<p dir="ltr">Machine learning and deep learning techniques now provide innovative tools for addressing problems in biological, engineering, and physical systems. Physics-informed neural networks (PINNs) are a type of neural network that incorporate physical laws described by partial differential equations (PDEs) into their supervised learning tasks. This dissertation aims to enhance PINNs with improved training techniques and loss functions to tackle the complex physics of viscoelastic flow and rheology more effectively. The focus areas of the dissertation are listed as follows: i) Assigning relative weights to loss terms in training physics-informed neural networks (PINNs) is complex. We propose a solution using numerical integration via backward Euler discretization to leverage statistical properties of data for determining loss weights. Our study focuses on two and three-dimensional Navier-Stokes equations, using spatio-temporal velocity and pressure data to ascertain kinematic viscosity. We examine two-dimensional flow past a cylinder and three-dimensional flow within an aneurysm. Our method, tested for sensitivity and robustness against various factors, converges faster and more accurately than traditional PINNs, especially for three-dimensional Navier-Stokes equations. We validated our approach with experimental data, using the velocity field from PIV channel flow measurements to generate a reference pressure field and determine water viscosity at room temperature. Results showed strong performance with experimental datasets. Our proposed method is a promising solution for ’stiff’ PDEs and scenarios requiring numerous constraints where traditional PINNs struggle. ii) Machine learning algorithms are valuable for fluid mechanics, but high data costs limit their practicality. To address this, we present viscoelasticNet, a Physics-Informed Neural Network (PINN) framework that selects the appropriate viscoelastic constitutive model and learns the stress field from a given velocity flow field. We incorporate three non-linear viscoelastic models: Oldroyd-B, Giesekus, and Linear PTT. Our framework uses neural networks to represent velocity, pressure, and stress fields and employs the backward Euler method to construct PINNs for the viscoelastic model. The approach is multistage: first, it solves for stress, then uses stress and velocity fields to solve for pressure. ViscoelasticNet effectively learned the parameters of the viscoelastic constitutive model on noisy and sparse datasets. Applied to a two-dimensional stenosis geometry and cross-slot flow, our framework accurately learned constitutive equation parameters, though it struggled with peak stress at cross-slot corners. We suggest addressing this by exploring smaller domains. ViscoelasticNet can extend to other rheological models like FENE-P and extended Pom-Pom and learn entire equations, not just parameters. Future research could explore more complex geometries and three-dimensional cases. Complementing Particle Image Velocimetry (PIV), our method can determine pressure and stress fields once the constitutive equation is learned, allowing the modeling of future fluid applications. iii) Physics-Informed Neural Networks (PINNs) are widely used for solving inverse and forward problems in various scientific and engineering fields. However, most PINNs frameworks operate within the Eulerian domain, where physical quantities are described at fixed points in space. We explore coupling Eulerian and Lagrangian domains using PINNs. By tracking particles in the Lagrangian domain, we aim to learn the velocity field in the Eulerian domain. We begin with a sensitivity analysis, focusing on the time-step size of particle data and the number of particles. Initial tests with external flow past a cylinder show that smaller time-step sizes yield better results, while the number of particles has little effect on accuracy. We then extend our analysis to a real-world scenario: the interior of an airplane cabin. Here, we successfully reconstruct the velocity field by tracking passive particles. Our findings suggest that this coupled Eulerian-Lagrangian PINNs framework is a promising tool for enhancing traditional experimental techniques like particle tracking. It can be extended to learn additional flow properties, such as the pressure field for three-dimensional internal flows, and infer viscosity from passive particle tracking, providing deeper insights into complex fluids and their constitutive models. iv) Time-fractional differential equations are widely used across various fields but often present computational and stability challenges, especially in inverse problems. Leveraging Physics-Informed Neural Networks (PINNs) offers a promising solution for these issues. PINNs efficiently compute fractional time derivatives using finite differences and handle other derivatives via automatic differentiation. This study addresses two inverse problems: (1) anomalous diffusion and (2) fractional viscoelasticity. Our approach defines residual loss scaled with the standard deviation of observed data, using numerically generated and experimental datasets to learn fractional coefficients and calibrate parameters for the fractional Maxwell model. Our framework demonstrated robust performance for anomalous diffusion, maintaining less than 10% relative error in predicting the generalized diffusion coefficient and the fractional derivative order, even with 25% Gaussian noise added to the dataset. This highlights the framework’s resilience and accuracy in noisy conditions. We also validated our approach by predicting relaxation moduli for pig tissue samples, achieving relative errors below 10% compared to literature values. This underscores the efficacy of our fractional model with fewer parameters. Our method can be extended to model non-linear fractional viscoelasticity, incorporate experimental data for anomalous diffusion, and apply it to three-dimensional scenarios, broadening its practical applications.</p>

The author presents some generalized Burgers' equations for incompressible and isothermal flow of viscous non-Newtonian fluids based on the Cross model, the Carreau model, and the Power-Law model and some simple assumptions on the flows. The author numerically solves the traveling wave equations for the Cross model, the Carreau model, the Power-Law model by using industrial data. The author proves existence and uniqueness of solutions to the traveling wave equations of each of the three models. The author also provides numerical estimates of the shock thickness as well as maximum strain $\varepsilon_{11}$ for each of the fluids.

