Jump to content

Diana M.

Members
  • Content Count

    14
  • Joined

  • Last visited

  • Days Won

    1

Diana M. last won the day on May 9 2018

Diana M. had the most liked content!

1 Follower

About Diana M.

  • Rank
    Beginner

Profile Information

  • Are you University user?
    No

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Hello, It seems there is an error with the solver execution and consequently with the responses extraction. Did you look at the error file for more information? If you want you can export the archive (from menu File) and share it with me to check. Kind regards, Diana
  2. You are welcome. Thank you for your feedback!
  3. Hi Rhodel, Thank you for your interest in HyperStudy and GRSM. I thought you may be interested in looking at this white paper dealing with HyperStudy optimization algorithms benchmark: https://www.altair.com/resource/benchmark-of-hyperstudy-optimization-algorithms Kind regards, Diana
  4. Hi, Please find below the answers of your questions. Should i pick them both? Yes, if you are interested to figure out the optimal values for two of them. Once variables are imported in HyperStudy, you can specify the variables types (discrete or continuos) as well as the ranges of variations (it defines the search space for the optimizer). What about response? Once you have run the evaluation (OptiStruct solving), you can define the responses of interest. Should i find h3d which hyperstudy made and define element number and it stresses? The results can be extracted from the h3d generated by OptiStruct. This h3d file is easily read in HyperStudy by using the Readsim reader. I'd suggest you to Drag&Drop the h3d in HyperStudy working area. It expands the File assistant that guides you in the responses creation. For the stress results, you can either define a request with specific element numbers or keep the default First-Last. can i pick global stresses? Yes, You can extract any values available in the h3d. If needed you can manage vectors using the HyperStudy tools. Kind regards,
  5. Hi, Can you please specify the solver you are launching from HyperStudy, the tool versions, and what the expected output (or output file is)? If I understand correctly, the steps Define Models and Import Variables are done, but the Execution is failing and outputs are not generated. Is this correct? Best regards,
  6. Hi, Regarding the message in your the task_exe_stdout, it seems that the execution in MotionSolve has failed. In such case the task Execute in HyperStudy is displayed in red. Is this what you see? If you still face the issue to get *.mrf files written in the run folder, can you please specify the MotionSolve and HyperStudy versions you are using. Concerning the use of MotionSolve Data set options as design variables in HyperStudy study, I would say Yes if they are available for selection on MotionSolve side. Kind regards, Diana
  7. Hi Damian, I follow up on this request. Do you still encounter the issue to get *.mrf files written in the run folders? If yes, can you please look at the task_execute_stderr.txt and task_execute_stdout.txt files, which are available in the run folders. Can you please check if there are any errors recovered from MotionSolve execution? It can explain why *.mrf files are not written. I suggest you also to try the tutorial HS-1030, which is available in the HyperStudy help. It's another example with MotionSolve. Can you please check if the behavior is the same with this model? Kind regards, Diana
  8. For information, this modification has been introduced in FluxMotor 2019.1 version. Please see the details here http://fluxmotordoc.altair.com/helponlines/helponline/598/18
  9. Hi Roberto, I've just seen your replies as I didn't receive notification of yours. Sorry for that. The message Error 12 that stops your optimization in case of failed FluxMotor runs is due to the solver script. Please go in Edit->Register solver script and set the FluxMotor solver script to point to FluxMotors.exe. Note that there are two .exe in the same directory, one FluxMotor.exe and another FluxMotors.exe with "s". You need to point to the one with "s". Please let me know if it fix the issue with stopped optimization. Kind regards, Diana
  10. Hi, Could you check if you have the option "Ignore failed evaluations" enabled in Specifications ? Also, can you tell me which versions are you using? Regards,
  11. Hi Amasker, To smooth the curve in HyperStudy, you can use saefilter or polyfit functions available in the Expression builder. You would need to create data sources for each vector Time and Velocity and then create a Response defined with the filter expression. Hope it helps. Let us know if further questions. Diana
  12. Hello, I confirm you that you can create a response of f(x1,x2) as (var_2-(5*var_1^2)/(4*pi^2)+var_1/pi-6)^2+10*(1-1/8*pi)*cos(var_1)+10, where var_1 is x1 and var_2 is x2 It works: f(1,1)=36.4 What was not working in your case? Best regards
  13. Hello, In addition of Rahul's answers I would like to provide some additional information. 1) Let me come back to your initial request. You said that your optimization with GRSM did not provide a satisfying solution. I would suggest to investigate this. Please, find attached some good practices on GRSM use that may help you to get insight on your optimization. 2) Menu File --> Export archive allows packaging the study files in archive file (.hstx). 3) The values precision for continuous dvs is managed internally by HyperStudy. If you would like that dvs take specific values (with less decimals) you may prefer using discrete variables instead of continuous ones. Thus, you can specify the variation step (ex. 0.1) in the allowed range. Hope it helps. GRSM_good_practices.pdf
  14. Hello, I would like to suggest you using Altair HyperStudy optimization tool, which replaces GOT-It in the Altair software suite. The HyperStudy-Flux connection is operational with version 2018 and allows you performing the optimization you are describing. Please let us know if you need more information. Kind regards
×
×
  • Create New...