Burak Aksoylu
Texas A&M University-San Antonio, TX·Mathematics & Statistics
Computer simulations of materials and fluids often fail where the physics turns abrupt, such as at a crack in a solid or a sharp jump in a flow. Methods that handle these features well tend to be very expensive to run.
- Fields
Mathematics & Statistics
Engineering
Computer Science
- Topics
scientific machine learning
SciML
peridynamics
- Grants
- 1 known grant, through Mar 2027
What they work on
- Use machine learning to improve iterative local-to-nonlocal coupling, so that peridynamics models only the region where fracture occurs and cheaper local models handle the rest.
- Develop neural network methods for complex fluid flow that use the peridynamic differential operator, a tool that turns derivatives into integrals, to capture sharp gradients and singularities.
- Build undergraduate research opportunities and curriculum in scientific machine learning through a partnership with the Institute for Mathematical and Statistical Innovation.
More about the research
This project combines machine learning (ML) with peridynamics (PD), a mathematical approach built on integrals rather than derivatives, to make such problems practical to compute. It also builds research capacity in scientific machine learning (SciML) for undergraduates at a primarily undergraduate, Hispanic-serving institution.
Why it matters
Simulating fractures and sharp-gradient fluid flows is costly, and current physics-guided machine learning can lose accuracy in exactly these situations. By making such simulations more feasible and accurate, the work could help engineers who model cracking materials and complex flows. It also trains undergraduates at a Hispanic-serving institution in computational and coding skills for STEM careers.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Tamer Zaki
Johns Hopkins University, MD·Engineering
Develop evolutional deep neural network (EDNN) algorithms that solve high-dimensional partial differential equations by evolving network parameters according to the governing physics, without relying on training data.
Sri Hari Krishn Narayanan
University of Chicago, IL·Computer Science
Develop an open-source, general-purpose automatic differentiation framework in Julia that computes exact derivatives of program code for both large-scale Earth system models and scientific machine learning methods.
Grady Wright
Boise State University, ID·Mathematics & Statistics
Develop the Unified Neural Operator formalism, which splits neural operators into an input encoder, output basis functions and a projection operator, so that existing architectures fit one framework and new ones can be designed.
The school
Texas A&M University-San Antonio: 2 labs on PI Pages.
Reach out
A start for your email. Fill in the brackets.