Nesse trabalho, foi desenvolvido um simulador numérico (C/C++) para a resolução de escoamentos de fluidos newtonianos incompressíveis, baseado no método de partículas Lagrangiano, livre de malhas, Smoothed Particle Hydrodynamics (SPH). Tradicionalmente, duas estratégias são utilizadas na determinação do campo de pressões de forma a garantir-se a condição de incompressibilidade do fluido. A primeira delas é a formulação chamada Weak Compressible Smoothed Particle Hydrodynamics (WCSPH), onde uma equação de estado para um fluido quase-incompressível é utilizada na determinação do campo de pressões. A segunda, emprega o Método da Projeção e o campo de pressões é obtido mediante a resolução de uma equação de Poisson. No estudo aqui desenvolvido, propõe-se três métodos iterativos, baseados noMétodo da Projeção, para o cálculo do campo de pressões, Incompressible Smoothed Particle Hydrodynamics (ISPH). A fim de validar os métodos iterativos e o código computacional, foram simulados dois problemas unidimensionais: os escoamentos de Couette entre duas placas planas paralelas infinitas e de Poiseuille em um duto infinito e foram usadas condições de contorno do tipo periódicas e partículas fantasmas. Um problema bidimensional, o escoamento no interior de uma cavidade com a parede superior posta em movimento, também foi considerado. Na resolução deste problema foi utilizado o reposicionamento periódico de partículas e partículas fantasmas. / In this work, we have developed a numerical simulator (C/C++) to solve incompressible Newtonian fluid flows, based on the meshfree Lagrangian Smoothed Particle Hydrodynamics (SPH) Method. Traditionally, two methods have been used to determine the pressure field to ensure the incompressibility of the fluid flow. The first is calledWeak Compressible Smoothed Particle Hydrodynamics (WCSPH) Method, in which an equation of state for a quasi-incompressible fluid is used to determine the pressure field. The second employs the Projection Method and the pressure field is obtained by solving a Poissons equation. In the study developed here, we have proposed three iterative methods based on the Projection Method to calculate the pressure field, Incompressible Smoothed Particle Hydrodynamics (ISPH) Method. In order to validate the iterative methods and the computational code we have simulated two one-dimensional problems: the Couette flow between two infinite parallel flat plates and the Poiseuille flow in a infinite duct, and periodic boundary conditions and ghost particles have been used. A two-dimensional problem, the lid-driven cavity flow, has also been considered. In solving this problem we have used a periodic repositioning technique and ghost particles.

Nesse trabalho, foi desenvolvido um simulador numérico (C/C++) para a resolução de escoamentos de fluidos newtonianos incompressíveis, baseado no método de partículas Lagrangiano, livre de malhas, Smoothed Particle Hydrodynamics (SPH). Tradicionalmente, duas estratégias são utilizadas na determinação do campo de pressões de forma a garantir-se a condição de incompressibilidade do fluido. A primeira delas é a formulação chamada Weak Compressible Smoothed Particle Hydrodynamics (WCSPH), onde uma equação de estado para um fluido quase-incompressível é utilizada na determinação do campo de pressões. A segunda, emprega o Método da Projeção e o campo de pressões é obtido mediante a resolução de uma equação de Poisson. No estudo aqui desenvolvido, propõe-se três métodos iterativos, baseados noMétodo da Projeção, para o cálculo do campo de pressões, Incompressible Smoothed Particle Hydrodynamics (ISPH). A fim de validar os métodos iterativos e o código computacional, foram simulados dois problemas unidimensionais: os escoamentos de Couette entre duas placas planas paralelas infinitas e de Poiseuille em um duto infinito e foram usadas condições de contorno do tipo periódicas e partículas fantasmas. Um problema bidimensional, o escoamento no interior de uma cavidade com a parede superior posta em movimento, também foi considerado. Na resolução deste problema foi utilizado o reposicionamento periódico de partículas e partículas fantasmas. / In this work, we have developed a numerical simulator (C/C++) to solve incompressible Newtonian fluid flows, based on the meshfree Lagrangian Smoothed Particle Hydrodynamics (SPH) Method. Traditionally, two methods have been used to determine the pressure field to ensure the incompressibility of the fluid flow. The first is calledWeak Compressible Smoothed Particle Hydrodynamics (WCSPH) Method, in which an equation of state for a quasi-incompressible fluid is used to determine the pressure field. The second employs the Projection Method and the pressure field is obtained by solving a Poissons equation. In the study developed here, we have proposed three iterative methods based on the Projection Method to calculate the pressure field, Incompressible Smoothed Particle Hydrodynamics (ISPH) Method. In order to validate the iterative methods and the computational code we have simulated two one-dimensional problems: the Couette flow between two infinite parallel flat plates and the Poiseuille flow in a infinite duct, and periodic boundary conditions and ghost particles have been used. A two-dimensional problem, the lid-driven cavity flow, has also been considered. In solving this problem we have used a periodic repositioning technique and ghost particles.

