High-resolution Finite Difference Schemes Research Articles

High resolution finite difference schemes for a size structured coagulation-fragmentation model in the space of radon measures.

In this paper, we develop explicit and semi-implicit second-order high-resolution finite difference schemes for a structured coagulation-fragmentation model formulated on the space of Radon measures. We prove the convergence of each of the two schemes to the unique weak solution of the model. We perform numerical simulations to demonstrate that the second order accuracy in the Bounded-Lipschitz norm is achieved by both schemes.

A hybrid lattice Boltzmann model for simulating viscoelastic instabilities

Low-Reynolds-number viscoelastic fluids exhibit a chaotic behaviour linked to the growth of elastic instabilities when the polymer molecules are stretched excessively. Despite offering various exciting benefits, the numerical difficulties associated with simulating this complex regime have left a lot still unexplored regarding the transition and onset of these low-Reynolds number viscoelastic instabilities. Recently, methods based on the lattice Boltzmann method (LBM) have emerged as a viable numerical tool for studying the behaviour of viscoelastic fluids largely due to the simplicity, adaptability, and intrinsic parallel features of LBM. However, extensive memory requirements and the inability to preserve numerical stability have restricted previous attempts from simulating viscoelastic instabilities. In this work, a hybrid lattice Boltzmann model is applied to simulate low-Reynolds number viscoelastic instabilities. In this approach, the hydrodynamic field is solved using a lattice Boltzmann model, whereas the polymer field is solved separately using a high-resolution finite difference scheme with the logarithmic Cholesky decomposition. The model is first validated for steady viscoelastic flows using the four-roll mill case, where the results are compared against the analytical solution and previous numerical studies. The model is then applied to simulate a chaotic low-Reynolds number viscoelastic instability by initialising a small perturbation to the isotropic polymer stress. The results at different Weissenberg numbers experience complex flow dynamics at sufficiently long time durations, which transition from quasi-periodic to periodic to aperiodic states with increasing elastic effects. The quasi-periodic regime was found to experience the fastest transition into the transient regime, with the transitioning time being delayed as the Weissenberg number increased. This work proves the capability of applying a coupled lattice Boltzmann method for exploring the complex flow behaviour of elastic instabilities, allowing the potential to explore more challenging and practical viscoelastic instability cases.

Mathematical modeling and computation of a dam–reservoir system balancing environmental management and hydropower generation

We propose a simple but non-trivial mathematical model of a dam–reservoir system balancing between the downstream environmental management and hydropower production. We consider a stochastic control problem of the system to find its balanced operation policy. We point out that finding the optimal policy reduces to solving a nonlinear degenerate elliptic equation. This equation is exactly solvable in a reduced case without hydropower production, while it requires numerical computation in more complicated cases. A demonstrative application example with a high-resolution finite difference scheme is finally presented.

Direct numerical simulation of a fully developed compressible wall turbulence over a wavy wall

ABSTRACTCompressible turbulent channel flow over a wavy surface is investigated by direct numerical simulations using high-resolution finite difference schemes. The Reynolds number considered in the present paper is 3380 based on the bulk velocity, the channel half-width and the kinetic viscosity at the wall. Four test cases are simulated and analysed at Mam = 0.33, 0.8, 1.2, 1.5 based on the bulk velocity and the speed of sound at the wall. We mainly focus on the curvature and the Mach number effects on the compressible turbulent flows. Numerical results show that although the wavy wall has effects on the mean and fluctuation quantities, log law still exists in the distribution of the wave-averaged streamwise velocity if the roughness effects are taken into consideration in the scaling of it. Near-wall streaks are broken by the wavy surface and near-wall quasi-streamwise vortices mostly begin at the upslope of the wave and pass over the crest of it. The wavy wall makes the turbulence more active and the flow easier to be blended. From the viewpoint of turbulent kinetic budgets, curvature effects strengthen both the diffusion terms and the dissipation terms. At the same time, they change the properties of the compressibility-related terms and promote more inner energy transferring into turbulent kinetic energy. As the Mach number increases, the reattachment of the mean flow is delayed, which indicates the mean separation bubble becomes larger. Concerning the near-wall coherent structures, the vortices are more sparsely distributed with the increasing of the Mach number. For the supersonic cases, shock waves appear. Though they have little effects on the mean turbulent quantities, they change the structures of the flow fields and induce local separations at the upper wall of the channel.

