Effect of a Low Volume Fraction

exx077
exx077 Altair Community Member
edited October 2020 in Community Q&A

Hello,

 

I am running a topology optimisation using Optistruct. I am trying to look at the key structural members of an aerospace component.

I am attempting to use a low volume fraction of ~3-6%.

When the volume fraction is particularly low does this effect the solver's ability to converge?

 

Thank you in advance

Tagged:

Answers

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited April 2018

    Hi,

     

    I think with lower volume fraction constraint solver has the freedom to use more volume for optimization. 

  • exx077
    exx077 Altair Community Member
    edited April 2018

    Hi @Prakash Pagadala,

     

    Could you explain what you mean by this?

    Altair Forum User said:

    has the freedom to use more volume for optimization

     

    Thanks

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited April 2018

    Meaning you are allowing more material to use for optimization, this may end up with a slightly heavier structure 

  • exx077
    exx077 Altair Community Member
    edited April 2018

    Hi @Prakash Pagadala,

     

    I think I understand.

    So rather than inspecting the topology of results with variable volume fractions with the same density cut off, is it more appropriate to inspect the results for a specific volume fraction optimisation with the same corresponding density value.

     

    I.e. If the volume fraction is 0.3, the density cut off for iso should also be 0.3?

     

    Thanks

  • Altair Forum User
    Altair Forum User
    Altair Employee
    edited April 2018
    Altair Forum User said:

    I.e. If the volume fraction is 0.3, the density cut off for iso should also be 0.3?

     

    So it is arriving at the best design. But relating ISO with Volume Fraction may not help always. 

     

    Looking at the results and judging the design or it should be based on experience. 

     

    I have gone through the model and I see you have used a VF with upper bound of 0.5 and no lower bound. Is there a reason?

  • exx077
    exx077 Altair Community Member
    edited April 2018

    @Prakash Pagadala,

     

    Altair Forum User said:

    I see you have used a VF with upper bound of 0.5 and no lower bound. Is there a reason?

    There is no particular reason for doing this.

    My understanding was that you set the upper limit of the volume fraction and the optimiser will be constrained to this value.

    I thought that if i set a upper and lower limit, then the simulation would just work to the upper limit of the volume fraction as this would enable it to minimise the compliance better.

     

    Thanks

Categories