# Shape optimization in FEM

**URL:** <https://discourse.julialang.org/t/shape-optimization-in-fem/107635>\
**Category:** Modelling & Simulations\
**Tags:** question, mesh, finite-element, autodiff, differentialequation\
**Created:** [December 14, 2023, 10:41pm UTC](https://discourse.julialang.org/t/shape-optimization-in-fem/107635 "2023-12-14T22:41:36Z")\
**Posts on this page:** 1\
**Showing post:** 7

<div class="post-metadata">

**Author:** ![instant0n](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/instant0n/32/205850_2.png) [@instant0n](https://discourse.julialang.org/u/instant0n)\
**Post date:** [December 29, 2023, 4:01pm UTC](https://discourse.julialang.org/t/shape-optimization-in-fem/107635/7 "2023-12-29T16:01:19Z")

</div>

Hi @tmigot, can the objective functional have finite support (point evaluations, rather than integrals over the entire domain)? And how is speed?

It sounds like this might be a solution for the problem I described here [Computing sensitivity to Dirichlet data in Gridap.jl](https://discourse.julialang.org/t/computing-sensitivity-to-dirichlet-data-in-gridap-jl/108145)

---

_[View the full topic](https://discourse.julialang.org/t/shape-optimization-in-fem/107635)._