Influence of Chemical and Radiation on an Unsteady MHD Oscillatory

Flow using Artificial Neural Network (ANN)

R. KAVITHA1, M. MAHENDRAN2

SRM Institute of Science and Technology,

Kattankulathur Campus, Chengalpattu,

Abstract: – This paper delves into the intricate interplay between chemical and thermal radiation in the context

of an unstable magnetohydrodynamic(MHD) oscillatory flow through a porous medium. The fluid under

investigation is presumed to be incompressible, electrically conductive, and radiating with the additional

influence of a homogeneous magnetic field applied perpendicular to the channel’s plane. Analytical closed-

form solutions are derived for the momentum, energy, and concentration equations providing a comprehensive

understanding of the system’s behavior. The investigation systematically explores the impact of various flow

factors, presenting their effects through graphical representations. The governing partial differential equations

(PDE) of the boundary layer are transformed into a set of coupled nonlinear ordinary differential equations

(ODE) using a closed-form method. Subsequently, an artificial neural network (ANN) is applied to these

approach leverages the power of artificial intelligence to enhance our understanding of complex fluid

magnetohydrodynamics and porous media flows.

Key-Words: - MHD, Oscillatory flow, Porous medium, Chemical Reaction, Thermal Radiation, HMT –

Received: March 6, 2023. Revised: December 19, 2023. Accepted: February 21, 2024. Published: April 2, 2024.

1 Introduction

employed in a wide assortment of businesses. To

begin, businesses use galvanized products to create

corrosion-resistant steel. In industry, electrolysis is

used to generate pure quality commodities and

minerals. Chemical reactions are used in industry to

recover minerals from ores. In [1], author

contributes to the understanding of MHD effects on

oscillatory flow around a semi-infinite plate. In [2],

the impact of MHD unsteady oscillatory flow

through a porous medium was explored. In [3], the

effects was discussed in [6]. The heat transfer

capabilities of hybrid and traditional nanofluids were

compared in [7]. In [8], the viscous dissipation of an

unsteady third-grade fluid was studied. In [9], the

author explored the dufour effect and emission

discharge through porous skin. Because of its

significance in soil mechanics, the study of

oscillatory flow in a porous channel has received a

lot of interest recently. In [11], researchers

investigated the effects of chemical processes on

oscillatory MHD flow in an asymmetric channel. In

[12], authors investigated MHD channel flow

through a porous medium – numerical approach.

This paper deals with radiative heat and convective

cooling.

Carlo Ray Tracing (MCRT) method. In [14]

explored heat enhancement using the ANN

approach. In [15] delved into the thermos-

bioconvection model using computational fluid

dynamics (CFD) and artificial intelligence. In [16]

employed a CFD machine learning approach to

study modeling strategies, reducing computational

costs and evaluating the machine learning model. In

[17] developed artificial intelligence machine

learning approaches for simulating combustion

and enhancement of momentum equation, energy

equation, and concentration equation in the realm of

fluid dynamics. Accurate modeling and prediction of

fluid dynamics are vital in a variety of engineering

applications, including aeronautical and

environmental sciences.

This research will investigate how a chemical

process and heat radiation influence an unstable

oscillatory magnetohydrodynamic (MHD) flow in

porous media. The governing equations for fluid

magetohydrodynamic (MHD) properties, electrical

conductivity, and chemical reactivity through a

porous media positioned between two infinite

vertical porous plates at a distance ‘d’. The channel

is oriented with the X-axis vertically and the Y-axis

perpendicular to the plane of the plates. Several

assumptions have been made:

considered constant.

considered when the permeability of the

porous media is constant.

the typical Boussinesq approximation, are

provided as follows:

The flow's governing equations are:

Momentum Equation

 

 

parameters such as Peclet number, Magnetic

number, Schmidt number, chemical process

parameter and so on.

Figure 1 depicts the fluid temperature for

various levels of thermal radiation. It is obvious that

an increase in heat radiation raises the temperature.

This is because increasing the radiation parameter

causes heat energy to be released into the fluid. This

is consistent with the radiation parameter’s

fundamental physical behavior. As the radiation

improves. Thermal radiation values increase the heat

of the working fluid more, raising the temperature

and thermal boundary layer thickness. The thermal

radiation parameter does not affect the concentration

distribution in flow.

Figure 2 depicts the effect of the Peclet number

on temperature. The Peclet number is defined as the

ratio of advective transport to diffusive transport. A

low Peclet number (Pe < 1) indicated that diffusion

is dominant over advection. In such cases, the

increasing Peclet number.

Figure 3 displays the fluid velocity for various

magnetic parameter values. Increasing magnetic

parameters reduces the fluid velocity. This is due to

the increase in resistive type force known as Lorentz

force caused by the transverse magnetic field on an

accelerated, the fluid concentration decreases. This is

because boosting the chemical process speeds up the

reactant rate process on the flow, lowering the

concentration of the reacting species. Interfacial MT

is accelerated by the chemical process. The chemical

reaction decreases local concentration while

increasing concentration gradient and flow. The

chemical reaction parameter has little to no effect on

the flow's temperature profile. The effect of Schmidt

number on concentration profiles is depicted in

offers exact solutions. MATLAB is used to visually

analyze velocity, temperature, and concentration

profiles for a variety of flow parameters.

The recent analysis found that velocity rises with

decreasing magnetic field. Temperature profiles

fluctuate as radiation parameter N changes. It has

also been discovered that temperature profiles grow

in proportion to the heat source. Concentration

patterns drop as chemical reaction parameters grow,

but reverse as Schmidt number Sc increases. It is also

Fig. 1: Impact of thermal radiation on temperature

Fig. 2: Impact of Peclet number on temperature

6 Modelling - Artificial Neural

Artificial neural networks, inspired by biological

networks of neurons, are designed to perform

specific tasks such as grouping, classification, and

pattern recognition. The activation of a neuron is

determined by weights, and the output of an artificial

neuron is computed using another function.

