Incompressible Navier-Stokes Flow Research Articles

TetraFEM: Numerical Solution of Partial Differential Equations Using Tensor Train Finite Element Method

In this paper, we present a methodology for the numerical solving of partial differential equations in 2D geometries with piecewise smooth boundaries via finite element method (FEM) using a Quantized Tensor Train (QTT) format. During the calculations, all the operators and data are assembled and represented in a compressed tensor format. We introduce an efficient assembly procedure of FEM matrices in the QTT format for curvilinear domains. The features of our approach include efficiency in terms of memory consumption and potential expansion to quantum computers. We demonstrate the correctness and advantages of the method by solving a number of problems, including nonlinear incompressible Navier–Stokes flow, in differently shaped domains.

A multiblock (MIB) finite element method for accurate and efficient blood flow simulation

We present an efficient and accurate immersed boundary (IB) finite element (FE) method for internal flow problems with complex geometries (e.g. blood flow in the vascular system). In this study, we use a voxelized flow domain (discretized with hexahedral and tetrahedral elements) instead of a box domain, which is frequently used in IB methods. We highlight the applicability and efficiency of the proposed method in generating the flow domain mesh and removing unnecessary portions of the domain, compared to the standard IB methods that rely on a full Cartesian mesh. Our approach is attractive for internal flow problems that use IB methods. The proposed method accurately satisfies the boundary conditions on the immersed object using velocity and pressure correction procedures. The velocity correction applies implicitly such that the velocity on the immersed boundary (Lagrangian points) interpolated from the corrected velocity values computed on the mesh nodes (Eulerian nodes) accurately satisfies the prescribed velocity boundary conditions. The proposed method utilizes the well-established incremental pressure correction scheme (IPCS) FE solver, and the boundary condition-enforced IB (BCE-IB) method to numerically solve the transient, incompressible Navier-Stokes (NS) flow equations. We verify the accuracy of our numerical method using the analytical solution for the Poiseuille flow in a cylinder, and the available experimental data (laser Doppler velocimetry) for the flow in a three-dimensional 90° angle tube bend. We further examine the accuracy and applicability of the proposed method by considering flow within complex geometries, such as blood flow in aneurysmal vessels and the aorta, flow configurations that would otherwise be difficult to solve by most IB methods. Our method offers high accuracy, as demonstrated by the verification examples, and high applicability, as demonstrated through the solution of blood flow within complex geometry. The proposed method is efficient, since it is as fast as the traditional finite element method used to solve the Navier–Stokes flow equations, with a small overhead (not more than 5%) due to the numerical solution of a linear system formulated for the IB method.

Unraveling hydrodynamic interactions in fish schools: A three-dimensional computational study of in-line and side-by-side configurations

We numerically investigate the hydrodynamic interactions between a pair of three-dimensional (3D) fish-like bodies arranged in both in-line and side-by-side configurations. The morphology and kinematics of these fish-like bodies are modeled on a live rainbow trout (Oncorhynchus mykiss) observed during steady swimming in the laboratory. An immersed-boundary-method-based incompressible Navier–Stokes flow solver is employed to capture the flow dynamics around the fish-like bodies accurately. Our findings indicate that hydrodynamic performance of individual fish in both arrangements is influenced by their spatial separation when in close proximity as well as by the relative phase difference between the two fish. In the case of in-phase in-line schools, the leading fish experiences up to 5.3% increase in propulsive efficiency, attributed to the water blockage effect caused by the following fish. In comparison, the following fish experiences an increase in drag and power consumption along its body. Detailed analysis reveals that this rise in drag primarily results from an increase in friction drag (89%), driven by the amplified velocity field around the fish's body. Furthermore, altering the phase difference between the fish can help reduce pressure drag on the following fish by affecting the interaction between incoming vortex rings and its trunk. In side-by-side schools with in-phase swimming, a reduction of 6.8% in power consumption on the caudal fin is achieved for each fish when the transverse distance is maintained at 0.25 body lengths. Flow analysis reveals that the decrease in power usage is attributed to a diminished velocity field between the caudal fins, facilitating flow separation and subsequently reducing energy expenditure required for generating comparative thrust. For the out-of-phase swimming, the side-by-side school system experiences enhanced thrust production, owing to a wake energy recapture mechanism. The degree of enhancement varies for each fish and is determined by the specific phase difference. These insights obtained from our study hold the potential to inform the design and navigation strategies of underwater robotic swarms.