A high resolution finite difference method for a model of structured susceptible‐infected populations coupled with the environment

We develop a general model describing a structured susceptible‐infected (SI) population coupled with the environment. This model applies to problems arising in ecology, epidemiology, and cell biology. The model consists of a system of quasilinear hyperbolic partial differential equations coupled with a system of nonlinear ordinary differential equations that represents the environment. We develop a second‐order high resolution finite difference scheme to numerically solve the model. Convergence of this scheme to a weak solution with bounded total variation is proved. Numerical simulations are provided to demonstrate the high‐resolution property of the scheme and an application to a multi‐host wildlife disease model is explored.© 2017 Wiley Periodicals, Inc. Numer Methods Partial Differential Eq 33: 1420–1458, 2017

Оne-dimensional gas dynamics problems and their solution based on high-resolution finite difference schemes

One-dimensional unsteady gas dynamics problems are revealing tests for the accuracy estimation of numerical solution with respect to simulation of supersonic flows of inviscid compressible gas. Numerical solution of Euler equations describing flows of inviscid compressible gas and conceding continuous and discontinuous solutions is considered. Discretization of Euler equations is based on finite volume method and WENO finite difference schemes. The numerical solutions computed are compared with the exact solution of Riemann problem. Monotonic correction of derivatives makes possible avoiding new extremes and ensures monotonicity of the numerical solution near the discontinuity, but it leads to the smoothness of the existing minimums and maximums and to the accuracy loss. Calculations with the use of WENO schemes give the possibility for obtaining accurate and monotonic solution with the presence of weak and strong gas dynamical discontinuities.

A second-order high-resolution finite difference scheme for a size-structured model for the spread of Mycobacterium marinum

We present a second-order high-resolution finite difference scheme to approximate the solution of a mathematical model of the transmission dynamics of Mycobacterium marinum (Mm) in an aquatic environment. This work extends the numerical theory and continues the preliminary studies on the model first developed in Ackleh et al. [Structured models for the spread of Mycobacterium marinum: foundations for a numerical approximation scheme, Math. Biosci. Eng. 11 (2014), pp. 679–721]. Numerical simulations demonstrating the accuracy of the method are presented, and we compare this scheme to the first-order scheme developed in Ackleh et al. [Structured models for the spread of Mycobacterium marinum: foundations for a numerical approximation scheme, Math. Biosci. Eng. 11 (2014), pp. 679–721] to show that the first-order method requires significantly more computational time to provide solutions with a similar accuracy. We also demonstrated that the model can be a tool to understand surprising or nonintuitive phenomena regarding competitive advantage in the context of biologically realistic growth, birth and death rates.

Developing Hybrid cell-edge and cell-node Dissipative Compact Scheme for Complex Geometry Flows

Developing high resolution finite difference scheme and enabling the use of this scheme on complex geometry are the aims of this study. High resolution has been achieved by Dissipative Compact Schemes (DCS), however, according to the recent research, applications of DCS on complex geometry may have serious problem for that the Geometric Conservation Law (GCL) is not satisfied, and this may cause numerical instability. To cope with this problem, a new scheme named Hybrid cell-edge and cell-node Dissipative Compact Scheme (HDCS) has been formulated. The formulation of the HDCS contains two steps. First, a new central compact scheme is formulated for the purpose of conveniently fulfilling the GCL, and then dissipation is added on the central scheme by high-order dissipative interpolation of cell-edge variables. The solutions of Euler and Navier-Stokes equations show that the HDCS can be applied successfully on complex geometry, while the DCS may suffer numerical instabilities. Moreover, high resolution of the HDCS may be observed in the test of scattering of acoustic waves by multiple cylinders.

