Computational sumulations and artificial inteligence for solving non-newtonian fluid flows

• Grant number: 21/07034-4
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: August 01, 2021
• End date: June 30, 2026
• Field of knowledge: Physical Sciences and Mathematics - Mathematics - Applied Mathematics
• Principal Investigator: Cassio Machiaveli Oishi
• Grantee: Fabio Vinicius Goes Amaral
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil
• Associated research grant: 13/07375-0 - CeMEAI - Center for Mathematical Sciences Applied to Industry, AP.CEPID
• Associated scholarship(s): 23/06035-2 - Sparse identification and machine learning techniques on viscoelastic fluid flows, BE.EP.DR

Abstract

Computational simulations of complex problems governed by differential equations have been significantly improved with advances in Artificial Intelligence (AI) techniques, which can be combined with classical numerical methods. In the current proposal, different AI techniques will be investigated, such as Machine/Deep Learning (ML/DL), Physics Informed Neural Network (PINN), Convolutional Neural Network (CNN) for numerical solutions of mathematical models adopted in incompressible non-newtonian fluid flows. It is worth noting that both the scientific community and industrial sectors have a strong interest in understanding the behavior of non-newtonian fluids that appear in different applications. The study proposal of this project aims to add improvements in finite difference methods for solving viscoelastic fluid flows in confined geometries and domains with moving boundary. AI techniques will be studied not only to accelerate the numerical convergence of classical methods, but also to estimate some dimensionless parameters used in the governing equations. (AU)