Numerical investigation of vortex-induced forces on a wind turbine tower segment at very high Reynolds numbers

The vortex-induced forces on an extruded cylinder with a span of two diameters representative of a finite segment of a non-tapered wind turbine tower at a very high Reynolds number (Re = 8.0×106) are numerically investigated using an incompressible Navier-Stokes flow solver with an Improved Delayed Detached Eddy Simulation (IDDES) turbulence model and correlation-based boundary layer transition modelling. The solution shows spanwise correlated structured vortex shedding with the Strouhal number St = 0.48. The boundary layer transition is found to occur at θ transition = 70 ◦ , and boundary layer separation is found to occur at θ separation = 120 ◦ . Results from the grid dependency study strongly imply that when using IDDES, the Strouhal number converges to higher values than previously reported by the literature as the grid is refined, with results ranging from St ∼ 0.44 using a grid with 4.2 × 106 cells, to St = 0.48 for the finest considered grid with 33 × 106 cells. This behaviour is not seen for URANS, where St = 0.33 for the finest grid.

Axial Green Function Method in an Efficient Projection Scheme for Incompressible Navier–Stokes Flow in Arbitrary Domains

We introduce a new approach to solving incompressible Navier–Stokes flow. This method combines a projection scheme with the Axial Green Function Method (AGM). Based on the Kim and Moin methods, our methodology employs a predictor–corrector mechanism to achieve stable and accurate velocity corrections. Using axial Green functions, we transform complex differential equations into simpler one-dimensional integral equations. These are strategically placed along a minimal axis-parallel lines, known as axial lines, within the flow domain. This transformation makes computation and analysis more efficient from a numerical viewpoint. A significant innovation in our approach is using one-dimensional axial Green functions tailored explicitly for the reaction–diffusion ordinary differential operator. These functions efficiently handle the discrete-time derivative and viscous terms of the momentum equation. Furthermore, our approach allows for the arbitrary construction of axis-parallel lines, facilitating analysis near critical flow regions and even enabling the random distribution of these lines. We validate our proposed method through numerical examples, demonstrating the convergence of numerical solutions, the effectiveness of arbitrarily constructed axis-parallel lines, and the potential extension of our method to three-dimensional flow problems. Additionally, this study provides a robust and adaptable alternative way to solve incompressible Navier–Stokes flows, proving its effectiveness in practical applications such as Tesla valves.

A pressure-projection pre-conditioning multi-fractional-step method for Navier–Stokes Flow in Porous Media

A new pressure-projection pre-conditioning multi-fractional-step (PPP-MFS) method for incompressible Navier–Stokes flow in porous media is presented. This fractional step method is applied to decouple the pressure and the velocity; thereby, overcoming the computational costs and difficulty associated with the resolution of the nonlinear term in the Navier–Stokes equation for fine geometric models. Specifically, time evolution is decomposed into a sequence of multi-fractional solution steps. In the first step, an elliptic problem is solved for pressure (p) with a no-slip boundary condition. This gives the Stokes pressure and velocity fields. In the second step, the obtained pressure p is then projected onto the field p∗ and used to solve for velocity field (u) required for the pre-conditioning of the solution to the Navier–Stokes equation. The pressure and velocity fields are obtained from the solution of the Navier–Stokes equation in the third step. Numerical and geometric discretisation of porous samples were carried out using finite-element method. For flow in simple channel models represented by two- and three-dimensions and in systems with high conductivity, the Stokes and Navier–Stokes numerical solutions produced close pressure and velocity field approximations. For flow around a cylinder, computation time is consistently higher in the Navier–Stokes equation by a factor of 2 with a pronounced non-symmetric pressure field at high mesh refinements. This computation time is desirable given that Navier–Stokes computation without preconditioning can be orders of magnitude more expensive.

