calc of root mean square after random analysis

Johannes_ACKVA

Bonjour,

(question 1)

after a random analysis (DYNA_ALEA_MODAL) and transforming back the spectra to physical coordinates (with REST_SPEC_PHYS) I make

Table=POST_DYNA_ALEA(..)

IMPR_TABLE(TABLE=Table)

to get the statistical properties. There I do not find the root-mean-square value. But it seems that the parameter ECART equals the root-mean-square multiplied with sqrt(2). Can somebody please confirm this?

I m a bit surprised, because to my knowlegde until now I believe(d?) that the rms-value is one of the most important outputs after the random analysis.

(question 2)

If I want to make a coulored plot of the rms-values of stress then there is no CALC_CHAMP doing this. It seems that I must calc the spectra at all elements, postprocess them all with POST_DYNA_ALEA, and then extract all the ECART-values and build a field with them. Am I wrong (I hope!) and is there an easier way? Is there an example in astest (I didn't find it)?

Regards

Johannes_ACKVA

______________________________________________________________________

CODE-ASTER-courses at Ingenieurbüro für Mechanik, Germany

*** CODE-ASTER INTRO + CONTACT + MATERIAL

5 days, Mon 08 to Fri 12 of Dec 2014

*** CODE-ASTER DYNAMIC ANALYSIS

2 days, Thur 15 to Fri 16 of Jan 2015

*** CODE-ASTER PYTHON

3 days, 25 to 27 of Feb 2015

zhudongyu

Johannes_ACKVA Excuse me, do you know how to calculate the stress results of random vibration? Is it convenient to share an example?

razvan.curtean

I'm also using the Code Aster to simulate random vibrations. I can confirm that the ECART value it is standard deviation (1 sigma ) and in hypothesis of random vibration (mean value it is always zero) it is equal with rms value as well.

To have coloured pictures you can scale displacement calculated in modal analysis considering 1 sigma displacement and can read 1 sigma stress as well. By the way, how did you calculated 1 sigma stress? If you want I can give you one small example.

Have a nice day!

Razvan

zhudongyu

razvan.curtean Excuse me, do you know how to calculate the stress results of random vibration? Is it convenient to share an example?

Johannes_ACKVA

hallo Razvan,

many thank You for Your answer.

(1)

I have still doubts about ECART and the rms-value, I still think they are not identic but differ by a factor sqrt(2)=1.414. I have prepared a Code-Aster run creating a very simple Power Spectal Density: Fmin=10, Fmax=110, Amplitude=1=constant. As far as I know the rms-value is the square root of the integration area under the spectrum. This area is 100, the square root is 10. But in .resu You can find: ECART=1.41421E+01

(2)

About the colored plots I would be very thankful if You can give me an example, I hope then to understand or will come back asking here,

Regards

Johannes_ACKVA

______________________________________________________________________

CODE-ASTER-courses at Ingenieurbüro für Mechanik, Germany

*** CODE-ASTER INTRO + CONTACT + MATERIAL

5 days, Mon 08 to Fri 12 of Dec 2014

*** CODE-ASTER DYNAMIC ANALYSIS

2 days, Thur 15 to Fri 16 of Jan 2015

*** CODE-ASTER PYTHON

3 days, 25 to 27 of Feb 2015

Attached file:
create-PSD-spectra_ToForum.tar.gz, 5194kb

razvan.curtean

A quick answer.

You are right, but the RMS can be defined also as the second momentum formula (see attached) and medium value it is always zero. Than RMS = 1 sigma.

Actually is similar with harmonic, 1 sigma of acceleration (or measure unit) input and 1 sigma of acceleration output (or measure unit).

I will send you one of my examples.

Razvan

Attached file:
rms.png, 18966kb

Mitchel

Hi Johannes_ACKVA and razvan.curtean,

I'm sorry to intrude but I'm facing the same problem concerning the rms stress plot, could you please explane a bit more how you faced the teme.

Regards,

Mitchel

zhudongyu

Mitchel you can see SDLD23 example and Spectral moment n°0 value is root mean square.the paramer is LAMBDA_00.
now i have a question is how to get all nodes and get med file data.and render in paravis