Scholarlink

<p>This paper explores artificial neural networks (ANN) technique to investigate thermo-viscous fluid motion in between horizontal porous stretched plates bounded in a medium of permeability. The fluid motion is governed by the nonlinear coupled PDE’s governed by the velocity and temperature’s subjected to the associated boundary conditions. Runge Kutta 6th order method via shooting techniques using ND solver developed in Mathematica software was utilized to get the numerical results of the flow regulating equations including temperature and velocity. The algorithms on Feed-forward neural networks (FFNN) are used to find and study the behavior of the governed flow motion. The neural network with multi-layer perceptron (MLP) is employed to get the trial functions. The optimization algorithm, ADAMS estimation is used to find the adjustable factors. The Runge–Kutta and ANN results have been visually depicted and presented in tabular form for numerous values of physical coefficients like injection/suction parameter, Darcy’s permeable parameter, thermal coefficient of conductivity and thermo interaction stress coefficient. With the various values of hidden neurons and space points, the analysis of convergence has been studied for both the velocities and temperature predictions between boundary of the plates. The effectiveness of the results has raised within the networks for the increase of neuron’s and space points. The maximum R-Squared values were attained for fluid velocity (0.99898) and temperature (0.9997) predictions. The advantage of employing these models are, it will take less processing time and less storage space required for predictions. The present ANN models will be utilized to evaluate more complicated representations and models. The present machine learning algorithms are utilized to predict the temperature fields or heat transfer rates in systems with intricate temperature-viscosity connections. In domains including energy systems, materials processing, biomedical applications, and microfluidics, this analysis is extremely pertinent.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial Neural Networks Approach to Analyze Thermo-viscous Fluid Motion between Horizontal Porous Stretched Plates Bounded in a Porous Medium

  • 期刊论文

  • 发表日期：

    2025年10月21日

查看全文
  • N. Pothanna,
  • V. Ganesh Kumar,
  • G. Jithender Reddy,
  • L. Sandeep Raj

摘要

This paper explores artificial neural networks (ANN) technique to investigate thermo-viscous fluid motion in between horizontal porous stretched plates bounded in a medium of permeability. The fluid motion is governed by the nonlinear coupled PDE’s governed by the velocity and temperature’s subjected to the associated boundary conditions. Runge Kutta 6th order method via shooting techniques using ND solver developed in Mathematica software was utilized to get the numerical results of the flow regulating equations including temperature and velocity. The algorithms on Feed-forward neural networks (FFNN) are used to find and study the behavior of the governed flow motion. The neural network with multi-layer perceptron (MLP) is employed to get the trial functions. The optimization algorithm, ADAMS estimation is used to find the adjustable factors. The Runge–Kutta and ANN results have been visually depicted and presented in tabular form for numerous values of physical coefficients like injection/suction parameter, Darcy’s permeable parameter, thermal coefficient of conductivity and thermo interaction stress coefficient. With the various values of hidden neurons and space points, the analysis of convergence has been studied for both the velocities and temperature predictions between boundary of the plates. The effectiveness of the results has raised within the networks for the increase of neuron’s and space points. The maximum R-Squared values were attained for fluid velocity (0.99898) and temperature (0.9997) predictions. The advantage of employing these models are, it will take less processing time and less storage space required for predictions. The present ANN models will be utilized to evaluate more complicated representations and models. The present machine learning algorithms are utilized to predict the temperature fields or heat transfer rates in systems with intricate temperature-viscosity connections. In domains including energy systems, materials processing, biomedical applications, and microfluidics, this analysis is extremely pertinent.