Topology optimization for heat conduction using generative design algorithms

Danny J. Lohan, Ercan M. Dede, James T. Allison

Research output: Contribution to journalArticlepeer-review


In this article we present a new approach to topological design for steady-state heat conduction. The method capitalizes on the use of a generative algorithm to represent topology, resulting in a decrease in the number of variables in the design description. Using a generative algorithm as a design abstraction, the optimization technique is targeted to dendritic topologies that are known to perform well for heat conduction. Specifically, a traditional topology optimization technique (SIMP) is confirmed to produce branching characteristics in optimal designs. The Space Colonization Algorithm, which can generate similar topological patterns, is selected for in-depth investigation. A genetic algorithm drives generation of design candidates, providing a highly diversified search of the target design space. Finally, several synthesized optimal designs for steady-state heat conduction, derived using the described algorithms, are compared using commercial finite element software.

Original languageEnglish (US)
Pages (from-to)1063-1077
Number of pages15
JournalStructural and Multidisciplinary Optimization
Issue number3
StatePublished - Mar 1 2017


  • Conductive heat transfer
  • Generative algorithms
  • Topology optimization

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Control and Optimization


Dive into the research topics of 'Topology optimization for heat conduction using generative design algorithms'. Together they form a unique fingerprint.

Cite this