Data-driven and machine learning in non-Newtonian fluid mechanics

• Grant number: 21/13833-7
• Support Opportunities: Scholarships abroad - Research
• Start date: August 01, 2022
• End date: July 31, 2023
• Field of knowledge: Physical Sciences and Mathematics - Mathematics - Applied Mathematics
• Principal Investigator: Cassio Machiaveli Oishi
• Grantee: Cassio Machiaveli Oishi
• Host Investigator: Steven Brunton
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil
• Institution abroad: University of Washington, United States
• Associated research grant: 13/07375-0 - CeMEAI - Center for Mathematical Sciences Applied to Industry, AP.CEPID

Abstract

This proposal sets out a research plan to investigate data-driven, machine learning and numerical methods for dealing with non-Newtonian fluid flows, as for instance, viscoelastic fluids. While the application of machine learning has been already considered as an important tool in fluid mechanics researches for Newtonian fluids, the construction and application of frameworks for non-Newtonian fluids are in early progress stages and remain largely unexplored. Based on simulation data of computational benchmark problems, the flow dynamics for viscoelastic fluids will be studied by learning algorithms. Moreover, to identify underlying functional form of the nonlinear physics observed in non-Newtonian fluid flows, as for instance viscoelastic flow instabilities and elastic turbulence, sparse identification modelling frameworks will be investigated. Fundamental numerical analysis, for instance convergence analysis, interpretability and generalizability of the methods will be discussed since these topics remain an open challenge in modeling physical systems using machine learning. In addition to improving data driven methods, this project has as the main objective the construction of new frameworks for boosting simulations of complex fluids in complex geometries, including effective data-driven strategies to solve applications from the non-Newtonian fluid mechanics. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

GONCALVES, M. B.; GUDINO, E.; MAIA, M.; OISHI, C. M.. Mathematical modeling for drug delivery and inflammation process: An application in macular edema. Applied Mathematical Modelling, v. 121, p. 22-pg., 2023-05-27. (13/07375-0, 21/13833-7)

EVANS, J. D.; PALHARES JUNIOR, I. L.; OISHI, C. M.; RUANO NETO, F.. Numerical verification of sharp corner behavior for Giesekus and Phan-Thien-Tanner fluids. Physics of Fluids, v. 34, n. 11, p. 28-pg., 2022-11-01. (21/13833-7, 21/05727-2, 13/07375-0)