In recent years, machine learning has been used to create data-driven solutions to problems for which an algorithmic solution is intractable, as well as fine-tuning existing algorithms. This research applies machine learning to the development of an improved finite-volume method for simulating PDEs with discontinuous solutions. Shock-capturing methods make use of nonlinear switching functions that

Turbulence models facilitated by Kolmogorov’s theory play an important role for compressible flows. Typically the basis of these models is the power spectrum of the velocity \({\mathbf {u}}\) or of the density-weighted velocity \({\mathbf {w}}\equiv \rho ^{1/3}{\mathbf {u}}\). While for incompressible flow the quantity turbulent kinetic energy characterises turbulent motions, from the thermodynamic

Previous research studies indicate that the proportion of Magnus force at the spinning projectile tail position is very high. Meanwhile, the large mutations of aerodynamic characteristics are found after adding boattail structures. In order to study the influences of boattail structures on aerodynamics of a spinning projectile, a 6.37-diameter long tangential-ogive-cylinder projectile is selected as

We present results of three-dimensional direct numerical simulations (DNS) and large eddy simulations (LES) of turbulent gravity currents with a discontinuous Galerkin finite elements method. In particular, we consider the lock-exchange test case as a benchmark for gravity currents. Since, to the best of our knowledge, non-Boussinesq three-dimensional reference DNS are not available in the literature

We propose a method to construct a reduced order model with machine learning for unsteady flows. The present machine-learned reduced order model (ML-ROM) is constructed by combining a convolutional neural network autoencoder (CNN-AE) and a long short-term memory (LSTM), which are trained in a sequential manner. First, the CNN-AE is trained using direct numerical simulation (DNS) data so as to map the

We develop an unsupervised machine learning algorithm for the automated discovery and identification of traveling waves in spatiotemporal systems governed by partial differential equations (PDEs). Our method uses sparse regression and subspace clustering to robustly identify translational invariances that can be leveraged to build improved reduced-order models (ROMs). Invariances, whether translational

Many previous studies have shown that the fidelity of three-dimensional cardiovascular flow simulations depends strongly on inflow and outflow boundary conditions that accurately describe the characteristics of the larger vascular network. These boundary conditions are generally based on lower-dimensional models that represent the upstream or downstream flow behavior in some aggregated fashion. However

This paper considers membranes of globular structure in the framework of the cell model technique. The flow of a micropolar fluid through a spherical cell consisting of a solid core, porous layer and liquid envelope is modeled using coupled micropolar and Brinkman-type equations. The solution is obtained in analytical form. Boundary value problems with different conditions on the hypothetical cell

We present a systematic approach for determining the optimal actuator location for separation control from input–output response data, gathered from numerical simulations or physical experiments. The Eigensystem realization algorithm is used to extract state-space descriptions from the response data associated with a candidate set of actuator locations. These system realizations are then used to determine

We introduce a new approach for adaptive mesh refinement in which adaptivity is driven by low rank decomposition and optimal sensing of the dynamically evolving flow field. This method seeks an ordered set of locations for mesh adaptation from the instantaneous data-driven basis of an online proper orthogonal decomposition of the velocity, which organizes features into sparse optimal orthogonal modes

An inviscid vortex shedding model for separated vortices from a solid body is studied. The model describes the separated vortices by vortex sheets and the attached flow via conformal mapping. We develop a computational model to simulate the vortex shedding of a moving body, with varying angle. An unsteady Kutta condition is imposed on the edges of the plate to determine the edge circulations and velocities

We numerically study the dynamics of a hollow water droplet falling in the air under the action of gravity. The focus of our study is to investigate the effects of the difference in radii (thickness) of the hollow droplet, gravity and surface tension at the air–water interface on shape oscillations and the breakup dynamics of the hollow droplet. We found that the oscillations of the inner interface

Thermomagnetic convection is based on the use of external magnetic fields to better control heat transfer fluxes in ferrofluids, finding important applications in engineering and many related areas. The improvement of such methods relies on fundamentally understanding the flow of ferrofluids under temperature gradients and external magnetic fields. However, the underlying physics of this phenomenon

The numerical study of the wake of full-scale slender devices such as aircraft wings and wind turbine blades requires high-fidelity large eddy simulation tools. The broad spectrum of scales involved entails the use of coarse-grain models for the devices. Actuator disk or line methods have been developed for that purpose, and are to date the most employed in that context. These methods transfer the

Surface acoustic waves have gained much attention in flow control given the effects arising from acoustic streaming. In this study, the hydrodynamic interference of a drop under surface acoustic waves is comprehensively investigated and the contact angle hysteresis effects are considered, too. This paper reveals the effects of some control parameters such as wave amplitude and wave frequency on the

The present article deals with the spatiotemporal reduction of a reacting, supersonic, turbulent jet. A flowfield dataset is first obtained from a LES simulation including chemical reactions. The spatial reduction is accomplished by performing successively a Fourier transform in the azimuthal direction and a proper orthogonal decomposition (POD), while the temporal reduction is obtained through a selection