Fast acoustic streaming in standing waves: Generation of an additional outer streaming cell

Rayleigh streaming in a cylindrical acoustic standing waveguide is studied both experimentally and numerically for nonlinear Reynolds numbers from 1 to 30 [Re(NL)=(U0/c0)(2)(R/δν)(2), with U0 the acoustic velocity amplitude at the velocity antinode, c0 the speed of sound, R the tube radius, and δν the acoustic boundary layer thickness]. Streaming velocity is measured by means of laser Doppler velocimetry in a cylindrical resonator filled with air at atmospheric pressure at high intensity sound levels. The compressible Navier-Stokes equations are solved numerically with high resolution finite difference schemes. The resonator is excited by shaking it along the axis at imposed frequency. Results of measurements and of numerical calculation are compared with results given in the literature and with each other. As expected, the axial streaming velocity measured and calculated agrees reasonably well with the slow streaming theory for small ReNL but deviates significantly from such predictions for fast streaming (ReNL>1). Both experimental and numerical results show that when ReNL is increased, the center of the outer streaming cells are pushed toward the acoustic velocity nodes until counter-rotating additional vortices are generated near the acoustic velocity antinodes.

A second-order high resolution finite difference scheme for a structured erythropoiesis model subject to malaria infection

We develop a second-order high-resolution finite difference scheme to approximate the solution of a mathematical model describing the within-host dynamics of malaria infection. The model consists of two nonlinear partial differential equations coupled with three nonlinear ordinary differential equations. Convergence of the numerical method to the unique weak solution with bounded total variation is proved. Numerical simulations demonstrating the achievement of the designed accuracy are presented.

Dynamic Modeling of Phase Crossings in Two-Phase Flow

AbstractTwo-phase flow and heat transfer, such as boiling and condensing flows, are complicated physical phenomena that generally prohibit an exact solution and even pose severe challenges for numerical approaches. If numerical solution time is also an issue the challenge increases even further. We present here a numerical implementation and novel study of a fully distributed dynamic one-dimensional model of two-phase flow in a tube, including pressure drop, heat transfer, and variations in tube cross-section. The model is based on a homogeneous formulation of the governing equations, discretized by a high resolution finite difference scheme due to Kurganov and Tadmore.The homogeneous formulation requires a set of thermodynamic relations to cover the entire range from liquid to gas state. This leads a number of numerical challenges since these relations introduce discontinuities in the derivative of the variables and are usually very slow to evaluate. To overcome these challenges, we use an interpolation scheme with local refinement.The simulations show that the method handles crossing of the saturation lines for both liquid to two-phase and two-phase to gas regions. Furthermore, a novel result obtained in this work, the method is stable towards dynamic transitions of the inlet/outlet boundaries across the saturation lines. Results for these cases are presented along with a numerical demonstration of conservation of mass under dynamically varying boundary conditions. Finally we present results for the stability of the code in a case of a tube with a narrow section.

Self-similar solutions for the diffraction of weak shocks

We numerically solve a problem for the unsteady transonic small disturbance equations that describes the diffraction of a weak shock into an expansion wave. In the context of a shock moving into a semi-infinite wall, this problem describes the interaction between the reflected part of the shock and the part that is transmitted beyond the wall. We formulate the equations in self-similar variables, and obtain numerical solutions using high resolution finite difference schemes. Our solutions appear to show that the shock dies out at the sonic line, rather than forms at an interior point of the supersonic region.

A bivariate population balance model for the microbial production of poly(3-hydroxybutyrate)

A bivariate population balance model is developed for describing the dynamic evolution of a PHB-producing microbial batch culture. The population balance model is solved with three different numerical methods namely, a continuous finite element method, a discrete-continuous finite element method and a discretized method with different high-resolution finite difference schemes for the calculation of the cell flux term. The various numerical methods are evaluated in terms of accuracy and computational requirements by a direct comparison of the leading moments of the bivariate distribution (i.e., M 10 and M 01) to the respective quantities calculated by a simple homogeneous model for the case of zero product inhibition. Furthermore, the results obtained from the solution of the 2-D population balance model are compared with the respective results of a 1-D PBM for the case of product inhibition of the residual biomass growth rate. It is shown that the 1-D PBM cannot accurately describe the dynamics of the microbial production system when the intracellularly accumulated PHB inhibits the residual biomass growth rate and, hence, a bivariate population balance modeling approach is required to describe the dynamic behavior of the microbial culture.