Monge–Ampère geometry and vortices

We introduce a new approach to Monge–Ampère geometry based on techniques from higher symplectic geometry. Our work is motivated by the application of Monge–Ampère geometry to the Poisson equation for the pressure that arises for incompressible Navier–Stokes flows. Whilst this equation constitutes an elliptic problem for the pressure, it can also be viewed as a non-linear partial differential equation connecting the pressure, the vorticity, and the rate-of-strain. As such, it is a key diagnostic relation in the quest to understand the formation of vortices in turbulent flows. We study this equation via an associated (higher) Lagrangian submanifold in the cotangent bundle to the configuration space of the fluid. Using our definition of a (higher) Monge–Ampère structure, we study an associated metric on the cotangent bundle together with its pull-back to the (higher) Lagrangian submanifold. The signatures of these metrics are dictated by the relationship between vorticity and rate-of-strain, and their scalar curvatures can be interpreted in a physical context in terms of the accumulation of vorticity, strain, and their gradients. We show explicity, in the case of two-dimensional flows, how topological information can be derived from the Monge–Ampère geometry of the Lagrangian submanifold. We also demonstrate how certain solutions to the three-dimensional incompressible Navier–Stokes equations, such as Hill’s spherical vortex and an integrable case of Arnol’d–Beltrami–Childress flow, have symmetries that facilitate a formulation of these solutions from the perspective of (higher) symplectic reduction.

Memory-Efficient Interpolatory Projection Techniques for the Stabilization of Incompressible Navier-Stokes Flows

In this article, we are proposing an updated form of Krylov subspace-based interpolatory projection techniques for the stabilization of incompressible Navier-Stokes flows. In the proposed techniques, we utilize the reduced-order modelling approach implicitly, where reduced-order matrices need not be gathered explicitly. To estimate the aimed optimal feedback matrix, only the factored solution of the desired continuous-time algebraic Riccati equation (CARE) needs to be stored through the classical eigenvalue decomposition. The sparse structure of the target systems will remain invariant within the matrix-vector operations for generating the bases of the projector matrices through Krylov subspace techniques, where a cohesive projection scheme will be incorporated to ensure the potency of the projector matrices. So, the proposed techniques will be feasible for memory allocation and enhance the rapid convergence of the simulation. Analysis of the target systems’ transient characteristics, such as eigenvalues and step-responses, will be used to ascertain the competence and reliability of the proposed techniques. Necessary computation will be done numerically through MATLAB. Stabilization of the transient behaviors and minimization of the simulation time are the prime concerns in this work. Eventually, by comparing with the contemporary techniques the advancement of the proposed techniques will be confirmed.

Effects of body shape on hydrodynamic interactions in a dense diamond fish school

To examine the effect that body shape plays in the interactions between fish in a dense fish school, fish-like undulating foils are numerically studied in a high-density diamond school. Shape parameters of leading edge radius, boattail angle, and maximum thickness location along the body are independently varied to control the body shape. A traveling wave is prescribed to the body, and the flow around the school is solved using an immersed boundary method-based incompressible Navier–Stokes flow solver. Our findings indicate that body shape does play a significant role in the performance of the school and varies school efficiency and thrust productions by as much as 7% and 40%, respectively, and changes the efficiency in individual swimmers within the school by up to 25%. The leading edge radius drives the scale of the anterior body suction experienced throughout the school. A rounder leading edge generates more suction but consequently consumes more power. Changes to the location of the maximum thickness along the foil significantly impact the wall effect in the school by changing the shape of the “wall.” A maximum thickness that occurs at or in front of the closest point of interaction between the body and tail is favored. Finally, changes to the boattail angle manipulate the shape of the active channels in the school altering vortex–body interactions and the wall effect. A large boattail angle leads to a pinching that prohibits significant flow in the channels, reducing performance.

Existence and regularity of steady-state solutions of the Navier-Stokes equations arising from irregular data

We analyze the forced incompressible stationary Navier-Stokes flow in R+n, n>2. Existence of a unique solution satisfying a global integrability property measured in a scale of tent spaces is established for small data in homogeneous Sobolev space with s=−12 degree of smoothness. The velocity field is shown to be locally Hölder continuous while the pressure belongs to Llocp for any p∈(1,∞). Our approach is based on the analysis of the inhomogeneous Stokes system for which we derive a new solvability result involving Dirichlet data in Triebel-Lizorkin classes with negative amount of smoothness and is of independent interest.

Physics-informed neural networks with domain decomposition for the incompressible Navier–Stokes equations