<p>Particle suspensions in viscoelastic fluids (e.g., polymeric fluids, liquid crystalline solutions, gels) are ubiquitous in industrial processes and in biology. In such fluids, particles often acquire lift forces that push them to preferential streamlines in the flow domain. This lift force depends greatly on the fluid’s rheology, and plays a vital role in many applications such as particle separations in microfluidic devices, particle rinsing on silicon wafers, and particle resuspension in enhanced oil recovery. Previous studies have provided understanding on how fluid rheology affects the motion of spherical particles in simple viscoelastic fluid flows such as shear flows. However, the combined effect of more complex flow profiles and particle shape is still under-explored. The main contribution of this thesis is to: (a) provide understanding on the migration and rotation dynamics of an arbitrary-shaped particle in complex flows of a viscoelastic fluid, and (b) develop guidelines for designing such suspensions for general applications.</p> <p><br></p> <p>In the first part of the thesis, we develop theories based on the second-order fluid (SOF) constitutive model to provide solutions for the polymeric force and torque on an arbitrary-shaped solid particle under a general quadratic flow field. When the first and second normal stress coefficients satisfy  <strong>Ψ</strong>₁  = −2 <strong>Ψ</strong> ₂ (corotational limit), the fluid viscoelasticity modifies only the fluid pressure and we provide exact solutions to the polymer force and torque on the particle. For a general SOF with  <strong>Ψ</strong> ₁ ≠  −2 <strong>Ψ</strong> ₂, fluid viscoelasticity modifies the shear stresses, and we provide a procedure for numerical solutions. General scaling laws are also identified to quantify the polymeric lift based on different particle shapes and orientation. We find that the particle migration speed is directly proportional to the length the particle spans in the shear gradient direction (L_sg), and that polymeric torques lead to unique orientation behavior under flow.</p> <p><br></p> <p>Secondly, we investigate the migration and rotational behavior of prolate and oblate spheroids in various viscoelastic, pressure-driven flows. In a 2-D slit flow, fluid viscoelasticity causes prolate particles to transition to a log-rolling motion where the particles orient perpendicular to the flow-flow gradient plane. This behavior leads to a slower overall migration speed (i.e., lift) of prolate particles towards the flow centerline compared to spherical particles of the same volume. In a circular tube flow, prolate particles align their long axis along the flow direction due to the extra polymer torque generated by the velocity curvature in all radial directions. Again, this effect causes prolate particles to migrate slower to the flow centerline than spheres of the same volume. For oblate particles, we quantify their long-time orientation and find that they migrate slower than spheres of the same volume, but exhibit larger migration speeds than prolate particles. Lastly, we examine the effect of normal stress ratio ? <strong>α</strong>  = <strong>Ψ</strong> ₂ /<strong>Ψ</strong>₁on the particle motion and find that this parameter only quantitatively impacts the particle migration velocity but has negligible effect on the rotational dynamics. We therefore can utilize the exact solution derived under the corotational limit (?<strong>α</strong> = −1/2) for a quick and reasonable prediction on the particle dynamics.</p> <p><br></p> <p>We next experimentally investigate the migration behavior of spheroidal particles in microfluidic systems and draw comparisons to our theoretical predictions. A dilute suspension of prolate/oblate microparticles in a density-matched 8% aqueous polyvinylpyrrolidone (PVP) solution is used as the model suspension system. Using brightfield microscopy, we qualitatively confirm our theoretical predictions for flow Deborah numbers 0 < De < 0.1 – i.e., that spherical particles show faster migration speed than prolate and oblate particles of the same volume in tube flows.</p> <p><br></p> <p>We finally design a holographic imaging method to capture the 3-D position and orientation of dynamic microparticles in microfluidic flow. We adopt in-line holography setup and propose a straightforward hologram reconstruction method to extract the 3-D position and orientation of a non-spherical particle. The method utilizes image moment to locate the particle and localize the detection region. We detect the particle position in the depth direction by quantifying the image sharpness at different depth position, and uses principal component analysis (PCA) to detect the orientation of the particle. For a semi-transparent particle that produces complex diffraction patterns, a mask based on the image moment information can be utilized during the image sharpness process to better resolve the particle position.</p> <p><br></p> <p>In the last part of this thesis, we conclude our work and discuss the future research perspectives. We also comment on the possible application of current work to various fields of research and industrial processes.</p> <p><br></p>

