傳遞路徑分析1(LMS網(wǎng)絡(luò)研討會(huì)講義)_圖文

1、yk The seminar will start at 10:00, Central European Standard Time (GMT +01:00, Paris Event number: 664 541 116 Voice over IP for the audio part or list of phone number in registration approval e-mail Transfer Path Analysis Web seminar 1/2 Methods and best practices for test-based Transfer Path Anal

2、ysis Laurent Britte Test Division Agenda 1 2 3 4 5 Introduction FRFs Measurement Loads Identification Multi-Reference TPA Challenges & Solution in LMS Test.Lab 2 copy right LMS International - 2008 Copyright LMS International 1 Agenda 1 2 3 4 5 Introduction FRFs Measurement Loads Identification

3、Multi-Reference TPA Challenges & Solution in LMS Test.Lab 3 copy right LMS International - 2008 Transfer Path Analysis Introduction structurestructure-borne airair-borne structurestructure-borne Transferpath analysis quantifies and visualizes the strengths of selected sources and their contribut

4、ion via multiple transmission paths to a selected receiver signal 4 copy right LMS International - 2008 Copyright LMS International 2 Transfer Path Analysis Source-transfer-receiver approach yk 80.00 20.00 Source (Fi,Qj X Transfer (NTF = Receiver (yk y k = NTFik * Fi + NTF jk * Q j i =1 j=1 n p | |

5、structural 5 copy right LMS International - 2008 airborne Transfer Path Analysis Source-transfer-receiver approach Which connections are putting most energy into the system ? Which path are transferring most energy ? How do the different path interact with each other ? Is it a airborne or a structur

6、e born problem ? Is it a system problem or a source problem ? ! Receiver Response: noise vibrations = = = Transmitter System characteristics: structural acoustic X X X Source Operating loads: structural acoustic critical loads critical dynamics worst case scenario 6 copy right LMS International - 20

7、08 Copyright LMS International Pa2 dB 3 Transfer Path Analysis Applications Upfront Upfront Engineering Engineering Test based applications Benchmarking Target setting CAE based applications Concept Analysis Load Identification Detailed Detailed Engineering Engineering CAE based contribution analysi

8、s Hybrid based contribution analysis System engineering What-if games and optimization Refinement Refinement Engineering Engineering (Fast Trouble shooting Critical component/path identification & assessment Auralization Efficient optimization Test Based TPA Accuracy requirements depending on ob

9、jectives Virtual Contribution Analysis Test Based TPA Hybrid TPA: Combining Test & CAE 7 copy right LMS International - 2008 Transfer Path Analysis Applications Upfront Upfront Engineering Engineering Detailed Detailed Engineering Engineering Refinement Refinement Engineering Engineering Benchma

10、rking Target setting Load Identification 8 copy right LMS International - 2008 Copyright LMS International 4 Transfer Path Analysis Applications Upfront Upfront Engineering Engineering Detailed Detailed Engineering Engineering Refinement Refinement Engineering Engineering System engineering What-if

11、engineering 9 copy right LMS International - 2008 Transfer Path Analysis Applications Upfront Upfront Engineering Engineering Detailed Detailed Engineering Engineering Refinement Refinement Engineering Engineering (Fast Trouble shooting Critical component/path identification & assessment Auraliz

12、ation Efficient optimization 10 copy right LMS International - 2008 Copyright LMS International 5 Measurements Transfer functions copy right LMS International -2008 EngineIntake GearBodyF 2F 1T = F 1x FRF 1+ F 2x FRF 2FRF measurement withoutremoving sourceOnly valid if Measurements Transfer function

13、s Removal of source Operational conditionCorrect method for FRF Measurements -LoadsStructural Acoustic MeasurementsiishelliXAQ&&&.=cIAcii1414=Force identification: mount stiffness methodForce identification: matrix inversion method right LMS International -2008Force identification: matri

