Scientific Machine Learning in Computational Fluid Dynamics
March 14, 2027
at ASTFE TFEC2027 Conference in person at Southern Methodist University, Dallas, TX, USA and partially online virtual via Zoom and Whova (from 9 AM to 1 PM)
Computational Fluid Dynamics (CFD) has long been a major tool of engineering design and analysis, helping engineers model and understand complex flow physics across many applications. High-fidelity simulation still comes at a high computational cost, especially for many-query problems like design optimization, digital twins, and multiscale physics. Scientific Machine Learning (SciML) can act as a helpful aid alongside conventional CFD by combining physics-based modeling with data-driven methods to lower cost, generalize across geometries, and support learned components alongside conventional solvers. This course introduces CFD practitioners a range of topics from foundational SciML concepts to state-of-the-art techniques, including embedding machine learning in CFD/FEM solvers, physics-informed neural networks, operator networks, differentiable solvers, and generative AI and agent-driven CFD.
KEY BENEFITS TO PARTICIPANTS
- Landscape of Scientific machine learning (SciML): what exists, and which approach fits which class of problem from operator networks and embedded solvers to differentiable solvers
- See case studies (coolant filling, groundwater flow) showing what these methods actually achieve and where they fall short
- Understand emerging directions including generative models, differentiable solvers and AI agents for CFD workflows
- Understand on how to frame decision framework for identifying where ML can meaningfully accelerate your own CFD work
WHO SHOULD ATTEND
- CFD practitioners, engineers, and computational physicists looking to evaluate or adopt ML-augmented simulation methods
- Simulation and design engineers working with computationally expensive or geometry-heavy CFD workflows
- Graduate students and research scholars working at the intersection of CFD and machine learning
Instructor(s)
AKSHAI RUNCHAL
MADHUKAR M. RAO
ENROLL FOR THE course
Workshop Outline
Lecture 1: Machine Learning for CFD: Motivation & Applications Lecture 2: Scientific Machine Learning & Neural Operators Lecture 3: Machine Learning Accelerated CFD & Surrogate Models Lecture 4: Agent-Driven CFD & Differentiable SolversRegistration
Fee: $100.00 - Participation fee
Registration will be limited to 50 participants.
Registration deadline: February 14th 2027
A certificate will be given upon completion of the course
ASTFE
Tel: +1 203 733 6275
Fax: +1 203 456 6167
© Copyright reserved ASTFE, 2026
American Society of
Thermal and Fluids Engineers
Fax: +1 203 456 6167
© Copyright reserved ASTFE, 2026