The aerodynamic characteristics of two neighboring airfoils are greatly different from those of a single airfoil, for both attached and detached flow conditions. In order to study the features of a two-airfoil arrangement with variations in the angle of attack and distances between the airfoils, and considering possible flow detachments, an adaptation of a discrete-time vortex numerical method is conducted

The need to numerically represent a free vortex system arises frequently in fundamental and applied research. Many possible techniques for realizing this vortex system exist but most tend to prioritize accuracy either inside or outside of the vortex core, which therefore makes them unsuitable for a stability analysis considering the entire flow field. In this article, a simple method is presented that

The present study aims to simulate a collision of two droplets containing immiscible liquids by employing a three- dimensional incompressible smoothed particle hydrodynamics (SPH) method, with models implemented for the computation. The simulation of a head-on collision of two droplets, both of which contain the same glycerol solution, showed that the droplet behavior agrees very well with that observed

The problem of finding optimal forcing and response for unbounded base flows, exemplified by the Blasius boundary layer, is assessed by means of a locally parallel resolvent analysis. A new analysis of previous results in the literature, which stated that a maximum resolvent gain occurs for spanwise wavenumber \(k_z \approx 0.2\), revealed that this result was not domain converged, and larger domains

Molecular dynamics simulation is used to study flow rate behavior of monoatomic fluid through carbon nanotube (CNT) against the pore diameter. All armchair and zigzag CNTs with diameters below 1.5 nm were considered. Fluid flow rate versus diameter is investigated, and discrepancy was observed in the results. Its non-monotonic behavior is reported and attributed to diameter-dependent potential energy

We apply supervised machine learning techniques to a number of regression problems in fluid dynamics. Four machine learning architectures are examined in terms of their characteristics, accuracy, computational cost, and robustness for canonical flow problems. We consider the estimation of force coefficients and wakes from a limited number of sensors on the surface for flows over a cylinder and NACA0012

In the present study, we investigate different data-driven parameterizations for large eddy simulation of two-dimensional turbulence in the a priori settings. These models utilize resolved flow field variables on the coarser grid to estimate the subgrid-scale stresses. We use data-driven closure models based on localized learning that employs a multilayer feedforward artificial neural network with

The structure of a shock wave is investigated using the continuum hypothesis for steady one-dimensional flow of a viscous non-ideal gas under heat conduction. The coefficients of viscosity and heat conductivity are assumed to be directly proportional to a power of the temperature and density of the gas. The simplified van der Waals equation of state for the non-ideal gas has been assumed in this work

In this work, we perform full-state LQR feedback control of fluid flows using non-intrusive data-driven reduced-order models. We propose a model reduction method called low-rank dynamic mode decomposition (lrDMD) that solves for a rank-constrained linear representation of the dynamical system. lrDMD is shown to have lower data reconstruction error compared to standard optimal mode decomposition (OMD)

The characteristics of the vortex flow with a free surface formed in a vertical cylindrical container filled with water, where the source of motion is the disk at the bottom endwall, are studied experimentally and analytically. Different types of oils and petroleum are used in experiments as admixtures. The problem of an “oil body” form in a complex vortex flow is considered on the basis of the analysis

The compressibility effects on the transition to turbulence in a spatially developing, compressible plane free shear layer are investigated via direct numerical simulation using a high-order discontinuous spectral element method for three different convective Mach numbers of 0.3, 0.5, and 0.7. The location of the laminar–turbulent transition zone is predicted by the analyses of vorticities, Reynolds

Periodic-forcing jets are applied to form a zero-net-mass-flux actuator for active flow control study on the \(25^{\circ }\) Ahmed body. Unlike traditional synthetic jet, this new proposed zero-net-mass-flux actuator has managed to divide the blowing phase and suction phase apart on different separation edges on the rear part of the body. Numerical simulation based on the large eddy simulation method

The control of complex systems is of critical importance in many branches of science, engineering, and industry, many of which are governed by nonlinear partial differential equations. Controlling an unsteady fluid flow is particularly important, as flow control is a key enabler for technologies in energy (e.g., wind, tidal, and combustion), transportation (e.g., planes, trains, and automobiles), security

An exegesis of a novel mechanism leading to vortex splitting and subsequent shedding that is valid for two-dimensional incompressible, inviscid or viscous, and external or internal or wall-bounded flows, is detailed in this research. The mechanism, termed the Vortex-Shedding Mechanism (VSM), is simple and intuitive, requiring only two coincident conditions in the flow: (1) the existence of a location

It has long been recognized that aortic root elasticity helps to ensure efficient aortic valve closure, but our understanding of the functional importance of the elasticity and geometry of the aortic root continues to e-volve as increasingly detailed in vivo imaging data become available. Herein, we describe a fluid-structure interaction model of the aortic root, including the aortic valve leaflets