Self-Supervised In-Context Operator Learning for Optimal Mixing in Incompressible Flows

DOI10.48550/ARXIV.2610.25556MaRDI QIDQ7401838arXivFDO

Authors Rongjie Lai, Weiwei Hu, Ziqian Li

Publication date 18 October 2026

Published in arXiv

Tldr We develop a self-supervised operator-learning framework that directly maps initial scalar fields to optimal mixing controls through Lagrangian reformulation of transport equations and particle-based Monte Carlo optimization, enabling efficient homogenization of passive scalars while minimizing kinetic energy in 2D and 3D incompressible flows.

Full work available at URL https://doi.org/10.48550/ARXIV.2610.25556

arXiv classification

math.NA

Cites work

• A multiscale measure for mixing
• A Structure-Preserving Numerical Scheme for Optimal Control and Design of Mixing in Incompressible Flows
• A survey of projection-based model reduction methods for parametric dynamical systems
• An approximating approach for boundary control of optimal mixing via Navier-Stokes flows
• Attention Is All You Need
• Cellular flow control design for mixing based on the least action principle
• Cellular mixing with bounded palenstrophy
• Control of stretching rate in time-periodic chaotic flows
• Exponential self-similar mixing by incompressible flows
• Feasibility, efficiency and transportability of short-horizon optimal mixing protocols
• Feedback control for fluid mixing via advection
• Finite Volume Methods for Hyperbolic Problems
• Fourier Neural Operator for Parametric Partial Differential Equations
• Geometrical structure of Laplacian eigenfunctions
• Hamiltonian Interface Dynamics for Reduced-Order Optimization of Incompressible Mixing
• In-Context Operator Learning with Data Prompts for Differential Equation Problems
• Mixing by stirring: optimizing shapes and strategies
• Mixing enhancement in binary fluids using optimised stirring strategies
• Mixing Versus Stirring
• Model predictive control: Theory and practice - a survey
• Numerical algorithms and simulations of boundary dynamic control for optimal mixing in unsteady Stokes flows
• Optimal mixing and optimal stirring for fixed energy, fixed power, or fixed palenstrophy flows
• Optimal stirring strategies for passive scalar mixing
• Quantification of mixing in aperiodic chaotic flows
• Symmetry concepts for the geometric analysis of mixing flows
• Topology optimization of fluids in Stokes flow
• Towards the design of an optimal mixer
• Unsupervised solution operator learning for mean-field games
• Using multiscale norms to quantify mixing and transport

This page was built for publication: Self-Supervised In-Context Operator Learning for Optimal Mixing in Incompressible Flows

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7401838)