Jump to content


  • Content Count

  • Joined

  • Last visited

About pshivashankar

  • Rank

Profile Information

  • Are you University user?
  1. Haha, it doesn't seem too likely that Hypermesh or Optistruct would be wrong. Maybe an understanding of how Optistruct segregates design and non-spaces could elp.
  2. There is no displacement constraint: The objective is to minimize stress is adhesive. The constraint set is maximum stress in the insert. I am trying to figure this out too. What constraint violation & actually means?
  3. But I get different masses as shown above. What might be the reason?
  4. Hi Tinh, The goal is to optimize the thickness of the insert region subject to stress and mass constraints (I have read that simultaneous definition of mass and stress constraints is not a good idea. I'm still trying to a figure out a way around this). The insert itself has a network of ribs and stiffeners inside which isn't seen in the picture below. During the DV definition, I have chosen the PSHELL cards corresponding to only the insert. This would mean that my design space refers only to the insert and everything else in the model is the non-design space. Am I correct? Moreover, if the above assumption is correct, the FEM mass of the insert as seen in Hypermesh and the 'total mass of design material' as seen in the *.out file must be equal. Isn't it? Any inputs or critiques are highly appreciated. Thank you!
  5. Hi Tinh, Does such a definition pose a conflict? Do you also have any comments on the large constraint violation %? In this case, the two constraints defined are the allowable stresses in the adhesive and the adherend. We would, however, like the stresses developed to be far below the allowables. The objective has been set to minimize stresses in the adhesive only.
  6. Update: During the FE modeling in Hypermesh, design space and non-design space were not explicitly defined by the user. What is the basis for Optistruct to decide and distinguish between design and non-design spaces?
  7. Greetings, I am performing a free-size optimization of an insert. The thickness of the insert is chosen as the design variable. A minimum and maximum value are set during the DV definition as sown below: However, what value of the thickness does Optistruct choose as the initial guess value for the first iteration?
  8. Greetings, I am performing a free-size optimization (iteration summary given below) and trying to understand the explanation behind maximum constraint violation %. So far the following is clear: If constraint violation > 1% : status is V (violated) If constraint violation < 1% : status is A (Active) Large constraint violation % could mean that the constraint is too tight and that the problem itself is ill-defined. the difference between regular and soft convergence is understood as well Problem definition: The thickness of an insert is defined to be the DV. Two stress constraints are defined. The objective is a stress minmax formulation. Question: I am unable to figure out whether the large constraint violation % implies a critical issue. If yes, how does one rectify the problem? Insert_FEM_y.out
  9. Greetings, I have a FEM model whose mass calculated by Hypermesh (under Tools > masscalc) is equal to 2.22 kg I am running a free-size optimization of this model. From the design parameters summary reflected in the *.out file (shown below), the sum of design and non-design mass is equal to 2.51524 kg: After the iterations begin, the mass registered in iteration 0 is 2.34766 kg Question: Is it wrong to expect the following: FEM mass = (design + non-design mass) = mass at iteration 0? Why are these values different?
  • Create New...