Implicit bulk-surface filtering method for node-based shape optimization and comparison of explicit and implicit filtering techniques

TL;DR

This paper introduces an implicit bulk-surface filtering method for shape optimization that improves boundary smoothness and mesh quality, demonstrating superior efficiency and consistency over explicit filters through numerical experiments.

Contribution

The work develops a novel implicit filtering technique based on PDEs for shape optimization, offering better efficiency and mesh independence compared to traditional explicit filters.

Findings

Implicit filtering is more efficient than explicit filtering.

The method preserves mesh quality during shape optimization.

Mesh-independent sensitivities can be achieved with proper scaling.

Abstract

This work studies shape filtering techniques, namely the convolution-based (explicit) and the PDE-based (implicit), and introduces an implicit bulk-surface filtering method to control the boundary smoothness and preserve the internal mesh quality simultaneously in the course of bulk (solid) shape optimization. To that end, volumetric mesh is filtered by the solution of pseudo-solid governing equations which are stiffened by the mesh-Jacobian and endowed with the Robin boundary condition which involves the Laplace-Beltrami operator on the mesh boundaries. Its superior performance from the non-simultaneous (sequential) treatment of boundary and internal meshes is demonstrated for the shape optimization of a complex solid structure. Well-established explicit filters, namely Gaussian and linear, and the Helmholtz/Sobolev-based (implicit) filter are critically examined in terms of consistency (rigid-body-movement production), geometric characteristics and the computational cost. It is shown that the implicit filtering is numerically more efficient and unconditionally consistent, compared to the explicit one. Supported by numerical experiments, a regularized Green's function is introduced as an equivalent explicit form of the Helmholtz/Sobolev filter. Furthermore, we give special attention to derive mesh-independent filtered sensitivities for node-based shape optimization with non-uniform meshes. It is shown that the mesh independent filtering can be achieved by scaling discrete sensitivities with the inverse of the mesh mass matrix.