Topology Optimization

  • Last Post 15 March 2020
fejaz1 posted this 24 February 2020


I am using the builtin tool in ANSYS for topology optimization. There is an objective called 'compliance' which is equal to the total strain energy OR the product of displacement of node with the nodal force (C = u' * F, or simply put, the stiffness of structure)

I am using a simple rectangular column and applying a force of 1 N in negative y direction (compressive force). I am also using volume constraint to retain 33% volume. Now, when I check the solution information, the objective value is different than the one I calculate manually by calculating the deformation/movement of node in negative y direction and multiplying by nodal force..

My question is, HOW ANSYS is calculating the objective or compliance? Can you please help me out.


Thank you

Order By: Standard | Newest | Votes
peteroznewman posted this 24 February 2020

Compliance is the inverse of Stiffness.

Every Optimization problem needs an objective. In some Topology Optimization problems, the objective is not that important, what is important is some constraint, such as a mass constraint or a stress constraint to keep one of those below a specified limit.

Were you able to bring the result from the optimization solution into a new geometry model?

fejaz1 posted this 15 March 2020

Hi Peter,

Thank you for responding. I think I did not articulate my problem properly.

While performing topology optimization, I am minimizing the compliance in my structure. Compliance is inverse of stiffness. When I click on SOLUTION INFORMATION tab, I check the final summary and objective is given there. I want to know, that 'objective' is inverse of stiffness or is it the strain energy density? One of your moderator answered here that it is strain energy density. Can you please clarify?

peteroznewman posted this 15 March 2020

The ANSYS Help system points to this reference, Bendsoe, M.P. and Sigmund O., Topology Optimization: Theory, Methods and Applications, Springer, Berlin, 2003, for information on the simple compliance optimality criteria used by the iterative solver.