Tag
This paper introduces a physics-aware autoencoder-based latent-space framework for reduced-order forward modeling and variational parameter estimation in parametric dynamical systems, demonstrated on computational fluid dynamics benchmarks. The method enables differentiable surrogate-based inverse modeling and shows improved calibration robustness under realistic noisy or partial observations.
The paper develops neural operator surrogates to predict hydrodynamic fields (velocity, vorticity, pressure) around immersed-boundary soft swimmers, achieving low global relative errors on held-out trajectories while identifying pressure accuracy and physical consistency as areas for further work.
Proposes MHLF, a multigrid-hierarchical learning framework that accelerates engineering-scale 3D aircraft CFD simulations by 3-8x while preserving high-fidelity accuracy across subsonic, transonic, and supersonic regimes.
This paper introduces 'Data-Guided FVM-PINN', a framework using finite-volume losses for 2D shallow water equations, demonstrating that sparse data guidance is crucial to prevent network collapse in rugged loss landscapes.
This paper presents a novel deep learning approach to predict inertial lift forces in microfluidic devices without explicit geometric parameters, enabling better generalization to unseen channel cross-sections compared to previous models.