arXiv subjects

Simon Gsell

Publications and source records attributed to Simon Gsell.

4 recordsLinked to original sources

arXiv2026

On the rectification of oscillatory flows by flexible leaflets in a confined geometry

Inspired by biological systems, extensive research has explored how fluid-structure interactions in compliant channels and confined geometries control fluid transport. While local nonlinearities can be induced by individual components, arranging these elements into larger architectures gives rise to increasingly complex, collective responses. Predicting these collective behaviors, however, remains largely restricted to steady-state characterization, as the dynamic coupling between time-varying flows and multiple interacting structures is difficult to model. In this paper, we investigate numerically the collective interaction of multiple asymmetric leaflets within a channel at low-Reynolds number. By utilizing symmetrically oscillating plates rather than a pressure-driven flow to isolate the system from background asymmetries, we characterize how these interacting structures generate a net fluid transport. We develop an analytical framework to evaluate transport in the steady limit, which we subsequently extend to account for time-dependent channel oscillations, providing a complete dynamic description of the coupled fluid-structure system. Our results demonstrate that high leaflet densities maximize collective interactions and net transport. Furthermore, we define an elastoviscous number comparing viscous hydrodynamic forces to the restorative elastic forces of the leaflets, and uncover an optimal value that maximizes the net flow. This framework establishes a foundation for analyzing how collective slender structures interact dynamically within viscous environments, laying the groundwork for future studies on flow control in biological fluid transport and microfluidic design.

physics.flu-dyn↗

arXiv2025

Inferring viscoplastic models from velocity fields: a physics-informed neural network approach

Fluid-like materials are ubiquitous, spanning from living biological tissues to geological formations, and across scales ranging from micrometers to kilometers. Inferring their rheological properties remains a major challenge, particularly when traditional rheometry fails to capture their complex, three-dimensional, and often heterogeneous behavior. This difficulty is exacerbated by system size, boundary conditions, and other material-specific physical, chemical, or thermal constraints. In this work, we explore whether rheological laws can be inferred directly from flow observations. We propose a physics-informed neural network (PINN) framework designed to learn constitutive viscoplastic laws from velocity field data alone. Our method uses a neural network to interpolate the velocity field, enabling the computation of velocity gradients via automatic differentiation. These gradients are used to estimate the residuals of the governing conservation laws, which implicitly depend on the unknown rheology. We jointly optimize both the constitutive model and the velocity field representation by minimizing the physical residuals and discrepancies from observed data. We validate our approach on synthetic velocity fields generated from numerical simulations using Herschel-Bulkley, Carreau and Panastasiou models under various flow conditions. The algorithm reliably infers rheological parameters, even in the presence of significant noise. We analyze the dependence of inference performance on flow geometry and sampling, highlighting the importance of shear rate distribution in the dataset. Finally, we explore preliminary strategies for model-agnostic inference via embedded model selection, demonstrating the potential of PINNs for identifying the most suitable rheological law from candidate models.

physics.flu-dyn↗

arXiv2025

Pilot-wave hydrodynamics of a particle in a density-stratified fluid

Inspired by bouncing drop experiments that revealed how macroscopic systems can exhibit wave-particle properties previously thought to be exclusive to quantum systems, we introduce here a new wave-particle system based on internal gravity waves propagating in density-stratified fluids. Recent experiments on particles (called ludions) oscillating in such a fluid medium suggest that wave-particle interactions can induce symmetry breaking, leading to spontaneous self-propulsion of the particle in the horizontal plane. Here, we propose a minimal hydrodynamic theory showing that this instability can be explained by a Doppler force emerging from interactions between the ludion and its own wave field. We validate our theoretical predictions using direct numerical simulations, which confirm that the growth of the instability is determined by the particle oscillation amplitude. In wall-bounded domains, reflections of the internal waves create a Casimir-like potential that rapidly develops and constrains the particle motion. Despite the presence of the Doppler force, this potential governs the ludion long-term dynamics, leading to capture in fixed points or chaotic attractors near the potential wells. We show that the essential features of this behavior are well captured by a minimal dynamical model. Our findings establish the ludion as a novel hydrodynamic pilot-wave system, offering a new platform for exploring macroscopic wave-particle behaviors, particularly in three-dimensional configurations.

physics.flu-dyn↗

arXiv2021

Phase separation dynamics in deformable droplets

Phase separation can drive spatial organization of multicomponent mixtures. For instance in developing animal embryos, effective phase separation descriptions have been used to account for the spatial organization of different tissue types. Similarly, separation of different tissue types and the emergence of a polar organization is also observed in cell aggregates mimicking early embryonic axis formation. Here, we describe such aggregates as deformable two-phase fluid droplets, which are suspended in a fluid environment (third phase). Using hybrid finite-volume Lattice-Boltzmann simulations, we numerically explore the out-of-equilibrium routes that can lead to the polar equilibrium state of such a droplet (Janus droplet). We focus on the interplay between spinodal decomposition and advection with hydrodynamic flows driven by interface tensions, which we characterize by a Peclet number $Pe$. Consistent with previous work, for large $Pe$ the coarsening process is generally accelerated. However, for intermediate $Pe$ we observe long-lived, strongly elongated droplets, where both phases form an alternating stripe pattern. We show that these ``croissant'' states are close to mechanical equilibrium and coarsen only slowly through diffusive fluxes in an Ostwald-ripening-like process. Finally, we show that a surface tension asymmetry between both droplet phases leads to transient, rotationally symmetric states whose resolution leads to flows reminiscent of Marangoni flows. Our work highlights the importance of advection for the phase separation process in finite, deformable systems.

cond-mat.soft↗