AI-driven feedforward neural networks for thermal performance prediction of ternary hybrid nanofluid flow in co-axial cylindrical systems

This work examines the behaviour of ternary nanofluid flow between two co-axial cylinders (CAC) with the impact of a magnetic field using a combination of Levenberg–Marquardt artificial neural network (LM-ANN). The governing flow model is resolved computationally using MATLAB’s bvp4c solver and further analyzed through the LM-ANN. For training, validation, and testing of the ANN model, a step size of 0.0075 is employed, generating a dataset of 100 points over the radial domain 0 . 3 7 5 ≤ 𝑟 ≤ 1 . 5 . Of these data, 70% are given for training, whilst the remaining 30% are equally divided between validating and testing. The obtained outcomes verify that as 𝑀 is augmented, it significantly reduces the flow velocity through Lorentz force action, nevertheless, the temperature field becomes increased due to more intensified nanoparticle interactions and higher thermal diffusion, with a rise in 𝐵 𝑟 . Also, the pressure distribution after undergoing a change due to the introduction of the 𝑀 is seen to show a decrease on the one hand, however, on the other hand, as the Reynolds number gets higher, the pressure increases due to enhanced inertial effects. The LM-ANN forecasts display an excellent agreement with the numerical counterparts with the errors in terms of absolute value being within the range of from 1 0 − 6 to 1 0 − 4 , thus verifying the model's accuracy and trustworthiness. The 10-neuron architecture yields a more effective generalization and a more consistent predictive accuracy than the one with 20 neurons. Moreover, the use of nanoparticles contributes to the improvement of mass and heat transmission, whereas the imposition of the magnetic field generates an opposite effect to the flow resistance but at the same time, also results in its intensification.

Authors

Institutions

Publication Details

Journal
Next Nanotechnology
Published
2026-11-18
DOI
https://doi.org/10.1016/j.nxnano.2026.100938
Primary Topic
Nanofluid Flow and Heat Transfer
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

AI-driven feedforward neural networks for thermal performance prediction of ternary hybrid nanofluid flow in co-axial cylindrical systems

Saima Zainab, Basim M. Makhdoum, Wasim Jamshed, Kiran Batool et al.
7 links
Next Nanotechnology
Nanofluid Flow and Heat Transfer
article

AI-driven feedforward neural networks for thermal performance prediction of ternary hybrid nanofluid flow in co-axial cylindrical systems

Saima Zainab, Basim M. Makhdoum, Wasim Jamshed, Kiran Batool, Sadia Shakir, Mohamed R. Eid
article en

Abstract

This work examines the behaviour of ternary nanofluid flow between two co-axial cylinders (CAC) with the impact of a magnetic field using a combination of Levenberg–Marquardt artificial neural network (LM-ANN). The governing flow model is resolved computationally using MATLAB’s bvp4c solver and further analyzed through the LM-ANN. For training, validation, and testing of the ANN model, a step size of 0.0075 is employed, generating a dataset of 100 points over the radial domain 0 . 3 7 5 ≤ 𝑟 ≤ 1 . 5 . Of these data, 70% are given for training, whilst the remaining 30% are equally divided between validating and testing. The obtained outcomes verify that as 𝑀 is augmented, it significantly reduces the flow velocity through Lorentz force action, nevertheless, the temperature field becomes increased due to more intensified nanoparticle interactions and higher thermal diffusion, with a rise in 𝐵 𝑟 . Also, the pressure distribution after undergoing a change due to the introduction of the 𝑀 is seen to show a decrease on the one hand, however, on the other hand, as the Reynolds number gets higher, the pressure increases due to enhanced inertial effects. The LM-ANN forecasts display an excellent agreement with the numerical counterparts with the errors in terms of absolute value being within the range of from 1 0 − 6 to 1 0 − 4 , thus verifying the model's accuracy and trustworthiness. The 10-neuron architecture yields a more effective generalization and a more consistent predictive accuracy than the one with 20 neurons. Moreover, the use of nanoparticles contributes to the improvement of mass and heat transmission, whereas the imposition of the magnetic field generates an opposite effect to the flow resistance but at the same time, also results in its intensification.

Next NanotechnologyVol. 10
Umm al-Qura University (SA), Biruni University (TR), The Women University Multan (PK), New Valley University (EG)
Nanofluid Flow and Heat Transfer
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon