Heat Conduction 3D

Heat Conduction 3D#

../../_images/heatconduction3d.png

Version

0

Design space

Box(0.0, 1.0, (51, 51, 51), float64)

Objectives

c: ↓

Conditions

volume: 0.3

area: 0.5

Dataset

IDEALLab/heat_conduction_3d_v0

Container

quay.io/dolfinadjoint/pyadjoint:master

Import

from engibench.problems.heatconduction3d.v0 import HeatConduction3D

HeatConduction 3D topology optimization problem.

Problem Description#

This problem simulates the performance of a Topology optimisation of heat conduction problems governed by the Poisson equation (https://github.com/dolfin-adjoint/pyadjoint/blob/master/examples/poisson-topology/poisson-topology.py)

Design space#

The design space is represented by a 3D numpy array which indicates the resolution.

Objectives#

The objective is defined and indexed as follows:

  1. c: Thermal compliance coefficient to minimize.

Conditions#

The conditions are defined by the following parameters:

  • volume: the volume limits on the material distributions

  • area: The area of the adiabatic region on the bottom side of the design domain.

Simulator#

The simulator is a docker container with the dolfin-adjoint software that computes the thermal compliance of the design. We convert use intermediary files to convert from and to the simulator that is run from a Docker image.

Dataset#

The dataset has been generated the dolfin-adjoint software. It is hosted on the Hugging Face Datasets Hub.

v0#

Fields#

The dataset only contains conditions and optimal designs (no objective).

Creation Method#

The creation method for the dataset is specified in the reference paper.

References#

If you use this problem in your research, please cite the following paper: Habibi, Milad, Shai Bernard, Jun Wang, and Mark Fuge, “Mean squared error may lead you astray when optimizing your inverse design methods” in JMD 2025. doi: https://doi.org/10.1115/1.4066102

Lead#

Milad Habibi @MIladHB

Contents