Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Criterion for Modell Validation/ callibration

Status
Not open for further replies.

engg35mechie

Automotive
Apr 28, 2015
14
0
0
DE
Hallo Group

I am new to NVH simulation.

I have a component tested for vibration. From the test involved measuring Frequency response function (Mobility)FRF on 6 points on the modell. So I have plots of FRF values at Six points on the modell. Using this I want to calibrate My FEM model (with hypermesh/optistruct/hyperstudy). What are techniques to callibrate such modell. I have come across Modal assurance criterion (MAC) and Model updating technique. can you enlighten me about such method.

Thanks and Regards
Pratik
 
Replies continue below

Recommended for you

Hello Greg


I have have been going through some post and then some scientific papers. And i believe that FDAC is possible way for correlating and Model updating in my case.
Reasons
1) It takes into account frequency shift
2)FRF can be directly used used for corelation without extracting modal parameter like (Mode shape , Damping etc)as against MAC. especially I dont know how exactly to extract mode shape from a FRF (Mobiltiy) data (attached file I have FRF at Six such points).

But i am confused as to how do you get the values of conjugate of the FRF (as refered in because all my FRF values are real coreesponding to each frequency.( I dont get which matrix should I transpose)
Also what exactly is FRF matrix (coloums= fRF value at various measured points and row = frequency steps where responses are measured

Thanks and Regards.
 
PS: I am validating the model in range 0 to 200 hz. Should I take the frequency values at resonance for validation with FDAC or any values at suitable interval will be good enough.
 
Status
Not open for further replies.
Back
Top