A “poor man’s” approach for high-resolution three-dimensional topology design for natural convection problems

Nicolo Pollini, Ole Sigmund, Casper Schousboe Andreasen, Joe Alexandersen*

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

This paper treats topology optimization of natural convection problems. A simplified model is suggested to describe the flow of a steady-state incompressible fluid, similar to Darcy’s law for porous media. By neglecting inertia and viscous boundary layers, the flow model is significantly simplified. The fluid flow is coupled to the thermal convection-diffusion equation through the Boussinesq approximation. The coupled non-linear system of equations is discretized with stabilized finite elements and solved in a parallel framework that allows for the optimization of high-resolution three-dimensional problems. A density-based topology optimization approach is used, where the permeability and conductivity of the distributed material is interpolated. Due to the simplified model, the proposed methodology significantly reduces the computational effort required in the optimization. At the same time, it is notably more accurate than even simpler models that rely on Newton’s law of cooling. The methodology discussed herein is applied to the optimization-based design of three-dimensional heat sinks. The final designs are compared with previous work obtained from solving the full set of Navier–Stokes equations, both in terms of design performance and computational cost. The computational time is shown to be decreased to around in terms of core-hours, allowing for the possibility of optimizing designs during the workday on a small computational cluster and overnight on a high-end desktop. However, due to the use of a simplified model, the performance of the final designs are evaluated using the full Navier–Stokes equations. This ensures verification of performance, as well as systematic comparison with reference results.
Original languageEnglish
Article number102736
JournalAdvances in Engineering Software
Volume140
Number of pages18
ISSN0965-9978
DOIs
Publication statusPublished - 2019

Keywords

  • Topology optimization
  • Conjugate heat transfer
  • Natural convection
  • High performance computing
  • Simplified model

Cite this

@article{95f50ae56e944274994ec550f4be71d4,
title = "A “poor man’s” approach for high-resolution three-dimensional topology design for natural convection problems",
abstract = "This paper treats topology optimization of natural convection problems. A simplified model is suggested to describe the flow of a steady-state incompressible fluid, similar to Darcy’s law for porous media. By neglecting inertia and viscous boundary layers, the flow model is significantly simplified. The fluid flow is coupled to the thermal convection-diffusion equation through the Boussinesq approximation. The coupled non-linear system of equations is discretized with stabilized finite elements and solved in a parallel framework that allows for the optimization of high-resolution three-dimensional problems. A density-based topology optimization approach is used, where the permeability and conductivity of the distributed material is interpolated. Due to the simplified model, the proposed methodology significantly reduces the computational effort required in the optimization. At the same time, it is notably more accurate than even simpler models that rely on Newton’s law of cooling. The methodology discussed herein is applied to the optimization-based design of three-dimensional heat sinks. The final designs are compared with previous work obtained from solving the full set of Navier–Stokes equations, both in terms of design performance and computational cost. The computational time is shown to be decreased to around in terms of core-hours, allowing for the possibility of optimizing designs during the workday on a small computational cluster and overnight on a high-end desktop. However, due to the use of a simplified model, the performance of the final designs are evaluated using the full Navier–Stokes equations. This ensures verification of performance, as well as systematic comparison with reference results.",
keywords = "Topology optimization, Conjugate heat transfer, Natural convection, High performance computing, Simplified model",
author = "Nicolo Pollini and Ole Sigmund and Andreasen, {Casper Schousboe} and Joe Alexandersen",
year = "2019",
doi = "10.1016/j.advengsoft.2019.102736",
language = "English",
volume = "140",
journal = "Advances in Engineering Software",
issn = "0965-9978",
publisher = "Pergamon Press",

}

A “poor man’s” approach for high-resolution three-dimensional topology design for natural convection problems. / Pollini, Nicolo; Sigmund, Ole; Andreasen, Casper Schousboe; Alexandersen, Joe.

In: Advances in Engineering Software, Vol. 140, 102736, 2019.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - A “poor man’s” approach for high-resolution three-dimensional topology design for natural convection problems

AU - Pollini, Nicolo

AU - Sigmund, Ole

AU - Andreasen, Casper Schousboe

AU - Alexandersen, Joe

PY - 2019

Y1 - 2019

N2 - This paper treats topology optimization of natural convection problems. A simplified model is suggested to describe the flow of a steady-state incompressible fluid, similar to Darcy’s law for porous media. By neglecting inertia and viscous boundary layers, the flow model is significantly simplified. The fluid flow is coupled to the thermal convection-diffusion equation through the Boussinesq approximation. The coupled non-linear system of equations is discretized with stabilized finite elements and solved in a parallel framework that allows for the optimization of high-resolution three-dimensional problems. A density-based topology optimization approach is used, where the permeability and conductivity of the distributed material is interpolated. Due to the simplified model, the proposed methodology significantly reduces the computational effort required in the optimization. At the same time, it is notably more accurate than even simpler models that rely on Newton’s law of cooling. The methodology discussed herein is applied to the optimization-based design of three-dimensional heat sinks. The final designs are compared with previous work obtained from solving the full set of Navier–Stokes equations, both in terms of design performance and computational cost. The computational time is shown to be decreased to around in terms of core-hours, allowing for the possibility of optimizing designs during the workday on a small computational cluster and overnight on a high-end desktop. However, due to the use of a simplified model, the performance of the final designs are evaluated using the full Navier–Stokes equations. This ensures verification of performance, as well as systematic comparison with reference results.

AB - This paper treats topology optimization of natural convection problems. A simplified model is suggested to describe the flow of a steady-state incompressible fluid, similar to Darcy’s law for porous media. By neglecting inertia and viscous boundary layers, the flow model is significantly simplified. The fluid flow is coupled to the thermal convection-diffusion equation through the Boussinesq approximation. The coupled non-linear system of equations is discretized with stabilized finite elements and solved in a parallel framework that allows for the optimization of high-resolution three-dimensional problems. A density-based topology optimization approach is used, where the permeability and conductivity of the distributed material is interpolated. Due to the simplified model, the proposed methodology significantly reduces the computational effort required in the optimization. At the same time, it is notably more accurate than even simpler models that rely on Newton’s law of cooling. The methodology discussed herein is applied to the optimization-based design of three-dimensional heat sinks. The final designs are compared with previous work obtained from solving the full set of Navier–Stokes equations, both in terms of design performance and computational cost. The computational time is shown to be decreased to around in terms of core-hours, allowing for the possibility of optimizing designs during the workday on a small computational cluster and overnight on a high-end desktop. However, due to the use of a simplified model, the performance of the final designs are evaluated using the full Navier–Stokes equations. This ensures verification of performance, as well as systematic comparison with reference results.

KW - Topology optimization

KW - Conjugate heat transfer

KW - Natural convection

KW - High performance computing

KW - Simplified model

U2 - 10.1016/j.advengsoft.2019.102736

DO - 10.1016/j.advengsoft.2019.102736

M3 - Journal article

VL - 140

JO - Advances in Engineering Software

JF - Advances in Engineering Software

SN - 0965-9978

M1 - 102736

ER -