arXiv subjects

Sushrut Kumar

Publications and source records attributed to Sushrut Kumar.

8 recordsLinked to original sources

arXiv2026

GPU-Accelerated Simulations of Moving Boundary Problems and Fluid-Structure Interaction at Extreme Scales

Computational fluid dynamics and fluid-structure interaction simulations involving moving and deforming bodies is extremely hard. In this work, we present a graphical processing unit (GPU) optimized implementation of the sharp-interface immersed boundary method. The method allows performing simulation around complex stationary as well as moving bodies on a Cartesian grid. We base our implementation on the ViCar3D framework and make use of OpenACC, CUDA, NCCL and MPI. We test the implementation across grid sizes ranging from O(10million) to O(1billion) points and achieved a 20X speedup compared to existing CPU implementation. We next present our multi-GPU implementation by utilizing CUDA streams and NCCL communicators. This enables us to obtain a >90% strong and weak scaling efficiencies. Next we demonstrate the capability of the developed software to simulate a turbulent fluid flow and coupled fluid-structure interaction in flapping bat wing in flight at Re=5000.

physics.comp-ph↗

arXiv2026

AeTHERON: Autoregressive Topology-aware Heterogeneous Graph Operator Network for Fluid-Structure Interaction

Surrogate modeling of body-driven fluid flows where immersed moving boundaries couple structural dynamics to chaotic, unsteady fluid phenomena remains a fundamental challenge for both computational physics and machine learning. We present AeTHERON, a heterogeneous graph neural operator whose architecture directly mirrors the structure of the sharp-interface immersed boundary method (IBM): a dual-graph representation separating fluid and structural domains, coupled through sparse cross-attention that reflects the compact support of IBM interpolation stencils. This physics-informed inductive bias enables AeTHERON to learn nonlinear fluid-structure coupling in a shared high-dimensional latent space, with continuous sinusoidal time embeddings providing temporal generalization across lead times. We evaluate AeTHERON on direct numerical simulations of a flapping flexible caudal fin, a canonical FSI benchmark featuring leading-edge vortex formation, large membrane deformation, and chaotic wake shedding across a 4x5 parameter grid of membrane thickness (h* = 0.01-0.04) and Strouhal number (St = 0.30-0.50). As a proof-of-concept, we train on the first 150 timesteps of a representative case using a 70/30 train/validation split and evaluate on the fully unseen extrapolation window t=150-200. AeTHERON captures large-scale vortex topology and wake structure with qualitative fidelity, achieving a mean extrapolation MAE of 0.168 without retraining, with error peaking near flapping half-cycle transitions where flow reorganization is most rapid -- a physically interpretable pattern consistent with the nonlinear fluid-membrane coupling. Inference requires milliseconds per timestep on a single GPU versus hours for equivalent DNS computation. This is a continuously developing preprint; results and figures will be updated in subsequent versions.

physics.comp-ph↗

arXiv2025

A GPU-Accelerated Sharp Interface Immersed Boundary Solver for Large Scale Flow Simulations

Immersed boundary methods (IBMs) facilitate the simulation of flows around stationary, moving, and deforming bodies on Cartesian grids. However, extending these simulations to the large grid sizes required for realistic flow problems remains a significant computational challenge. In this work, we present the implementation and acceleration of ViCar3D, a sharp-interface immersed boundary solver, on graphical processing units (GPUs). We utilize OpenACC, CUDA Fortran and MPI to reprogram \emph{ViCar3D}, a sharp-interface immersed boundary solver, on multi-GPU architectures. Verification and scalability studies are performed for two benchmark cases: two-dimensional flow past a circular cylinder and direct numerical simulation (DNS) of flow past a finite rectangular wing. For the latter, we observe an approximately 20X speedup (node-to-node comparison) relative to the CPU-based implementation. The GPU-accelerated solver is capable of simulating complex 3D flows with up to 200 million mesh points on a single node equipped with four GPUs. Strong and weak scaling tests demonstrate maximum scaling efficiencies of 92\% and 93\%, respectively, on multi-GPU systems. We further test the code to simulate fluid flow past complex-shaped single-body and multi-body cases.

physics.flu-dyn↗

arXiv2025

Flow-induced dorso-ventral deformation enhances propulsive efficiency in flexible caudal fins

Fish swim with flexible fins that stand in stark contrast to the rigid propulsors of engineered vehicles. Using numerical simulations of the dynamics of flow-structure interaction, we have found that dorso-ventral deformation in flexible caudal fins results in a 70% increase in efficiency of caudal fin swimmers compared to a rigid fin generating the same amount of thrust. By correlating fin deformation to the flow physics, we find that the greater power requirements of rigid fins can be largely attributed to their propensity to generate high-magnitude lateral forces. In contrast, flexible fins achieve high efficiency local-redirection of force where deformations orient pressure forces on the fin in fore-aft and dorso-ventral directions to reduce the power demand of generating thrust forces. These deformations occur at phases in the tail-beat cycle where the fin experiences large lateral velocities and pressure differentials and this reduces the net power expended by the flexible fins. In this way, the flexibility of a caudal fin offers a simple and elegant solution for efficient locomotion which does not require sensing, computation and control that might otherwise be provided by the nervous system of a fish or a computer within a underwater vehicle. These flow-induced dorso-ventral fin deformations therefore imbue a mechanical intelligence in these fins that provides propulsive advantages to caudal fin swimmers and they also offer solutions for efficient propulsion in engineered systems.

