A predictive surrogate model for heat transfer of an impinging jet on a concave surface

Sajad Salavatidezfouli; Saeed Rakhsha; Armin Sheidani; Giovanni Stabile; Gianluigi Rozza

doi:10.1016/j.ijheatmasstransfer.2025.127248

A predictive surrogate model for heat transfer of an impinging jet on a concave surface

Sajad Salavatidezfouli, Saeed Rakhsha, Armin Sheidani, Giovanni Stabile, Gianluigi Rozza^*

^*Corresponding author for this work

Department of Electrical and Computer Engineering - Signal Processing and Machine learning

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

3 Citations (Scopus)

Abstract

This paper aims to comprehensively investigate the efficacy of model order reduction and deep learning techniques in predicting heat transfer of pulsatile impinging jets on a concave surface. We introduce two predictive approaches: one employing a Fast Fourier Transform-augmented Artificial Neural Network for predicting the area-averaged Nusselt number under constant-frequency jet scenarios, and another comparing the performance of LSTM and Transformer neural networks for random-frequency jets. Results indicate that the Transformer significantly outperforms the LSTM, achieving higher accuracy and robustness in predicting Nusselt numbers — covering up to 50% of the cycle with precision, whereas the LSTM covers only 20% with greater error margins. Additionally, the integration of Proper Orthogonal Decomposition with Transformer networks yields a novel strategy for predicting local Nusselt number distributions. This method significantly reduces computational complexity while maintaining high accuracy, with a maximum prediction error of 5%. These findings demonstrate the efficacy of advanced deep learning techniques for temporal prediction of the Nusselt number on complex surfaces, suggesting further applicability in broader fluid dynamics and heat transfer analyses.

Original language	English
Article number	127248
Journal	International Journal of Heat and Mass Transfer
Volume	251
ISSN	0017-9310
DOIs	https://doi.org/10.1016/j.ijheatmasstransfer.2025.127248
Publication status	Published - Nov 2025

Keywords

Concave surface
Deep learning
Heat transfer
Model order reduction
Predictive surrogate model

Access to Document

10.1016/j.ijheatmasstransfer.2025.127248

Library access?

Check availability with the Royal Danish Library

Cite this

@article{79d1e11e01d646c594ca075c809900e4,

title = "A predictive surrogate model for heat transfer of an impinging jet on a concave surface",

abstract = "This paper aims to comprehensively investigate the efficacy of model order reduction and deep learning techniques in predicting heat transfer of pulsatile impinging jets on a concave surface. We introduce two predictive approaches: one employing a Fast Fourier Transform-augmented Artificial Neural Network for predicting the area-averaged Nusselt number under constant-frequency jet scenarios, and another comparing the performance of LSTM and Transformer neural networks for random-frequency jets. Results indicate that the Transformer significantly outperforms the LSTM, achieving higher accuracy and robustness in predicting Nusselt numbers — covering up to 50\% of the cycle with precision, whereas the LSTM covers only 20\% with greater error margins. Additionally, the integration of Proper Orthogonal Decomposition with Transformer networks yields a novel strategy for predicting local Nusselt number distributions. This method significantly reduces computational complexity while maintaining high accuracy, with a maximum prediction error of 5\%. These findings demonstrate the efficacy of advanced deep learning techniques for temporal prediction of the Nusselt number on complex surfaces, suggesting further applicability in broader fluid dynamics and heat transfer analyses.",

keywords = "Concave surface, Deep learning, Heat transfer, Model order reduction, Predictive surrogate model",

author = "Sajad Salavatidezfouli and Saeed Rakhsha and Armin Sheidani and Giovanni Stabile and Gianluigi Rozza",

note = "Publisher Copyright: {\textcopyright} 2025 Elsevier Ltd",

year = "2025",

month = nov,

doi = "10.1016/j.ijheatmasstransfer.2025.127248",

language = "English",

volume = "251",

journal = "International Journal of Heat and Mass Transfer",

issn = "0017-9310",

publisher = "Elsevier Ltd",

}

A predictive surrogate model for heat transfer of an impinging jet on a concave surface. / Salavatidezfouli, Sajad; Rakhsha, Saeed; Sheidani, Armin et al.
In: International Journal of Heat and Mass Transfer, Vol. 251, 127248, 11.2025.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

TY - JOUR

T1 - A predictive surrogate model for heat transfer of an impinging jet on a concave surface

AU - Salavatidezfouli, Sajad

AU - Rakhsha, Saeed

AU - Sheidani, Armin

AU - Stabile, Giovanni

AU - Rozza, Gianluigi

PY - 2025/11

Y1 - 2025/11

N2 - This paper aims to comprehensively investigate the efficacy of model order reduction and deep learning techniques in predicting heat transfer of pulsatile impinging jets on a concave surface. We introduce two predictive approaches: one employing a Fast Fourier Transform-augmented Artificial Neural Network for predicting the area-averaged Nusselt number under constant-frequency jet scenarios, and another comparing the performance of LSTM and Transformer neural networks for random-frequency jets. Results indicate that the Transformer significantly outperforms the LSTM, achieving higher accuracy and robustness in predicting Nusselt numbers — covering up to 50% of the cycle with precision, whereas the LSTM covers only 20% with greater error margins. Additionally, the integration of Proper Orthogonal Decomposition with Transformer networks yields a novel strategy for predicting local Nusselt number distributions. This method significantly reduces computational complexity while maintaining high accuracy, with a maximum prediction error of 5%. These findings demonstrate the efficacy of advanced deep learning techniques for temporal prediction of the Nusselt number on complex surfaces, suggesting further applicability in broader fluid dynamics and heat transfer analyses.

AB - This paper aims to comprehensively investigate the efficacy of model order reduction and deep learning techniques in predicting heat transfer of pulsatile impinging jets on a concave surface. We introduce two predictive approaches: one employing a Fast Fourier Transform-augmented Artificial Neural Network for predicting the area-averaged Nusselt number under constant-frequency jet scenarios, and another comparing the performance of LSTM and Transformer neural networks for random-frequency jets. Results indicate that the Transformer significantly outperforms the LSTM, achieving higher accuracy and robustness in predicting Nusselt numbers — covering up to 50% of the cycle with precision, whereas the LSTM covers only 20% with greater error margins. Additionally, the integration of Proper Orthogonal Decomposition with Transformer networks yields a novel strategy for predicting local Nusselt number distributions. This method significantly reduces computational complexity while maintaining high accuracy, with a maximum prediction error of 5%. These findings demonstrate the efficacy of advanced deep learning techniques for temporal prediction of the Nusselt number on complex surfaces, suggesting further applicability in broader fluid dynamics and heat transfer analyses.

KW - Concave surface

KW - Deep learning

KW - Heat transfer

KW - Model order reduction

KW - Predictive surrogate model

UR - https://www.scopus.com/pages/publications/105007059601

U2 - 10.1016/j.ijheatmasstransfer.2025.127248

DO - 10.1016/j.ijheatmasstransfer.2025.127248

M3 - Journal article

AN - SCOPUS:105007059601

SN - 0017-9310

VL - 251

JO - International Journal of Heat and Mass Transfer

JF - International Journal of Heat and Mass Transfer

M1 - 127248

ER -

A predictive surrogate model for heat transfer of an impinging jet on a concave surface

Abstract

Keywords

Access to Document

Library access?

Other files and links

Fingerprint

Cite this