arXiv subjects

Mahmoud Golestanian

Publications and source records attributed to Mahmoud Golestanian.

4 recordsLinked to original sources

arXiv2026

AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling

High-fidelity CFD is essential for aerodynamic design, but repeated simulations are computationally expensive, motivating surrogate models for rapid evaluation across geometries and operating conditions. Most existing surrogates are designed for direct field regression, requiring the evaluation of millions of field points even when only an aerodynamic quantity or a localized region is needed, while their internal representations are not intended for direct use in downstream tasks. We introduce AeroJEPA, a framework inspired by joint-embedding predictive architectures that represents the problem in two distinct latent spaces: context tokens encode geometry, while predicted tokens encode the aerodynamic state. Both representations remain directly accessible for downstream tasks, such as linear readouts of design variables and aerodynamic quantities without decoding and integrating the full field. When spatial detail is needed, a continuous implicit decoder evaluates the field only at the requested coordinates while reusing the encoded geometry. We evaluate AeroJEPA on HiLiftAeroML, with multi-million-point fields, and SuperWing, which spans a broad family of transonic wings. Compared with state-of-the-art direct-regression surrogates, AeroJEPA trades peak full-field accuracy for compact, reusable representations. In our selective-decoding experiment, however, AeroJEPA substantially outperforms the evaluated direct-regression surrogates while avoiding predictions over the remainder of the aircraft. The learned representations further support controlled interpolation, concept-vector arithmetic, and preliminary constrained latent-space optimization. These results show how predictive representations can support aerodynamic analysis with or without full-field reconstruction.

cs.LG↗

arXiv2026

Agentic Exploration of PDE Spaces using Latent Foundation Models for Parameterized Simulations

Flow physics and more broadly physical phenomena governed by partial differential equations (PDEs), are inherently continuous, high-dimensional and often chaotic in nature. Traditionally, researchers have explored these rich spatiotemporal PDE solution spaces using laboratory experiments and/or computationally expensive numerical simulations. This severely limits automated and large-scale exploration, unlike domains such as drug discovery or materials science, where discrete, tokenizable representations naturally interface with large language models. We address this by coupling multi-agent LLMs with latent foundation models (LFMs), a generative model over parametrised simulations, that learns explicit, compact and disentangled latent representations of flow fields, enabling continuous exploration across governing PDE parameters and boundary conditions. The LFM serves as an on-demand surrogate simulator, allowing agents to query arbitrary parameter configurations at negligible cost. A hierarchical agent architecture orchestrates exploration through a closed loop of hypothesis, experimentation, analysis and verification, with a tool-modular interface requiring no user support. Applied to flow past tandem cylinders at Re = 500, the framework autonomously evaluates over 1,600 parameter-location pairs and discovers divergent scaling laws: a regime-dependent two-mode structure for minimum displacement thickness and a robust linear scaling for maximum momentum thickness, with both landscapes exhibiting a dual-extrema structure that emerges at the near-wake to co-shedding regime transition. The coupling of the learned physical representations with agentic reasoning establishes a general paradigm for automated scientific discovery in PDE-governed systems.

cs.AI↗

arXiv2023

Vanquishing volumetric locking in quadratic NURBS-based discretizations of nearly-incompressible linear elasticity: CAS elements

Quadratic NURBS-based discretizations of the Galerkin method suffer from volumetric locking when applied to nearly-incompressible linear elasticity. Volumetric locking causes not only smaller displacements than expected, but also large-amplitude spurious oscillations of normal stresses. Continuous-assumed-strain (CAS) elements have been recently introduced to remove membrane locking in quadratic NURBS-based discretizations of linear plane curved Kirchhoff rods (Casquero et al., CMAME, 2022). In this work, we propose two generalizations of CAS elements (named CAS1 and CAS2 elements) to overcome volumetric locking in quadratic NURBS-based discretizations of nearly-incompressible linear elasticity. CAS1 elements linearly interpolate the strains at the knots in each direction for the term in the variational form involving the first Lamé parameter while CAS2 elements linearly interpolate the dilatational strains at the knots in each direction. For both element types, a displacement vector with C1 continuity across element boundaries results in assumed strains with C0 continuity across element boundaries. In addition, the implementation of the two locking treatments proposed in this work does not require any additional global or element matrix operations such as matrix inversions or matrix multiplications. The locking treatments are applied at the element level and the nonzero pattern of the global stiffness matrix is preserved. The numerical examples solved in this work show that CAS1 and CAS2 elements, using either two or three Gauss-Legrendre quadrature points per direction, are effective locking treatments since they not only result in more accurate displacements for coarse meshes, but also remove the spurious oscillations of normal stresses.

cs.CE↗

arXiv2022

Removing membrane locking in quadratic NURBS-based discretizations of linear plane Kirchhoff rods: CAS elements

NURBS-based discretizations suffer from membrane locking when applied to primal formulations of curved thin-walled structures. We consider linear plane curved Kirchhoff rods as a model problem to study how to remove membrane locking from NURBS-based discretizations. In this work, we propose continuous-assumed-strain (CAS) elements, an assumed strain treatment that removes membrane locking from quadratic NURBS for an ample range of slenderness ratios. CAS elements take advantage of the C1 inter-element continuity of the displacement vector given by quadratic NURBS to interpolate the membrane strain using linear Lagrange polynomials while preserving the C0 inter-element continuity of the membrane strain. CAS elements are the first NURBS-based element type able to remove membrane locking for a broad range of slenderness ratios that combines the following characteristics: (1) No additional degrees of freedom are added, (2) No additional systems of algebraic equations need to be solved, and (3) The nonzero pattern of the stiffness matrix is preserved. Since the only additional computations required by the proposed element type are to evaluate the derivatives of the basis functions and the unit tangent vector at the knots, the proposed scheme barely increases the computational cost with respect to the locking-prone NURBS-based discretization of the primal formulation. The benchmark problems show that the convergence of CAS elements is independent of the slenderness ratio while the convergence of quadratic NURBS elements, local Bbar elements, and local ANS elements depends heavily on the slenderness ratio. The numerical examples also show how CAS elements remove the spurious oscillations in stress resultants caused by membrane locking while quadratic NURBS elements, local Bbar elements, and local ANS elements suffer from large-amplitude spurious oscillations in stress resultants.

cs.CE↗