Multipoint base excitation random analysis Ansys V11 1

[srjose]

Dear all.
Reading the Ansys Help I find out that in order to take into account the cross correlations between different signals in a multipoint Random Analysis I should use following commands:

Command(s):

COVAL for cospectral values
QDVAL for quadspectral values
PSDSPL for a spatial relationship
PSDWAV for a wave propagation relationship

Could any expert in this topics explain me how to use this parameters in order to take into account the cross correlations during a multipoint base excitation random analysis?

If none of previous commands is defined, how are the signals? correlated or uncorrelated?

Thanks everybody in advance.

 

[Transient1]

I've never done multipoint random analysis. However, I do remember from the help that the results are uncorrelated if applied in multiple axes at once (i.e. x,y,z PSD's).

 

[srjose]

Thank you.

So you mean that if I don't specify anything Ansys understands that the signals are fully uncorrelated, I mean the option by default is that the signals are uncorrelated, aren't they?

If that's right one might assure that the simulation by default is the most conservative, isn't it?

 

[cbrn]

Hi,
as far as I know, the signals would be full-uncorrelated.
Whether it is the most conservative or not, instead, is another problem and, I'd say, you can't know that "a priori".
Without more information, what I'd personally do is to use some kind of correlation if-and-only-if I know that the "reality" of the problem needs it. If I have no information I'd rather go with "no correlation".

Regards