cancel
Showing results for 
Search instead for 
Did you mean: 

Multiple random analyses in the same deck

Pioneer
Pioneer

Hello

 

In the Nastran v11 release notes they say you can now do multiple random analyses in the same deck. They provide an example.

 

In NX Nastran 11, the ANALYSIS case control command now includes the RANDOM subcase type for SOLs 108 and 111. With the new RANDOM type, you can streamline the process of performing multiple random analyses with the RANDOM and optionally RCROSS case control commands over frequency response subcases of the same frequencies.

For example, in SOL 111: 

SDAMPING=101
SPC=2
ACCELERATION(RPRINT)=ALL
STRESS(SORT2,PSDF)=ALL
RCROSS(PRINT)=1
METHOD=1
$
SUBCASE 1 $ Normal Modes
  ANALYSIS=MODES
  DISP=ALL
SUBCASE 2 $ Frequency Response using FREQ set 13
  FREQUENCY=13
  DLOAD=111
SUBCASE 3 $ Frequency Response using FREQ set 13
  FREQUENCY=13
  DLOAD=211
SUBCASE 4 $ Random Response referencing FRF from subcases 2 and 3
  ANALYSIS=RANDOM
  RANDOM=100
SUBCASE 5 $ Random Response referencing FRF from subcases 2 and 3
  ANALYSIS=RANDOM
  RANDOM=200
$ Change frequencies
SUBCASE 11 $ Frequency Response using FREQ set 23
  FREQUENCY=23
  DLOAD=111
SUBCASE 12 $ Frequency Response using FREQ set 23
  FREQUENCY=23
  DLOAD=211
SUBCASE 14 $ Random Response referencing FRF from subcases 11 and 12
  ANALYSIS=RANDOM
  RANDOM=1100  
SUBCASE 15 $ Random Response referencing FRF from subcases 11 and 12
  ANALYSIS=RANDOM
  RANDOM=1200

 

 

Has anyone managed to do this?

Even if I structure my case control as in the example, I cannot get any output for the random analysis even though there are no FATAL errors in the f06 file.

 

Any ideas what I am doing wrong?

Can you run multiple random analyses in the same deck?

8 REPLIES

Re: Multiple random analyses in the same deck

Phenom
Phenom

Have you got all the cards defined? can you post the full input deck (no model required)

Production: NX9.0.3.4, NX10.0.2.6
Development: VB.NET (amateur level !)

Re: Multiple random analyses in the same deck

Legend
Legend
We're too old school to have evolved to the fancy random vibe subcase way (at least we're not using seismic mass anymore :-)), but let us give it a shot and we'll report back...

Re: Multiple random analyses in the same deck

Pioneer
Pioneer

Morning

Sorry, I was away for a week.

The input deck is as per listing above, but referencing my DLOAD, RANDOM, SPC and METHOD card ids.

I get no random output in f06 file. Output is only the eigenvector tables.

 

Peter

Re: Multiple random analyses in the same deck

Phenom
Phenom

Not sure what you mean by random output. rms, zero crossing?

I think you only get random data if you use XYPLOT card. something like

 

                            X Y - O U T P U T  S U M M A R Y  ( R E S P O N S E )
SUBCASE  CURVE FRAME    CURVE ID./       XMIN-FRAME/    XMAX-FRAME/    YMIN-FRAME/      X FOR        YMAX-FRAME/      X FOR
   ID     TYPE   NO.  PANEL  : GRID ID     ALL DATA       ALL DATA       ALL DATA       YMIN           ALL DATA       YMAX
      1  ACCE      0      101(  4)      2.000000E+00   5.020000E+02   2.500000E-01   6.438968E+00   2.500000E-01   2.000000E+00
                                        2.000000E+00   5.020000E+02   2.500000E-01   6.438968E+00   2.500000E-01   2.000000E+00
      1  ACCE      0      149(  4)      2.000000E+00   5.020000E+02   1.527553E-01   5.020000E+02   3.915089E+00   1.198768E+01
                                        2.000000E+00   5.020000E+02   1.527553E-01   5.020000E+02   3.915089E+00   1.198768E+01

 

Having said that if you use say

ACCELERATION(SORT2,PLOT,REAL,RALL)

then you should get the PSD data point

 

 POINT-ID =         1
                                            A C C E L E R A T I O N    V E C T O R
                                                ( POWER SPECTRAL DENSITY FUNCTION )
 
      FREQUENCY   TYPE          T1             T2             T3             R1             R2             R3
   1.000000E+01     G      2.150474E-22   1.356574E-23   9.752309E+06   6.971814E-02   3.980307E-05   1.898200E-26
   1.200000E+01     G      4.459262E-22   2.813017E-23   9.811733E+06   1.453133E-01   8.302587E-05   3.936142E-26
   1.400000E+01     G      8.261416E-22   5.211516E-23   9.882576E+06   2.708569E-01   1.548983E-04   7.292264E-26

Production: NX9.0.3.4, NX10.0.2.6
Development: VB.NET (amateur level !)

Re: Multiple random analyses in the same deck

Pioneer
Pioneer

Hi

The request ACCELERATION(RPRINT)=1, where SET 1 = 100001 would give you output in the f06 for a normal random analysis

                                            A C C E L E R A T I O N    V E C T O R
                                      ( ROOT MEAN SQUARE; RMSSF SCALE FACTOR =  1.00E+00 )
 
      POINT ID.   TYPE          T1             T2             T3             R1             R2             R3
       1000001      G      7.589160E+01   0.0            0.0            0.0            0.0            0.0
1    IBDM VERSION 01-06                                                      AUGUST   8, 2017  NX NASTRAN 11/ 1/16   PAGE    23
      RANDOM DLL EXCITATION IN X AXIS                                                                                               
0                                                                                                            RANDOM  4000           
      FREQUENCY =  0.000000E+00
                                            A C C E L E R A T I O N    V E C T O R
                                                   ( NUMBER OF ZERO CROSSINGS )
 
      POINT ID.   TYPE          T1             T2             T3             R1             R2             R3
       1000001      G      7.643273E+02   0.0            0.0            0.0            0.0            0.0

In the multiple random analysis you get nothing.

 

Peter

Re: Multiple random analyses in the same deck

Pioneer
Pioneer

Wondering whether someone from NX Nastran would like to comment?

Does this feature work?

Re: Multiple random analyses in the same deck

Legend
Legend

PeterE wrote:

Wondering whether someone from NX Nastran would like to comment?

Does this feature work?


Sorry, we got distracted with work ( hate when this happens :-) ).  Let me get to it today and I'll report back. 

Re: Multiple random analyses in the same deck

Pioneer
Pioneer

I have attempted to attach sample runs. One with a single case which works, and another with multiple cases which doesn't.

Maybe you can show me where I am going wrong.