Physics-informed neural network (PINN) has emerged as a promising approach for solving differential equations in recent years. However, their application to large-scale complex problems has faced challenges regarding accuracy and efficiency. To address these limitations, domain decomposition has gained popularity as an effective strategy. This paper studies a domain decomposition PINN method for solving incompressible Navier–Stokes equations. We assess the method's predicted accuracy, convergence, and the impact of different strategies on performance. In the domain decomposition PINN method, individual PINN is employed for each subdomain to compute local solutions, which are seamlessly connected by enforcing additional continuity conditions at the interfaces. To improve the method's performance, we investigate various continuity conditions at the interfaces and analyze their influence on the predictive accuracy and interface continuity. Furthermore, we introduce two approaches: the dynamic weight method and a novel neural network architecture incorporating attention mechanisms, both aimed at mitigating gradient pathologies commonly encountered in PINN methods. To demonstrate the effectiveness of the proposed method, we apply it to a range of forward and inverse problems involving diverse incompressible Navier–Stokes flow scenarios. This includes solving benchmark problems such as the two-dimensional (2D) Kovasznay flow, the three-dimensional (3D) Beltrami flow, the 2D lid-driven cavity flow, and the 2D cylinder wake. Additionally, we conduct 3D blood flow simulations for synthetic flow geometries and real blood vessels. The experimental results demonstrate the capability and versatility of the domain decomposition PINN method in accurately solving incompressible Navier–Stokes flow problems.

SVD-Krylov-based sparsity-preserving techniques to optimally stabilize the incompressible Navier–Stokes flows

SVD-Krylov-based sparsity-preserving techniques to optimally stabilize the incompressible Navier–Stokes flows

Sparsity-Preserving Two-Sided Iterative Algorithm for Riccati-Based Boundary Feedback Stabilization of the Incompressible Navier–Stokes Flow

In this paper, we explore the Riccati-based boundary feedback stabilization of the incompressible Navier–Stokes flow via the Krylov subspace techniques. Since the volume of data derived from the original models is gigantic, the feedback stabilization process through the Riccati equation is always infeasible. We apply aℋ2optimal model-order reduction scheme for reduced-order modeling, preserving the sparsity of the system. An extended form of the Krylov subspace-based two-sided iterative algorithm (TSIA) is implemented, where the computation of an equivalent Sylvester equation is included for minimizing the computation time and enhancing the stability of the reduced-order models with satisfying the Wilson conditions. Inverse projection approaches are applied to get the optimal feedback matrix from the reduced-order models. To validate the efficiency of the proposed techniques, transient behaviors of the target systems are observed incorporating the tabular and figurative comparisons with MATLAB simulations. Finally, to reveal the advancement of the proposed techniques, we compare our work with some existing works.

Attractors for the inhomogeneous incompressible Navier–Stokes flows

This paper is devoted to the evolution of Lions’s weak solutions to the inhomogeneous Navier–Stokes equations. After proving that the kinetic energy is eventually bounded, we obtain a weakly compact global attractor that all Lions’s weak solutions approach as time tends to infinity. Furthermore, the existence of attracting sets in strong topology is established for short trajectories satisfying an additional compactness condition on the density.

Dynamics for 2D incompressible Navier–Stokes flow coupled with time-dependent Darcy flow

In this paper, we use the method of evolutionary systems introduced by Cheskidov and Foias to describe the existence of global attractor for 2D incompressible Navier–Stokes flow coupled with time-dependent Darcy flow. Furthermore, stationary statistical solutions of this system are constructed from the global attractor.

Fluid dynamic shape optimization using self-adapting nonlinear extension operators with multigrid preconditioners

In this article we propose a scalable shape optimization algorithm which is tailored for large scale problems and geometries represented by hierarchically refined meshes. Weak scalability and grid independent convergence is achieved via a combination of multigrid schemes for the simulation of the PDEs and quasi Newton methods on the optimization side. For this purpose a self-adapting, nonlinear extension operator is proposed within the framework of the method of mappings. This operator is demonstrated to identify critical regions in the reference configuration where geometric singularities have to arise or vanish. Thereby the set of admissible transformations is adapted to the underlying shape optimization situation. The performance of the proposed method is demonstrated for the example of drag minimization of an obstacle within a stationary, incompressible Navier–Stokes flow.

Energy Dissipation and Nonpotential Effects in Wave Breaking

This paper presents a comparative numerical study of the energy dissipation process in the breaking of focused waves by using a potential flow model and a coupled potential/viscous flow model. An empirical eddy viscosity term is introduced to the fully-nonlinear potential (FNP) flow model to account for the breaking wave energy dissipation. The FNP model is further coupled with an incompressible two-phase NavierStokes (NS) flow solver through a one-way linkage to generate and propagate focused waves in the domain. Numerical absorbing regions are placed in front of the outlet boundaries to dampen wave reflection. The standalone FNP model and the coupled FNP+NS model are applied to deal with each scenario comparatively. This enables the accurate quantification and comparison of the energy loss of breaking waves calculated by the two numerical models. The velocity field is decomposed into potential component, which is reconstructed from the corresponding free surface elevation computed in the coupled model by using the weakly-nonlinear wave theory, and non-potential rotational component. Detailed analysis of the numerical results shows that: (1) energy loss is closely related to wave steepness, (2) mild rotational motion produced by a non-breaking wave is local in time with a short life-span, (3) strong non-potential motion triggered by wave breaking is not local in time but persists in the flow for dozens of or even many more wave periods.

