Local Search Framework for Topology Optimization Based on Singular Value Decomposition
Year: 2026
Type: article
Abstract: Topology optimization frequently produces multiple structurally different solutions with comparable performance, thereby making essential geometric features difficult to identify. This study developed a local search framework based on singular value decomposition to systematically extract dominant shape features from multiple optimized designs and construct a reduced design space. This reduced space not only helps identify essential geometric features but can also be reused when design requirements are slightly modified. Numerical results for wireless power transfer devices showed that high efficiency was maintained even under an additional constraint.
Source: IEEE Transactions on Magnetics
Institutions: Nagasaki University, Hokkaido University, Omron (Japan)
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,645
OpenAlex ID: W7214768930