Adaptive scaled boundary finite element method for two/three-dimensional structural topology optimization based on dynamic responses
Rut Su,Xiaoran Zhang,Sawekchai Tangaramvong,Chongmin Song
DOI: https://doi.org/10.1016/j.cma.2024.116966
IF: 6.588
2024-04-08
Computer Methods in Applied Mechanics and Engineering
Abstract:This paper presents an efficient solution for designing the topology of structures that can withstand dynamic loads. The method, called image/stereolithography (STL)-based adaptive scaled boundary finite element (SBFE), is a novel approach to topology optimization (TO) that is particularly effective for designing two/three-dimensional structures. The SBFE-based TO algorithm adopts a bi-evolutionary structural optimization (BESO) approach, which utilizes image/STL-based procedures to automatically adapt meshes and minimize mechanical compliance under varying volume fractions in the time domain. A key advantage of this approach is that it uses images of TO responses, rather than specific error functions, to guide the automatic adaptive mesh schemes. The method also incorporates a quadtree/octree mesh refinement to address the problem of hanging nodes, ensuring fast mesh convergence during the optimization process. Furthermore, a high-order implicit time integration technique provides a close approximation of equivalent static loads, allowing for the accurate mapping of displacement fields under actual dynamic responses at each time step. The effectiveness of this method is demonstrated through a variety of numerical examples and benchmarks.
mechanics,engineering, multidisciplinary,mathematics, interdisciplinary applications