Intelligent computing technique based supervised learning for squeezing flow model

Maryam Mabrook Almalki; Eman Salem Alaidarous; Dalal Adnan Maturi; Muhammad Asif Zahoor Raja; Muhammad Shoaib

doi:10.1038/s41598-021-99108-z

Intelligent computing technique based supervised learning for squeezing flow model

Sci Rep. 2021 Oct 1;11(1):19597. doi: 10.1038/s41598-021-99108-z.

Authors

Maryam Mabrook Almalki^{1

2}, Eman Salem Alaidarous³, Dalal Adnan Maturi³, Muhammad Asif Zahoor Raja⁴, Muhammad Shoaib⁵

Affiliations

¹ Department of Mathematics, Faculty of Science, Umm Al-Qura University, Makkah, 24211, Saudi Arabia. mmmalki@uqu.edu.sa.
² Department of Mathematics, Faculty of Science, King Abdulaziz University, Jeddah, 21589, Saudi Arabia. mmmalki@uqu.edu.sa.
³ Department of Mathematics, Faculty of Science, King Abdulaziz University, Jeddah, 21589, Saudi Arabia.
⁴ Future Technology Research Center, National Yunlin University of Science and Technology, Douliu, 64002, Taiwan.
⁵ Department of Mathematics, COMSATS University Islamabad, Attock Campus, Attock, 43600, Pakistan.

Abstract

In this study, the unsteady squeezing flow between circular parallel plates (USF-CPP) is investigated through the intelligent computing paradigm of Levenberg-Marquard backpropagation neural networks (LMBNN). Similarity transformation introduces the fluidic system of the governing partial differential equations into nonlinear ordinary differential equations. A dataset is generated based on squeezing fluid flow system USF-CPP for the LMBNN through the Runge-Kutta method by the suitable variations of Reynolds number and volume flow rate. To attain approximation solutions for USF-CPP to different scenarios and cases of LMBNN, the operations of training, testing, and validation are prepared and then the outcomes are compared with the reference data set to ensure the suggested model's accuracy. The output of LMBNN is discussed by the mean square error, dynamics of state transition, analysis of error histograms, and regression illustrations.