High-resolution finite difference schemes using curvilinear coordinate grids for DNS of compressible turbulent flow over wavy walls

Direct numerical simulations of compressible turbulent flow over wavy wall geometries have been carried out by solving N–S equations on general curvilinear coordinates. A 6th order WENO scheme with minimized dispersion and controllable dissipation is employed to compute the inviscid fluxes, a 4th order central difference scheme is applied to compute the viscous fluxes, and a 6th order conservative compact scheme is used for computing the geometrical metrics. An implicit LU-SGS method is used for time integration to improve computational efficiency over the explicit schemes such as the Runge–Kutta approach. The validity and applicability of the present algorithm is confirmed by comparing our results with laboratory experimental measurements and DNS results in the literature.

A Probability Density Approach to Modeling Local Control of Calcium-Induced Calcium Release in Cardiac Myocytes

We present a probability density approach to modeling localized Ca 2+ influx via L-type Ca 2+ channels and Ca 2+-induced Ca 2+ release mediated by clusters of ryanodine receptors during excitation-contraction coupling in cardiac myocytes. Coupled advection-reaction equations are derived relating the time-dependent probability density of subsarcolemmal subspace and junctional sarcoplasmic reticulum [Ca 2+] conditioned on “Ca 2+ release unit” state. When these equations are solved numerically using a high-resolution finite difference scheme and the resulting probability densities are coupled to ordinary differential equations for the bulk myoplasmic and sarcoplasmic reticulum [Ca 2+], a realistic but minimal model of cardiac excitation-contraction coupling is produced. Modeling Ca 2+ release unit activity using this probability density approach avoids the computationally demanding task of resolving spatial aspects of global Ca 2+ signaling, while accurately representing heterogeneous local Ca 2+ signals in a population of diadic subspaces and junctional sarcoplasmic reticulum depletion domains. The probability density approach is validated for a physiologically realistic number of Ca 2+ release units and benchmarked for computational efficiency by comparison to traditional Monte Carlo simulations. In simulated voltage-clamp protocols, both the probability density and Monte Carlo approaches to modeling local control of excitation-contraction coupling produce high-gain Ca 2+ release that is graded with changes in membrane potential, a phenomenon not exhibited by so-called “common pool” models. However, a probability density calculation can be significantly faster than the corresponding Monte Carlo simulation, especially when cellular parameters are such that diadic subspace [Ca 2+] is in quasistatic equilibrium with junctional sarcoplasmic reticulum [Ca 2+] and, consequently, univariate rather than multivariate probability densities may be employed.

Rarefied Flow Computations Using Nonlinear Model Boltzmann Equations

High resolution finite difference schemes for solving the nonlinear model Boltzmann equations are presented for the computations of rarefied gas flows. The discrete ordinate method is first applied to remove the velocity space dependency of the distribution function which renders the model Boltzmann equation in phase space to a set of hyperbolic conservation laws with source terms in physical space. Then a high order essentially nonoscillatory method due to Harten et al. (J. Comput. Phys. 71, 231, 1987) is adapted and extended to solve them. Explicit methods using operator splitting and implicit methods using the lower-upper factorization are described to treat multidimensional problems. The methods are tested for both steady and unsteady rarefied gas flows to illustrate its potential use. The computed results using model Boltzmann equations are found to compare well both with those using the direct simulation Monte Carlo results in the transitional regime flows and those with the continuum Navier-Stokes calculations in near continuum regime flows.

A High-Resolution Finite Difference Scheme for Supersonic Wet-Stream Flows.

A high-resolution finite difference method to simulate the gas-liquid two-phase flow is proposed taking account of the apparent compressibility which characterizes high -speed two -phase flow phenomena. The speed of sound and the total energy for the gas-liquid two-phase fluid flows are re-estimated based on the assumption of homogeneous and the thermodynamic relations. Here, the homogeneous two-phase flow model assuming a local equilibrium is adopted. The basic equations for the two-phase mixture flow similar to those of the ordinary single-phase compressible gas flow are solved using the upwind finite difference scheme based on the theory of characteristics. To elucidate the applicability of the present scheme, some numerical calculations of wet-steam nozzle flows with condensation were implemented. The calculated results are in good agreement with the existing experimental data.