arXiv subjects

Gokberk Kabacaoglu

Publications and source records attributed to Gokberk Kabacaoglu.

6 recordsLinked to original sources

arXiv2026

VesNet: Neural network accelerated solver for simulating Stokesian vesicle suspensions

Numerical simulation of deformable particle suspensions in Stokes flow is computationally expensive due to nonlinear fluid-structure interactions, evolving interfaces, and multiscale hydrodynamics. We present VesNet, a hybrid framework that accelerates two-dimensional vesicle suspension simulations by approximating vesicle self interactions, including background flow coupling and short-range lubrication forces, while retaining conventional modules for boundary reparameterization and far-field hydrodynamics. A GPU-accelerated implementation achieves over 100x speedup compared to a multithreaded MATLAB CPU boundary integral solver and about 5x relative to its GPU counterpart. VesNet accurately captures key dynamics, including single-vesicle phase behavior, pair interactions, and large-scale suspensions in Taylor-Green and Poiseuille flows, enabling efficient simulations of thousands of vesicles on modest computational resources.

physics.flu-dyn↗

arXiv2019

Machine learning acceleration of simulations of Stokesian suspensions

Particulate Stokesian flows describe the hydrodynamics of rigid or deformable particles in Stokes flows. Due to highly nonlinear fluid-structure interaction dynamics, moving interfaces, and multiple scales, numerical simulations of such flows are challenging and expensive. In this Letter, we propose a generic machine-learning-augmented reduced model for these flows. Our model replaces expensive parts of a numerical scheme with multilayer perceptrons. Given the physical parameters of the particle, our model generalizes to arbitrary geometries and boundary conditions without the need to retrain the regression function. It is 10 times faster than a state-of-the-art numerical scheme having the same number of degrees of freedom and can reproduce several features of the flow quite accurately. We illustrate the performance of our model on integral equation formulation of vesicle suspensions in two dimensions.

physics.comp-ph↗

arXiv2019

Optimal design of deterministic lateral displacement device for viscosity contrast based cell sorting

We solve a design optimization problem for deterministic lateral displacement (DLD) device to sort same-size biological cells by their deformability, in particular to sort red blood cells (RBCs) by their viscosity contrast between the fluid in the interior and the exterior of the cells. A DLD device optimized for efficient cell sorting enables rapid medical diagnoses of several diseases such as malaria since infected cells are stiffer than their healthy counterparts. The device consists of pillar arrays in which pillar rows are tilted and hence are not orthogonal to the columns. This arrangement leads cells to have different final vertical displacements depending on their deformability, therefore, it vertically separates the cells. Pillar cross section, tilt angle of the pillar rows and center-to-center distances between pillars define a unique device. For a given pair of viscosity contrast values of the cells we seek optimal DLD designs by fixing the tilt angle and the center-to-center distances. So the only design parameter is the pillar cross section which we parameterize with uniform 5th order B-splines. We propose an objective function to try to capture efficient cell sorting. The objective function is evaluated by simulating the cell flows through a device using our 2D model based on a boundary integral method (Kabacaoglu et al. Journal of Computational Physics, 357:43-77, 2018). We solve the optimization problem using the covariance matrix adaptation evolution strategy (CMA-ES), which is a stochastic, derivative-free algorithm. We present several scenarios where solving the optimization problem finds designs that can separate cells with similar viscosity contrast values. To the best of our knowledge, this is the first study which poses designing a DLD device as a constrained optimization problem and shows that solving this problem systematically discovers optimal designs.

physics.comp-ph↗

arXiv2018

Sorting same-size red blood cells in deep deterministic lateral displacement devices

We study deformability-based sorting of same-size RBCs via DLD using an in-house integral equation solver. Our goal is to quantitatively characterize the physical mechanisms that enable the cell separation. To this end, we systematically investigate the effects of the interior fluid viscosity and membrane elasticity of a cell on its behavior. In particular, we consider deep devices in which a cell can show rich dynamics such as taking a particular angular orientation depending on its mechanical property. We have found out that cells moving with a sufficiently high positive inclination angle with respect to the flow direction displace laterally while those with smaller angles travel with the flow streamlines. Thereby, deformability-based cell sorting is possible. The underlying mechanism here is cell migration due to the cell's positive inclination and the shear gradient. The higher the inclination is, the farther the cell can travel laterally. We also assess the efficiency of the technique for dense suspensions. It turns out that most of the cells in dense suspensions does not displace in the lateral direction no matter what their deformability is. As a result, separating cells using a DLD device becomes harder.

physics.comp-ph↗

arXiv2017

Low-resolution simulations of vesicle suspensions in 2D

Vesicle suspensions appear in many biological and industrial applications. These suspensions are characterized by rich and complex dynamics of vesicles due to their interaction with the bulk fluid, and their large deformations and nonlinear elastic properties. Many existing state-of-the-art numerical schemes can resolve such complex vesicle flows. However, even when using provably optimal algorithms, these simulations can be computationally expensive, especially for suspensions with a large number of vesicles. In this paper, we investigate the effect of a number of algorithmic empirical fixes in an attempt to make low-resolution simulations more stable and more predictive. Based on our empirical studies for a number of flow configurations, we propose a scheme that attempts to integrate these fixes in a systematic way. This low-resolution scheme is an extension of our previous work Quaife and Biros (2014) and Quaife and Biros (2016). Our low-resolution correction algorithms (LRCA) include anti-aliasing and membrane reparametrization for avoiding spurious oscillations in vesicles' membranes, adaptive time stepping and a repulsion force for handling vesicle collisions and, correction of vesicles' area and arc-length for maintaining physical vesicle shapes. We perform a systematic error analysis by comparing the low-resolution simulations of dilute and dense suspensions with their high-fidelity, fully resolved, counterparts. We observe that the LRCA enables both efficient and statistically accurate low-resolution simulations of vesicle suspensions, while it can be 10x to 100x faster.

physics.flu-dyn↗

arXiv2016

Quantification of mixing in vesicle suspensions using numerical simulations in two dimensions

We study mixing in Stokesian vesicle suspensions in two dimensions on a cylindrical Couette apparatus using numerical simulations. The vesicle flow simulation is done using a boundary integral method and the advection-diffusion equation for the mixing of the solute is solved using a pseudo-spectral scheme. We study the effect of the area fraction, the viscosity contrast between the inside (the vesicles) and the outside (the bulk) fluid, the initial condition of the solute, and the mixing metric. We compare mixing in the suspension with mixing in the Couette apparatus without vesicles. On the one hand, the presence of vesicles in most cases, slightly suppresses mixing. This is because the solute can be only diffused across the vesicle interface and not advected. On the other hand, there exist spatial distributions of the solute for which the unperturbed Couette flow completely fails to mix whereas the presence of vesicles enables mixing. We derive a simple condition that relates the velocity and solute and can be used to characterize the cases in which the presence of vesicles promotes mixing.

physics.flu-dyn↗