Construction of low dissipative high-order well-balanced filter schemes for non-equilibrium flows

The goal of this paper is to generalize the well-balanced approach for non-equilibrium flow studied by Wang et al. (2009) [29] to a class of low dissipative high-order shock-capturing filter schemes and to explore more advantages of well-balanced schemes in reacting flows. More general 1D and 2D reacting flow models and new examples of shock turbulence interactions are provided to demonstrate the advantage of well-balanced schemes. The class of filter schemes developed by Yee et al. (1999) [33], Sjögreen and Yee (2004) [27] and Yee and Sjögreen (2007) [38] consist of two steps, a full time step of spatially high-order non-dissipative base scheme and an adaptive non-linear filter containing shock-capturing dissipation. A good property of the filter scheme is that the base scheme and the filter are stand-alone modules in designing. Therefore, the idea of designing a well-balanced filter scheme is straightforward, i.e. choosing a well-balanced base scheme with a well-balanced filter (both with high-order accuracy). A typical class of these schemes shown in this paper is the high-order central difference schemes/predictor–corrector (PC) schemes with a high-order well-balanced WENO filter. The new filter scheme with the well-balanced property will gather the features of both filter methods and well-balanced properties: it can preserve certain steady-state solutions exactly; it is able to capture small perturbations, e.g. turbulence fluctuations; and it adaptively controls numerical dissipation. Thus it shows high accuracy, efficiency and stability in shock/turbulence interactions. Numerical examples containing 1D and 2D smooth problems, 1D stationary contact discontinuity problem and 1D turbulence/shock interactions are included to verify the improved accuracy, in addition to the well-balanced behavior.     

Assessment of localized artificial diffusivity scheme for large-eddy simulation of compressible turbulent flows

The localized artificial diffusivity method is investigated in the context of large-eddy simulation of compressible turbulent flows. The performance of different artificial bulk viscosity models are evaluated through detailed results from the evolution of decaying compressible isotropic turbulence with eddy shocklets and supersonic turbulent boundary layer. Effects of subgrid-scale (SGS) models and implicit time-integration scheme/time-step size are also investigated within the framework of the numerical scheme used. The use of a shock sensor along with artificial bulk viscosity significantly improves the scheme for simulating turbulent flows involving shocks while retaining the shock-capturing capability. The proposed combination of Ducros-type sensor with a negative dilatation sensor removes unnecessary bulk viscosity within expansion and weakly compressible turbulence regions without shocks and allows it to localize near the shocks. It also eliminates the need for a wall-damping function for the bulk viscosity while simulating wall-bounded turbulent flows. For the numerical schemes used, better results are obtained without adding an explicit SGS model than with SGS model at moderate Reynolds number. Inclusion of a SGS model in addition to the low-pass filtering and artificial bulk viscosity results in additional damping of the resolved turbulence. However, investigations at higher Reynolds numbers suggest the need for an explicit SGS model. The flow statistics obtained using the second-order implicit time-integration scheme with three sub-iterations closely agrees with the explicit scheme if the maximum Courant–Friedrichs–Lewy is kept near unity.     

Assessment of high-resolution methods for numerical simulations of compressible turbulence with shock waves

Flows in which shock waves and turbulence are present and interact dynamically occur in a wide range of applications, including inertial confinement fusion, supernovae explosion, and scramjet propulsion. Accurate simulations of such problems are challenging because of the contradictory requirements of numerical methods used to simulate turbulence, which must minimize any numerical dissipation that would otherwise overwhelm the small scales, and shock-capturing schemes, which introduce numerical dissipation to stabilize the solution. The objective of the present work is to evaluate the performance of several numerical methods capable of simultaneously handling turbulence and shock waves. A comprehensive range of high-resolution methods (WENO, hybrid WENO/central difference, artificial diffusivity, adaptive characteristic-based filter, and shock fitting) and suite of test cases (Taylor–Green vortex, Shu–Osher problem, shock-vorticity/entropy wave interaction, Noh problem, compressible isotropic turbulence) relevant to problems with shocks and turbulence are considered. The results indicate that the WENO methods provide sharp shock profiles, but overwhelm the physical dissipation. The hybrid method is minimally dissipative and leads to sharp shocks and well-resolved broadband turbulence, but relies on an appropriate shock sensor. Artificial diffusivity methods in which the artificial bulk viscosity is based on the magnitude of the strain-rate tensor resolve vortical structures well but damp dilatational modes in compressible turbulence; dilatation-based artificial bulk viscosity methods significantly improve this behavior. For well-defined shocks, the shock fitting approach yields good results.     

