Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks

Taiba Kouser; Dilek Funda Kurtulus; Abdulrahman Aliyu; Imil Hamda Imran; Luai M. Alhems; Srikanth Goli

doi:10.1007/978-3-032-18834-2_22

Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks

Taiba Kouser^*
, Dilek Funda Kurtulus
, Abdulrahman Aliyu
, Imil Hamda Imran
, Luai M. Alhems
, Srikanth Goli

^*Corresponding author for this work

Applied Research Center for Metrology, Standards and Testing

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

Abstract

In this study, a two-layer artificial neural network (ANN) is used to predict the aerodynamic lift and drag coefficients for NACA0005. The first layer has nine neurons, while the number of neurons in the second layer is varied from 1 to 100 to find the best performance. Data from 28 angles of attack ranges 9^° ≤ α ≤ 11^°, and the Reynolds number between 1000 and 5000 are utilized. The ANN achieved R² > 0.99 for lift coefficient (C_l) and R² > 0.96 for drag coefficient (C_d), showing high prediction accuracy. Reynolds number of 1750 and 3000 are evaluated as validation and testing to assess the accuracy of ANN. The overall ANN predicts the computational fluid dynamics results, with errors below 1% for C_l and C_d in the testing phase.

Original language	English
Title of host publication	International Conference on Intelligent Systems and Computational Strategies in Engineering and Technology
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	327-334
Number of pages	8
DOIs	https://doi.org/10.1007/978-3-032-18834-2_22
State	Published - 2026

Publication series

Name	Signals and Communication Technology
Volume	Part F2435
ISSN (Print)	1860-4862
ISSN (Electronic)	1860-4870

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Keywords

Aerodynamic coefficients
Angle of attack
Artificial neural network (ANN)
Computational fluid dynamics (CFD)
NACA 0005
Reynolds number

ASJC Scopus subject areas

Control and Systems Engineering
Signal Processing
Computer Networks and Communications
Electrical and Electronic Engineering

Access to Document

10.1007/978-3-032-18834-2_22

Cite this

Kouser, T., Kurtulus, D. F., Aliyu, A., Imran, I. H., Alhems, L. M., & Goli, S. (2026). Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks. In International Conference on Intelligent Systems and Computational Strategies in Engineering and Technology (pp. 327-334). (Signals and Communication Technology; Vol. Part F2435). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-032-18834-2_22

@inbook{91fdc796b6bd4c48b7a5b35c3ef5e5b5,

title = "Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks",

abstract = "In this study, a two-layer artificial neural network (ANN) is used to predict the aerodynamic lift and drag coefficients for NACA0005. The first layer has nine neurons, while the number of neurons in the second layer is varied from 1 to 100 to find the best performance. Data from 28 angles of attack ranges 9° ≤ α ≤ 11°, and the Reynolds number between 1000 and 5000 are utilized. The ANN achieved R2 > 0.99 for lift coefficient (Cl) and R2 > 0.96 for drag coefficient (Cd), showing high prediction accuracy. Reynolds number of 1750 and 3000 are evaluated as validation and testing to assess the accuracy of ANN. The overall ANN predicts the computational fluid dynamics results, with errors below 1\% for Cl and Cd in the testing phase.",

keywords = "Aerodynamic coefficients, Angle of attack, Artificial neural network (ANN), Computational fluid dynamics (CFD), NACA 0005, Reynolds number",

author = "Taiba Kouser and Kurtulus, \{Dilek Funda\} and Abdulrahman Aliyu and Imran, \{Imil Hamda\} and Alhems, \{Luai M.\} and Srikanth Goli",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.",

year = "2026",

doi = "10.1007/978-3-032-18834-2\_22",

language = "English",

series = "Signals and Communication Technology",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "327--334",

booktitle = "International Conference on Intelligent Systems and Computational Strategies in Engineering and Technology",

address = "Germany",

}

Kouser, T, Kurtulus, DF, Aliyu, A, Imran, IH, Alhems, LM & Goli, S 2026, Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks. in International Conference on Intelligent Systems and Computational Strategies in Engineering and Technology. Signals and Communication Technology, vol. Part F2435, Springer Science and Business Media Deutschland GmbH, pp. 327-334. https://doi.org/10.1007/978-3-032-18834-2_22

Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks. / Kouser, Taiba; Kurtulus, Dilek Funda; Aliyu, Abdulrahman et al.
International Conference on Intelligent Systems and Computational Strategies in Engineering and Technology. Springer Science and Business Media Deutschland GmbH, 2026. p. 327-334 (Signals and Communication Technology; Vol. Part F2435).

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

TY - CHAP

T1 - Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks

AU - Kouser, Taiba

AU - Kurtulus, Dilek Funda

AU - Aliyu, Abdulrahman

AU - Imran, Imil Hamda

AU - Alhems, Luai M.

AU - Goli, Srikanth

PY - 2026

Y1 - 2026

N2 - In this study, a two-layer artificial neural network (ANN) is used to predict the aerodynamic lift and drag coefficients for NACA0005. The first layer has nine neurons, while the number of neurons in the second layer is varied from 1 to 100 to find the best performance. Data from 28 angles of attack ranges 9° ≤ α ≤ 11°, and the Reynolds number between 1000 and 5000 are utilized. The ANN achieved R2 > 0.99 for lift coefficient (Cl) and R2 > 0.96 for drag coefficient (Cd), showing high prediction accuracy. Reynolds number of 1750 and 3000 are evaluated as validation and testing to assess the accuracy of ANN. The overall ANN predicts the computational fluid dynamics results, with errors below 1% for Cl and Cd in the testing phase.

AB - In this study, a two-layer artificial neural network (ANN) is used to predict the aerodynamic lift and drag coefficients for NACA0005. The first layer has nine neurons, while the number of neurons in the second layer is varied from 1 to 100 to find the best performance. Data from 28 angles of attack ranges 9° ≤ α ≤ 11°, and the Reynolds number between 1000 and 5000 are utilized. The ANN achieved R2 > 0.99 for lift coefficient (Cl) and R2 > 0.96 for drag coefficient (Cd), showing high prediction accuracy. Reynolds number of 1750 and 3000 are evaluated as validation and testing to assess the accuracy of ANN. The overall ANN predicts the computational fluid dynamics results, with errors below 1% for Cl and Cd in the testing phase.

KW - Aerodynamic coefficients

KW - Angle of attack

KW - Artificial neural network (ANN)

KW - Computational fluid dynamics (CFD)

KW - NACA 0005

KW - Reynolds number

UR - https://www.scopus.com/pages/publications/105049363058

U2 - 10.1007/978-3-032-18834-2_22

DO - 10.1007/978-3-032-18834-2_22

M3 - Chapter

AN - SCOPUS:105049363058

T3 - Signals and Communication Technology

SP - 327

EP - 334

BT - International Conference on Intelligent Systems and Computational Strategies in Engineering and Technology

PB - Springer Science and Business Media Deutschland GmbH

ER -

Kouser T, Kurtulus DF, Aliyu A, Imran IH, Alhems LM, Goli S. Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks. In International Conference on Intelligent Systems and Computational Strategies in Engineering and Technology. Springer Science and Business Media Deutschland GmbH. 2026. p. 327-334. (Signals and Communication Technology). doi: 10.1007/978-3-032-18834-2_22

Data-Driven Modeling of NACA 0005 Airfoil Aerodynamics Using Neural Networks

Abstract

Publication series

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this