EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Optimisation of Heat Transfer in Perforated Fins under Forced Flow adopting a Radial Basis Function Neural Network

W. H. Al Doori, M. A. Kumbhalkar, S. A. Choudhari, S. B. Khaple, T. S. Sargar, S. V. Prayagi

AFRICAN JOURNAL OF APPLIED RESEARCH · 04.07.2026 · 12 (5) · 187-205

DOI: 10.26437/kry5xj80 ↗

Kurz gefasst

Fachartikel in der Zeitschrift AFRICAN JOURNAL OF APPLIED RESEARCH, online erschienen am 04.07.2026. Verfasst von W. H. Al Doori und 5 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

This study aims to develop an integrated computational framework that combines Computational Fluid Dynamics (CFD), Radial Basis Function Neural Networks (RBFNN), and multi-objective optimisation to predict and optimise the thermo-hydraulic performance of perforated fins under forced-convection conditions. Design/Methodology/Approach: The Radial Basis Function Neural Network (RBFNN) surrogate model was trained using high-fidelity Computational Fluid Dynamics (CFD) simulation data covering multiple perforation geometries and operating conditions. The developed surrogate achieved an average Nusselt-number prediction accuracy exceeding R² > 0.98 with relative prediction errors below 0.02%. To determine optimal fin configurations, the Non-Dominated Sorting Genetic Algorithm II (NSGA-II), a widely adopted multi-objective evolutionary optimisation algorithm, was employed to simultaneously maximise heat-transfer performance and minimise aerodynamic pressure drop under forced-convection conditions.

Findings

The optimal configuration was 15 rectangular perforations at an airflow of 1.8 m/s, with a Nu of 66.2 and a pressure drop of 12.3 Pa. Another design with 6 circular perforations at 0.7 m/s had a Nu of 38.1 and a pressure drop of 3.9 Pa. Sensitivity analysis showed that air velocity and perforation diameter contributed nearly 60% of the performance variation. Optimised designs enhanced heat transfer performance by up to 16% while reducing fin mass by 96% compared to non-perforated fins. Research Limitation: The optimisation process was restricted to selected geometric variables, namely perforation size, shape, and count, without considering manufacturing constraints or turbulent high-Reynolds-number flow conditions. Practical Implication: The proposed RBFNN-based optimisation framework significantly reduces computational cost associated with repeated CFD simulations while providing accurate thermo-fluid predictions. The methodology can support rapid engineering design of lightweight, high-efficiency cooling systems for aerospace, automotive, electronics, and compact heat exchanger applications. Social Implication: Enhanced thermal management systems directly improve energy efficiency and reduce operational energy consumption in electronic devices, transportation systems, and industrial equipment. Optimised lightweight perforated fins may therefore support more sustainable cooling technologies with lower material usage and reduced environmental impact. Originality / Value: This study presents a novel integrated framework that combines CFD-generated thermal-fluid datasets, Radial Basis Function Neural Networks (RBFNN), uncertainty quantification, and NSGA-II multi-objective optimisation for the design of perforated fins under forced convection.

Zitieren

APA

Doori, W. H. A., Kumbhalkar, M. A., Choudhari, S. A., Khaple, S. B., Sargar, T. S., & Prayagi, S. V. (2026). Optimisation of Heat Transfer in Perforated Fins under Forced Flow adopting a Radial Basis Function Neural Network. AFRICAN JOURNAL OF APPLIED RESEARCH, 12(5), 187-205. https://doi.org/10.26437/kry5xj80

Vancouver

Doori WHA, Kumbhalkar MA, Choudhari SA, Khaple SB, Sargar TS, Prayagi SV. Optimisation of Heat Transfer in Perforated Fins under Forced Flow adopting a Radial Basis Function Neural Network. AFRICAN JOURNAL OF APPLIED RESEARCH. 2026;12(5):187-205. doi:10.26437/kry5xj80

BibTeX
@article{Doori2026,
  author = {Doori, W. H. Al and Kumbhalkar, M. A. and Choudhari, S. A. and Khaple, S. B. and Sargar, T. S. and Prayagi, S. V.},
  title = {Optimisation of Heat Transfer in Perforated Fins under Forced Flow adopting a Radial Basis Function Neural Network},
  journal = {AFRICAN JOURNAL OF APPLIED RESEARCH},
  year = {2026},
  volume = {12},
  number = {5},
  pages = {187--205},
  doi = {10.26437/kry5xj80}
}
Kostenlos starten →