NanjingUniversityofAeronautics&Astronautics,Nanjing,210016,China)Abstract:Anoptimalexcitationsignalisproposedtoimprovetheeffectofdamagedetection.DispersionequationsanddispersioncurvesofLambwavepropagatinginalaminatedcompositeplateareutilizedtopredictthemodesofLambwaveandthedomainofthecenterfrequencyoftheinputexcitationsignal.Anexperimentaltestsystemissetup.Lambwavecanbegeneratedandreceivedbytwopiezoelectricce-ramicpatches(PZT)mountedonthesamesurfaceofcompositelaminateasanactuatorandasensor.TheLambwavepropagatingmodesareexperimentallyanalyzedunderthedifferentexcitationsignalpa-rametersthatincludethefunctionforms,centerfrequenciesandwavepeaknumber.ThecharacteristicsofsensorsignalsareextractedusingHilbert-HuangtransformandHilbertspectrum.BasedontheHilbertspectrum,twocriteriafortheenergyattenuationratioandthedamagesensitivityofLambwavepropagatinginthestructurearedefinedandusedtooptimizetheexcitationsignalexperimentally.Re-searchresultsshowthatthesinglemodeLambwave(S0)canonlybegeneratedinthestructurewiththeoptimalexcitationsignaltorealizethesuppressiononthemultipleLambwavemodes.TheLambwavegeneratedbytheoptimalsignalhasthelowestenergyattenuationratioandthehighestdamagesensitivi-ty,soitismoresuitablefordamagedetection.Keywords:oscillationandwave;structuralhealthmonitoring;ultrasonicLambwave;Hilbert-Huangtransform;compositematerial,,,-(Acousto-ultrasonics)Lamb,,,,[1,2]Lamb,,,Lamb,[3-7],Lamb,,,,Lamb,Lamb,,,LinX[7],VeidtM[8,9],Lamb,,Lamb;[10]Lamb,,Lamb,Lamb(PZT5),,Lamb,Lamb1LambLamb,Lamb1.1Lamb,,d,Lamb[9]tan(d/2)tan(d/2)=-4k2(k2-2)2(1a)tan(d/2)tan(d/2)=-(k2-2)24k2(1b):=(2/C2L)-k2(2)=(2/C2S)-k2(3),=2!f;k,k=/CP,CPLamb;CL;CSLamb,k,CP=/k,(1)LambCP-()fd,-,CG=d/dk(4)-k,(4),,[10]Lamb,Lamb1.2(2,3)(1),tan!fdC2P-C2SCPCStan!fdC2P-C2LCPCL=-4C3S(C2P-C2L)(C2P-C2S)CL(2CS-C2P)2(5a)tan!fdC2P-C2SCPCStan!fdC2P-C2LCPCL=-CL(2CS-C2P)24C3S(C2P-C2L)(C2P-C2S)(5b)(5),CLCS61438,CL=E(1-∀)#(1+∀)(1-2∀)(6)CS=E/[2#(1+∀)](7):E;#;∀,,1CL=3.5m/ms,CS=2.2m/ms1E/GPa#/(gcm-3)∀b/mm201.750.153,11-1:,,,;;S0A0,,Lamb(380kHz)S0A0,2Lamb,2YE5850PCNIPCI-6110I/OSMSV1.0,1000mm500mm3mm(),110mm1mm,150mm2:SMS,PCI-6110I/O,Lamb;,PCI-6110I/O,Lamb,SMS,33.1Lamb,,,,Lamb,Morlet,V(t)=AH(t)-H(t-nfc)(1-cos2!fctnsin2!fct(8):A;n;fc;H(t)HeavisideMorlet[4],Mor-let6155,:V(t)=Ae-2!(fctn-12)22cos(2!fct-n!)(9)(8,9),3A=10,fc=125kHzMorlet[11],n=53:,(),,,,30kHz,46kHz,,Lamb,Lamb,n,n,,,,,n,,n=53.2Lamb,,,300kHz25kHz250kHz,510kHz44Lamb,300kHz,A0S0Lamb,1,50kHz,LambA0,S0,,S05065kHz,,S0,A0,70kHz,S0,,50kHz70kHzA0,S0S0Lamb,A0,,,S0Lamb61638,S0,Lamb,,RALambDS,RARA=1-H2h(,t0)dH2e(,0)d(10)LambDSDS=1-H2d(,t0)dH2h(,t0)d(11):Hh(,t0)LambHHTHilbert[12];He(,0)HHTHilbert;Hd(,t0)LambHHTHilbert3.3HHTRADS,HHTHilbertHHTHilbert-HuangtransformHilbert-HuangNEHuang1998,HHT(Empiricalmodedecomposition,EMD)Hilbert,EMDf(t)(Intrinsicmodefunction,IMF)EMD,Cj(t)rn(t)f(t)=nj=1Cj(t)+rn(t)(12)IMFHilbertf(t)=Renj=1aj(t)eij(t)dt(13)aj(t)=Cj(t)2+CHj(t)2j(t)=d∃j(t)/dt∃j(t)=arctan[CHj(t)/Cj(t)]CHj(t)Cj(t)HilbertIMF-,HilbertH(,t)(HHT[12])HHTHilbert,(10,11)H2(,t0)dS0Lambt0,0,RAS0Lamb,;DSS0Lamb,75175kHz,,,:,,,Lamb,,,HHT,5RADS5125kHz,Lamb125kHz54(1),Lamb,;S0A0,,S0A0(2)Morlet,,,,6175,:(3),Lamb,,(50kHz),LambA0;(70kHz),S0;,,,S0(4),LambLamb,Lamb:[1]CawleyP,AlleyneD.TheuseofLambwavesforthelongrangeinspectionoflargestructures[J].Ul-trasonics,1996,34(3):287-290.[2]ToyamaN,TakatsuboJ.Lambwavemethodforquickinspectionofimpact-induceddelaminationincompositelaminates[J].CompositesScienceandTechnology,2004,64(9):1293-1300.[3]IhnJ,ChangFK.Detectionandmonitoringofhid-denfatiguecrackgrowthusingabuilt-inpiezoelectricsensor/actuatornetwork:I.Diagnostics[J].SmartMaterialandStructures,2004,13(3):609-620.[4]SuZ,YeL.Lambwave-basedquantitativeidentifi-cationofdelaminationinCF/EPcompositestruc-turesusingartificialneuralalgorithm[J].CompositeStructures,2004,66(4):627-637.[5]SohnH,ParkG,WaitJR.Wavelet-baseactivesensingfordelaminationdetectionincompositestructures[J].SmartMaterialandStructures,2004,13(1):153-160.[6]SuZ,YeL,BuX.Adamageidentificationtech-niqueforCF/EPcompositelaminatesusingdis-tributedpiezoelectrictransducers[J].CompositeStructures,2002,57(1):465-471.[7]LinX,YuanFG.DiagnosticLambwaveinaninte-gratedpiezoelectricsensor/actuatorplate:analyticalandexperimentalstudies[J].SmartMaterials