14、x inversion method Indicator MicrophonesQ inTransfer Path Analysis Force identification: matrix inversion method Structural Indirect : Indirect indicator Finput 1001 1 1002 Lab measurements Step 1 HXQ . Local FRFs are measured between suspension and indicator accelerometers Operational measurements

15、Step 2 Operating accelerations in the vicinity of suspension (Pa Step 3 &&1001 x & x&1 &&1002 x F = H susp 1 . xF . . x 1 F1oper x1 && F1 x && = 1001 F1 && x1002 F1 &&1 x x oper &&1001 &&1002oper x oper 21 copy right LMS Interna

16、tional - 2008 Transfer Path Analysis Force identification: OPAX Intake Engine aai Fi api Body . Gear active side LMS Test.Lab OPAX Reduced parametric load models (Broadband model Reduced number of indicators passive side Fi ( = f ( parameters , a ai (, a pi ( p = H i fi + H k Qk i 22 copy right LMS

17、International - 2008 k Copyright LMS International 11 Transfer Path Analysis Load identification methods Direct measured forces and acoustic strength Mount stiffness method (K.x Nozzle noise Single source, multiple indicators Matrix inversion Multiple source, Multiple indicators Note: multiple sets

18、can be defined in 1 analysis 23 copy right LMS International - 2008 Agenda 1 2 3 4 5 Introduction FRFs Measurement Loads Identification Multi-Reference TPA Challenges & Solution in LMS Test.Lab 24 copy right LMS International - 2008 Copyright LMS International 12 Multi-reference crosspower measu

19、rements PY = S2 S1 When using 1 reference X2 : average( XPPY APX 2 Part of signal PY non-coherent with X2 disappears 25 copy right LMS International - 2008 If references perfectly uncorrelated X1 = S1 X2 = S2 X 1' X 1'* 0 X ' X '* = L 0 0 ' '* X2 X2 L 0 '* Y1 X 1'* Y1

20、 X 2 '* L 0 Y X '* Y2 X 2 YX '* = 2 1 L L L L ' '* '* L X nX n Ym X 1'* Ym X 2 L 0 '* L Y1 X n '* L Y2 X n L L '* L Ym X n 26 copy right LMS International - 2008 Copyright LMS International 13 If references partially correlated X1 = S1 S2 X2 = S1 S2 * X1 X1 X

21、X* XX * = 2 1 L * X n X1 * * X 1X 2 L X1X n * * X 2X 2 L X2 Xn L L L * * X nX 2 L Xn Xn * Y1 X 1* Y1 X 2 * Y X * Y2 X 2 YX * = 2 1 L L * * Ym X 1 Ym X 2 * L Y1 X n * L Y2 X n L L * L Ym X n 27 copy right LMS International - 2008 Singular value decomposition “slave” measurements Y2 Y1 Y3 X2 X1 Y2 Pri

22、ncipal component decomposition Y1 Y3 X2 X 1 Partially coherent reference set Orthogonal reference set 28 copy right LMS International - 2008 Copyright LMS International 14 Reference selection: 2 approaches 1. Capture all energy that enters car References = wheel center vibrations PCs structural phen

23、omena PLUS - can also be used for ride comfort MIN - many structural phenomena (usually 6-8 - not all of them generate noise - ranking in importance for wheel center vibration energy 2. Capture all energy that is relevant inside the car References = interior microphones PCs acoustic phenomena PLUS -

24、 little number of acoustic phenomena (1-2 - ranking in importance of interior acoustic energy 29 copy right LMS International - 2008 Agenda 1 2 3 4 5 Introduction FRFs Measurement Loads Identification Multi-Reference TPA Challenges & Solution in LMS Test.Lab 30 copy right LMS International - 200

25、8 Copyright LMS International 15 Transfer Path Analysis Challenges and pains Transfer Path Analysis Pains Data management Complexity Pains Pains Evaluation of multiple designs 31 copy right LMS International - 2008 LMS Test.Lab Transfer Path Analysis TPA Model definition 32 copy right LMS Internatio