physics.flu-dyn↗

arXiv2025

Computational Modeling and Analysis of the Coupled Aero Structural Dynamics in Bat Inspired Wings

We employ a novel computational modeling framework to perform high-fidelity direct numerical simulations of aero-structural interactions in bat-inspired membrane wings. The wing of a bat consists of an elastic membrane supported by a highly articulated skeleton, enabling localized control over wing movement and deformation during flight. By modeling these complex deformations, along with realistic wing movements and interactions with the surrounding airflow, we expect to gain new insights into the performance of these unique wings. Our model achieves a high degree of realism by incorporating experimental measurements of the skeleton's joint movements to guide the fluid-structure interaction simulations. The simulations reveal that different segments of the wing undergo distinct aeroelastic deformations, impacting flow dynamics and aerodynamic loads. Specifically, the simulations show significant variations in the effectiveness of the wing in generating lift, drag, and thrust forces across different segments and regions of the wing. We employ a force partitioning method to analyze the causality of pressure loads over the wing, demonstrating that vortex-induced pressure forces are dominant while added mass contributions to aerodynamic loads are minimal. This approach also elucidates the role of various flow structures in shaping pressure distributions. Finally, we compare the fully articulated, flexible bat wing to equivalent stiff wings derived from the same kinematics, demonstrating the critical impact of wing articulation and deformation on aerodynamic efficiency.

physics.flu-dyn↗

arXiv2023

Force moment partitioning and scaling analysis of vortices shed by a 2D pitching wing in quiescent fluid

We experimentally study the dynamics and strength of vortices shed from a NACA 0012 wing undergoing sinusoidal pitching in quiescent water. We characterize the temporal evolution of the vortex trajectory and circulation over a range of pitching frequencies, amplitudes and pivot locations. By employing a physics-based force and moment partitioning method (FMPM), we estimate the vortex-induced aerodynamic moment from the velocity fields measured using particle image velocimetry. The vortex circulation, formation time and vorticity-induced moment are shown to follow scaling laws based on the feeding shear-layer velocity. The vortex dynamics, together with the spatial distribution of the vorticity-induced moment, provide quantitative explanations for the nonlinear behaviors observed in the fluid damping (Zhu et al., J. Fluid Mech., vol. 923, 2021, R2). The FMPM-estimated moment and damping are shown to match well in trend with direct force measurements, despite a discrepancy in magnitude. Our results demonstrate the powerful capability of the FMPM in dissecting experimental flow field data and providing valuable insights into the underlying flow physics.

physics.flu-dyn↗

arXiv2022

Contribution of spanwise and cross-span vortices to the lift generation of low-aspect-ratio wings: Insights from force partitioning

This study reports on the vortex-induced lift production mechanisms in low Reynolds number flows over low aspect-ratio rectangular wings. We use a rigorous force partitioning method which allows for the estimation of the pressure-induced aerodynamic loads due to distinct flow features or vortex structures in the flow around the wing. The specific focus of this work is on distinguishing the effect of spanwise and cross-span oriented vortex structures on pressure-induced lift production. We quantify the lift induced on the wing by these different vortices, and also estimate their influence within different regions of the flow-field around the wing and in the wake. By varying the aspect-ratio and angle-of-attack of the wing, we show that for most cases, the spanwise oriented vorticity contributes less to the total lift than cross-span oriented vortices. Furthermore, the spanwise vorticity in the near wake is capable of producing net negative lift on the wing and this is explained by separating and quantifying the influence of vortex cores and regions of strain in the wake. The results demonstrate the utility of the force partitioning method for dissecting the flow physics of vortex dominated flows.

physics.flu-dyn↗

arXiv2020

Parametric Analysis of Smartphone Camera for a Low Cost Particle Image Velocimetry System

This study focuses on assessing smartphone camera characteristics for developing an economic smartphone-based Particle Image Velocimetry (PIV) system. In the investigation, flow around a cylinder was visualized using two commercially-available smartphones (OnePlus 5T and iPhone X) cameras and low-intensity laser diodes. Hydrogen bubbles generated from electrolysis (termed Bubble Image Velocimetry) of aluminum electrodes were used as seeding medium. OpenPIV, an open-source toolbox, was used for processing captured images and obtaining the flow fields. A parametric analysis of the two smartphones was conducted across varying camera characteristics such as ISO, exposure compensation and frame rate. The results obtained through experimentation were compared with the results of a validated computational fluid dynamics (CFD) study with the same flow conditions and were found to be in good agreement, with deviation ranging from 1% to 3.5% for iPhone X and 1% to 7% for OnePlus 5T. It was observed that a higher frame rate results in greater accuracy of the measurement. Further, an exposure compensation of -1 EV and an ISO of 400 was found to produce results with the least error as compared to CFD values.

eess.IV↗