ANN have found widespread applications across

various domains due to their ability to model

effectiveness of ANN in solving complex problems

and improving decision-making processes across

diverse industries.

This study employs a multi-layer feed-forward

neural network with the Back Propagation training

algorithm. The multi-layer perception consists of at

least three layers: an input layer, an output layer and

one or more hidden layers. Weights are adjusted

through the Back Propagation training method to

minimize the difference between expected and

while 15% was allocated for validation, and another

15% was reserved for evaluating the model’s output.

Fig. 6: Graphical Representation of skin friction coefficients

The artificial neurons employed a sigmoid

function as their activation function, with training

conducted over a fixed number of epochs. For the

prediction of skin friction, Nusselt number and

Sherwood number three separate ANN models

were trained, tested, and validated using a total of

80 numerical results. In each case, the initial 50

data sets were allocated for training, the

subsequent 15 for validation, and the remaining

sets for testing the model’s outcomes. The

performance of the suggested ANN models on the

training, validation, and test sets for skin friction,

validation, and test sets focusing on skin friction

coefficient, Nusselt number, and Sherwood

number. Remarkably, the ANN models exhibited a

high degree of accuracy, successfully capturing the

intricate relationships between input and output

variables. The results obtained from the ANN

models closely align with the numerically derived

values.

7 Conclusion

This research investigates the combined influence

of chemical and thermal radiation on oscillatory

MHD flow in a porous media containing HMT.

Converting the controlling PDE to an ODE offers

exact solutions. MATLAB is used to visually

analyze velocity, temperature, and concentration

profiles for a variety of flow parameters.

The recent analysis found that,

In addition to the above conclusion, we found

the following agreement using the ANN Model.

The current study successfully employs the

ANN technique to simulate HMT in the MHD

chemical and thermal radiation flow through a

stretching/shrinking sheet, according to the results

prediction of skin friction, Nusselt number, and

Sherwood number fits the standard numerical data

well. The ANN model is a valuable tool and a

potential alternative to traditional time-consuming

numerical approaches since it offers quick, precise,

8 Future Work

Development of Hybrid Models: Explore the

development of hybrid models that combine

traditional mathematical models for fluid

dynamics, MHD, chemical reactions, and radiation

with ANN based models. This could improve the

accuracy and efficiency of predictions.

1977.

[2] M.Chitra and M.Suhasini, “Effect of

unsteady oscillatory MHD flow through a

porous medium in porous vertical channel

with chemical reaction and concentration”,

Journal of Physics Conference series, 1000

012039, 2018.

[3] R.Kavitha, L.Jeyanthi, S.M.Chithra and

Sherlin Nisha, “Numerical investigation of

MHD hybrid Cu-AL2O3/Water nanofluid

absorption”, Journal of Survey in Fisheries

Sciences, 10(2S), 2544-2551, 2023.

[5] A.M. Okedoye and S.O.Salawu, “Unsteady

oscillatory MHD boundary layer flow past a

moving plate with mass transfer and binary

chemical reaction”, SN Applied Science, 1,

1586, 2019.

[6] Raghunath Kodi and Obulesu Mopuri,

“Unsteady MHD oscillatory casson fluid

flow past an inclined vertical porous plate

Al2O3/water nanofluid flow over a

J. Nonlin. Sci. Numer. Simul. 17(5), 249–257

2016.

[8] I.Nayak, “Numerical Study of MHD flow

and heat transfer of an unsteady third grade

Sidik, “Unsteady MHD on convective flow

of a Newtonian fluid past an inclined plate

in presence of chemical reaction with

radiation absorption and Dufour effects”

CFD Letters, 14, 7, 2022.

[10] Pawan Kumar Sharma and Sudhir Kumar

Chauhan, “Effect of MHD on unsteady

oscillatory couette flow through porous

media”, International journal of applied

mechanics and engineering, 27, 1, 88-202,

ARPN Journal of Engineering and Applied

Sciences, Vol. 10, No. 4, pp. 1839 – 1845,

2015.

[12] O.D. Makinde and T. Chinyoka,

“Numerical investigation of transient heat

transfer to hydromagnetic channel flow

with radiative heat and convective cooling”,

Commun. Nonlinear Sci. Numer. Simul.

Vol. 15, 2010, pp. 3919–3930.

[13] Mehran Yarahmadi, J.Robert Mahan and

with heat generation/absorption and

chemical reaction”, International

communications in heat and mass transfer,

147, 106956, 2023.

[15] Shafqat Hussain, Fatih Ertam, Mohamed

F.Oztop, and Nidal H.Abu-Hamdeh,

“Passive Contro of energy storage of

NePCM, heat and mass transfer with

gamma-shaped baffle in a thermo –

bioconvection system using CFD and

simulation of mass transfer in membranes

using hybrid machine learning models and

combustion thermal analysis”, Heat transfer

engineering, 2023.

[18] Khalil Ur Rehman, Andac Batur Colak and

Wasfi Shatanawi, “Artificial Neural

Networking (ANN) model for convective

heat transfer in thermally magnetized

multiple flow regimes with temperature

stratification effects”, Mathematics, 10,14,

1-19, 2022.

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

- M.Mahendran was responsible writing-original

draft, writing-review and editing.

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

No funding was received for conducting this study.

The authors have no conflicts of interest to declare.

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US