26、nal - 2008 Copyright LMS International 16 LMS Test.Lab Transfer Path Analysis FRF Selection 33 copy right LMS International - 2008 LMS Test.Lab Transfer Path Analysis Operational Data Selection 34 copy right LMS International - 2008 Copyright LMS International 17 LMS Test.Lab Transfer Path Analysis

27、TPA Loads: Direct, OPA, OPAX, Mount Stiffness, Matrix Inversion 35 copy right LMS International - 2008 LMS Test.Lab Transfer Path Analysis TPA Results TPA Results: Single Path visualization TPA Results: Section Comparison 36 copy right LMS International - 2008 Copyright LMS International 18 LMS Test

28、.Lab Transfer Path Analysis TPA Results 4D-Display 37 copy right LMS International - 2008 LMS Test.Lab Transfer Path Analysis TPA Results 4D-Display Contr ibuti n at case ( path or group vs. rpm or fr equency o 60.00 Pa dB( A A Contr ibuti n at rpm or fr equency ( path or group vs. case o 60.00 A 10

29、.00 S pectr um PRCM 01:S : :00 S pectrum : PRCM:0001:S . S pectrum : PRCM:0002:S . S pectrum : PRCM:0003:S . S pectrum : PRCM:0004:S . S pectrum : PRCM:0005:S . S pectrum : PRCM:0006:S . RMS S um . 10.00 10.00 50.00 Measu red subf:502 0:X be_r:500 9:X subf :5020:Y subf:502 0:Z be_r: 9:X dB( A be_r :

30、5009:Y sh_r:503 2:X sh _r:32:Y sh_f:503 0:Z Sp ectrum: P RCM:0001:S . Pa Pa dB (A 9 1.00 Hz 350.00 60.00 Measu red be_f:1 8:X be_f:1 8:Z be_f :5018:Y be_r: 9:X be_r: 9:Z be_r :5009:Y sh_f:3 0:X sh_f:3 0:Z sh_f :5030:Y sh_r:3 2:X sh_r:3 2:Z sh_r :5032:Y subf:2 0:X subf:2 0:Z subf :5020:Y 10.00 -40.00

31、 Sp ectrum: P RCM:0002:S . Sp ectrum: P RCM:0003:S . Sp ectrum: P RCM:0004:S . Sp ectrum: P RCM:0005:S . Sp ectrum: P RCM:0006:S . RMS S um -40.00 Hz 350.00 Contribution Analysis Case by Case RPM or Frequency Contribution Analysis Contr ibuti n at path or grou p (case vs. rpm or fr equency o 60.0 0

32、dB( A Pa A V ector co ntribution dB (A 60.0 0 10.0 0 Pa A S pectru m P RCM:00 01:S : Spec trum: P RCM: 0001:S . Spec trum: P RCM: 0002:S . Spec trum: P RCM: 0003:S . Spec trum: P RCM: 0004:S . Spec trum: P RCM: 0005:S . Spec trum: P RCM: 0006:S . RM S Su m . 10.0 0 Measu red . Total be_f :5018:X . b

33、e_f:5018 :Y . sh_f :5030:X . sh_f:5030 :Y . be_r :5009:X . be_r :5009:Y . be . _f:18:X be_f:18 :Y . sh . _f:30:X sh_f:30 :Y . sh . _r:32:X be_f :5018:Z . sh_f :5030:Z . sh_r :32:Y . b . e_r:9:X be_r :9:Y . be_r :5009:Z . be . _f:18:Z sh . _f:30:Z b . e_r:9:Z sh_r :32:Z subf :5020:X . subf:5020 :Y . sh_r :5032:X . sh_r :5032:Y . su . bf:20:X subf:20 :Y . subf :5020:Z . sh_r :5032:Z . su . bf:20:Z sh . _r:32:Z dB( A 60.0 0 Pa -40.00 99. 00 0.0 0 Hz 50.00 b e_f:5018 :X 350.00 50.0 0 RMS S um Spect rum: PRCM:0001:S Spect rum: PRCM:0002:S d