<p>Free surface flows, characterized by a deformable interface between two immiscible fluids or between a liquid and a gas, play a pivotal role in numerous natural phenomena and industrial processes. The fluid-fluid interface dynamics, governed by the complex interplay of forces such as inertia, capillary force, viscous force, and possibly elastic force, significantly influence the behavior of the fluids involved. Examples of free surface flows can be observed in everyday situations, such as droplet formation from a faucet, propagation and breaking of ocean waves, and tear films that coat the eye. An in-depth understanding of free surface flows and fluid-fluid interface dynamics has extensive implications for optimizing applications like inkjet printing, coating, spraying, and droplet formation while providing insights into the intricate behavior of natural fluid systems. Most of these applications, except for coating, involve abrupt and catastrophic topological changes of interfaces present in processes such as drop breakup, sheet rupture, and drop impact, where small droplets form from liquid sheets or filaments.</p> <p>This thesis examines the dynamics of contracting liquid filaments through computational means. Previous computational simulations have assumed that initially the fluid within the filament is quiescent which, however, may not typically be the case in practical applications. Here, the effect of a realistic, non-zero initial velocity profile is considered with the hypothesis that the fact that the fluid is already in motion when it starts to contract may result in significant alterations in the filament’s final fate vis-a-vis whether it breaks up into multiple small droplets or contracts into a sphere as its ends retract toward each other. The transient system of governing equations, the three-dimensional but axisymmetric (3DA) Navier-Stokes and continuity equations subjected to interfacial boundary conditions, are solved using rigorous and robust numerical algorithms in both fully 3DA and one-dimensional (1D) settings using the Galerkin finite element (GFEM) method. The simulation results are then used to construct comprehensive phase diagrams to delineate regions where filaments break up into smaller droplets from those where filaments contract to spheres without breakup.</p> <p>Polymer additives are often present in practical applications involving filament contraction and breakup. The presence of polymer molecules in an otherwise Newtonian solvent gives rise to non-Newtonian rheology. In this thesis, the dynamics of filaments containing polymer additives are analyzed using a 1D algorithm that is developed specifically for simulating viscoelastic free surface flows where the fluid’s rheology is described by the oft-used Oldroyd-B model. In real-world applications, filaments produced from nozzles are expected to be prestressed at the instant when they are created and begin to contract. It is demonstrated that the retraction velocity of tips of highly viscous, prestressed filaments is significantly increased compared to filaments in which the polymer molecules are initially relaxed and Newtonian filaments. This enhancement is explained by examining the value of f σ: D (σ: Elastic stress; D: Rate-of-strain tensor), which can be positive or negative. This quantity is positive when the flow does work on the polymer molecules but negative when the molecules do work on the flow, i.e., when elastic recoiling or unloading takes place. In prestressed filaments, elastic unloading takes place because σ: D < 0. The elastic stresses work by pulling the fluid in axially and pushing it out radially, thereby drastically increasing the tip velocity.  However, this enhancement in contraction velocity is not observed in low to intermediate viscosity prestressed filaments and whose Newtonian counterparts typically experience end-pinching. It has been established by others that end-pinching can be precluded in either filaments of intermediate viscosity or surfactant-laden filaments of low viscosity through a process known as escape from end-pinching. In this study, we demonstrate that a similar escape can also occur in prestressed viscoelastic filaments of low-to-intermediate viscosity, as revealed by one-dimensional numerical simulations and rationalized by examining when and where the elastic recoil takes place.</p> <p>Beyond cylindrical filaments, thin liquid films or planar liquid sheets are also prevalent in atomization, curtain coating, and other processes where liquid sheet stability has been a subject of extensive research. Numerous authors have examined wave formation and growth leading to sheet breakup. Free liquid films or sheets without edges or caps at their two ends, which typically have two free surfaces and are surrounded by air or sometimes another liquid, can destabilize and rupture due to intermolecular van der Waals attractive forces, despite the stabilizing influence of surface tension. In this thesis, the dynamics of contracting free films or sheets with caps---two-dimensional (2D) drops---of Newtonian fluids is examined without considering van der Waals forces to confirm or refute the hypothesis that such systems can rupture due to finite-amplitude perturbations even in the absence of intermolecular forces. In particular, both two-dimensional and one-dimensional high-accuracy