Stabilization of exact coherent structures in two-dimensional turbulence using time-delayed feedback

Time-delayed feedback control, attributed to Pyragas (1992 Physics Letters 170(6) 421-428), is a method known to stabilise periodic orbits in low dimensional chaotic dynamical systems. A system of the form $\dot{\mathbf{x}}(t)=f(\mathbf{x})$ has an additional term $G(\mathbf{x}(t)-\mathbf{x}(t-T))$ introduced where $G$ is some `gain matrix' and $T$ a time delay. The form of the delay term is such that it will vanish for any orbit of period $T,$ therefore making it also an orbit of the uncontrolled system. This non-invasive feature makes the method attractive for stabilising exact coherent structures in fluid turbulence. Here we begin by validating the method for the basic flow in Kolmogorov flow; a two-dimensional incompressible Navier-Stokes flow with a sinusoidal body force. The linear predictions for stabilisation are well captured by direct numerical simulation. By applying an adaptive method to adjust the streamwise translation of the delay, a known travelling wave solution is able to be stabilised up to relatively high Reynolds number. We discover that the famous `odd-number' limitation of this time-delayed feedback method can be overcome in the fluid problem by using the symmetries of the system. This leads to the discovery of 8 additional exact coherent structures which can be stabilised with this approach. This means that certain unstable exact coherent structures can be obtained by simply time-stepping a modified set of equations, thus circumventing the usual convergence algorithms.

Immersed boundary finite element method for blood flow simulation

We present an efficient and accurate immersed boundary (IB) finite element (FE) method for numerically solving the incompressible Navier–Stokes equations. Particular emphasis is given to internal flows with complex geometries (blood flow in the vasculature system). IB methods are computationally costly for internal flows, mainly due to the large percentage of grid points that lie outside the flow domain. In this study, we apply a local refinement strategy, along with a domain reduction approach, in order to reduce the grid that covers the flow domain and increase the percentage of grid nodes that fall inside the flow domain. The proposed method utilizes an efficient and accurate FE solver with the incremental pressure correction scheme (IPCS), along with the boundary condition enforced IB method to numerically solve the transient, incompressible Navier–Stokes flow equations. We verify the accuracy of the numerical method using the analytical solution for Poiseuille flow in a cylinder. We further examine the accuracy and applicability of the proposed method by considering flow within complex geometries, such as blood flow in aneurysmal vessels and the aorta, flow configurations which would otherwise be extremely difficult to solve by most IB methods. Our method offers high accuracy, as demonstrated by the verification examples, and high efficiency, as demonstrated through the solution of blood flow within complex geometry on an off-the-shelf laptop computer.

Dynamics of diffusion flames in a very low strain rate flow field: from transient one-dimensional to stationary two-dimensional regime

The present work describes the transition of transient one-dimensional diffusion flame into a steady two-dimensional regime in a new flow field configuration. To that end, a cylindrical burner from which fuel is ejected radially and uniformly is positioned in the middle of two impinging flows. The chosen conditions are such that the strain rate is very low. The majority of the flame is located in a region of the flow field where spatial coordinates are scaled with the reciprocal of the square root of the strain rate, and the velocities are scaled with the square root of the strain rate. To simplify the model, a potential flow is assumed, with its results compared with those from a more detailed incompressible Navier–Stokes flow solution. The evolution of the flame is similar in both cases, which shows that the idealised potential flow describes well the flow field in such a geometry. Mixture fraction and excess enthalpy variables are employed to describe the infinitely fast chemical reaction, and therefore, fuel mass fraction, oxidiser mass fraction, and temperature fields. Results show that the initial flame displacement is controlled by the radial transport of fuel near the burner, where the impinging flows have a negligible influence. After that region, the flame is strongly influenced by the impinging flows where its acceleration is observed. Moreover, the proposed asymptotic solutions highlight the main transport mechanisms of reactants to the flame under different conditions and show the dependence of the flame on the chemical and flow field parameters. The stationary solution presents a diffusion flame with continuous geometric variation, from the counterflow to the coflow regime.