Shape and Layout Optimization of Bare-Tube Heat Exchangers Using the Adjoint Method and CAD-based Parametrization

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

1 Downloads (Pure)

Abstract

This study presents a Computational Fluid Dynamics (CFD)-based optimization framework to enhance the performance of bare-tube heat exchangers. The framework comprises a CFD solver, an adjoint solver, and a CAD-based parametrization tool. This framework simultaneously optimizes the tube shape and layout to obtain an optimum heat exchanger configuration with a higher heat transfer rate and lower pressure drop. Firstly, a parametric study of the longitudinal pitch is performed that shows that increasing the longitudinal pitch results in an increase in pressure drop and heat transfer rate. Furthermore, the proposed framework was applied to optimize an in-line elliptical tube configuration. The optimum geometry showed a performance improvement of 29% as compared to the baseline geometry while satisfying the constraint on the heat transfer rate.

Original languageEnglish
Title of host publicationProceedings of the 9th World Congress on Momentum, Heat and Mass Transfer, MHMT 2024
EditorsLixin Cheng, Tassos G. Karayiannis, Sohel Murshed
PublisherAvestia Publishing
Number of pages8
ISBN (Print)9781990800344
DOIs
Publication statusPublished - 2024
Event9th World Congress on Momentum, Heat and Mass Transfer, MHMT 2024 - London, United Kingdom
Duration: 11 Apr 202413 Apr 2024

Publication series

NameProceedings of the World Congress on Momentum, Heat and Mass Transfer
ISSN (Electronic)2371-5316

Conference

Conference9th World Congress on Momentum, Heat and Mass Transfer, MHMT 2024
Country/TerritoryUnited Kingdom
CityLondon
Period11/04/2413/04/24

Keywords

  • CFD
  • Design and Simulation
  • Drag Reduction
  • Shape Optimization
  • Turbulent Flow and Heat Transfer

Fingerprint

Dive into the research topics of 'Shape and Layout Optimization of Bare-Tube Heat Exchangers Using the Adjoint Method and CAD-based Parametrization'. Together they form a unique fingerprint.

Cite this