simulations are employed to demonstrate that unlike inviscid 2D drops that can rupture in the absence of van der Waals forces, 2D drops or sheets can escape from pinch-off due to the action of viscous forces which are present in real systems no matter how small their viscosity. The reopening of the interface and escape from pinch-off in 2D drops and sheets are explained by demonstrating the key role played by vorticity. New power-law relations or scaling laws are obtained as a function of Ohnesorge number (ratio of viscous to the square root of the product of inertial and capillary forces) for the value of the minimum film thickness for which 2D drops or sheets stop thinning and after which the interface begins to reopen. Simple yet powerful arguments are presented rationalizing these scaling laws. It is expected that these power-law relations should be of great interest to experimentalists who study such phenomena by high-speed visualization experiments.</p> <p>Some of the motivation for this thesis research comes from crop spraying applications in which achieving zero or negligible drift is highly desirable. To further the understanding of fluid mechanics underpinning current and future drift reduction technologies, a simplified experimental setup is adopted to generate liquid sheets and analyze their disintegration into droplets. This new setup is both simpler and more universal than commonly utilized experimental systems that use single or multiple nozzles to generate liquid sheets and spray droplets from the disintegration of free liquid films. In the current experiments, droplets of test fluids are made to collide with or impact the top planar surface of a solid cylinder or rod. A series of MATLAB codes are developed and employed to extract droplet size distributions from images that are obtained from high-speed visualization experiments. The experimental setup and the means of data analysis are then used to probe the effect of fluid properties on the dynamics of sheet disintegration and droplet size distributions. It is hoped that future researchers will be able to combine what has been done in this thesis by simulations and in this chapter via experimental observations to develop an improved mechanistic understanding of spray formation.</p>

<p>Fibre suspensions in the fluid medium are common in industry, biology, and the environment. Industrial examples of concentrated suspensions include fresh concrete, uncured solid rocket fuel, and biomass slurries; natural examples include silt transfer in rivers and red blood cells in the blood.  These suspensions often include a Newtonian fluid as their suspending medium; still, these suspensions exhibit a plethora of non-Newtonian properties, such as yield stresses, rate-dependent rheology, and normal stresses, to name a few. Other than volume fraction, the type of fiber material, the presence of fluid-fiber or fiber-fiber interactions such as hydrodynamic, Brownian, colloidal, frictional, chemical, and/or electrostatic determine the rheological behavior of suspension. The average inter-fiber gaps between the neighboring fibers decrease significantly as the suspension volume fraction move towards a concentrated regime. As a result, in this regime, inter-fiber interactions become dominant.  Moreover, the surface asperities are present on the fiber surface even in the case of so-called smooth fibers, as fibers in real suspensions are not perfectly smooth. Hence, contact forces arising from the direct touching of the fibers become one of the essential factors in determining the rheology of suspensions.</p> <p>We first describe the causes of yield stress, shear thinning, and normal stress differences in fibre suspensions. We model the fibers as inextensible continuous flexible slender bodies with the Euler-Bernoulli beam equation governing their dynamics suspended in an incompressible Newtonian fluid. The fiber dynamics and fluid flow coupling is achieved using the immersed boundary method (IBM). In addition, the fiber surface roughness lead to inter-fiber contacts resulting in normal and tangential forces between the fibers, which follow Coulomb’s law of<br> friction. The surface roughness is modeled as hemispherical protrusions on the fiber surfaces. In addition to the comparison of the computational model to the experimental results, we demonstrate that attractive interactions lead to yield stress and shear thinning rheology.</p> <p>Furthermore, we investigate the effects of fiber aspect ratio, roughness, flexibility, and volume fraction on the rheology of concentrated suspensions. We find that the suspension viscosity increases with increasing the volume fraction, roughness, fiber rigidity, and aspect ratio. The increase in relative viscosity is the macroscopic manifestation of a similar increase  in the microscopic contact contribution with these parameters. In addition, we observe positive and negative first and second normal stress differences, respectively, in agreement with previous experiments. Lastly, we propose a modified Maron-Pierce law to quantify the the jamming volume fraction with varying fiber aspect ratio and roughness. Additionally, we provide a constitutive model to calculate the viscosity at various volume fractions, aspect ratios, and shear rates.</p>