A Straightforward $hp$-Adaptivity Strategy for Shock-Capturing with High-Order Discontinuous Galerkin Methods

In this paper, high-order Discontinuous Galerkin (DG) method is used to solve the two-dimensional Euler equations. A shock-capturing method based on the artificial viscosity technique is employed to handle physical discontinuities. Numerical tests show that the shocks can be captured within one element even on very coarse grids. The thickness of the shocks is dominated by the local mesh size and the local order of the basis functions. In order to obtain better shock resolution, a straightforward $hp$-adaptivity strategy is introduced, which is based on the high-order contribution calculated using hierarchical basis. Numerical results indicate that the $hp$-adaptivity method is easy to implement and better shock resolution can be obtained with smaller local mesh size and higher local order.     

On high-order shock-fitting and front-tracking schemes for numerical simulation of shock–disturbance interactions

High-order methods that can resolve interactions of flow-disturbances with shock waves are critical for reliable numerical simulation of shock wave and turbulence interaction. Such problems are not well understood due to the limitations of numerical methods. Most of the popular shock-capturing methods are only first-order accurate at the shock and may incur spurious numerical oscillations near the shock. Shock-fitting algorithms have been proposed as an alternative which can achieve uniform high-order accuracy and can avoid possible spurious oscillations incurred in shock-capturing methods by treating shocks as sharp interfaces. We explore two ways for shock-fitting: conventional moving grid set-up and a new fixed grid set-up with front tracking. In the conventional shock-fitting method, a moving grid is fitted to the shock whereas in the newly developed fixed grid set-up the shock front is tracked using Lagrangian points and is free to move across the underlying fixed grid. Different implementations of shock-fitting methods have been published in the literature. However, uniform high-order accuracy of various shock-fitting methods has not been systematically established. In this paper, we carry out a rigorous grid-convergence analysis on different variations of shock-fitting methods with both moving and fixed grids. These shock-fitting methods consist of different combinations of numerical methods for computing flow away from the shock and those for computing the shock movement. Specifically, we consider fifth-order upwind finite-difference scheme and shock-capturing WENO schemes with conventional shock-fitting and show that a fifth-order convergence is indeed achieved for a canonical one-dimensional shock-entropy wave interaction problem. We also show that the method of finding shock velocity from one characteristic relation and Rankine–Hugoniot jump condition performs better than the other methods of computing shock velocities. A high-order front-tracking implementation of shock-fitting is also presented in this paper and nominal rate of convergence is shown. The front-tracking results are validated by comparing to results from the conventional shock-fitting method and a linear-interaction analysis for a two-dimensional shock disturbance interaction problem.     

Analysis of a second-order polarization-independent filter made of a single ring resonator and a Sagnac interferometer

A novel second-order polarization-independent filter made of a single ring resonator and a Sagnac interferometer (SRRSI) is proposed, and its filtering characteristics are investigated. By using birefringence in waveguide, a single ring resonator can be used to synthesize a filter with second-order response. Analytical formulas are derived for characteristics of the SRRSI varied with waveguide parameters, such as the coupling coefficient; and the critical condition of a second-order Butterworth filter is given. The influence of loss in the ring resonator is also analyzed.     

An interface capturing method for the simulation of multi-phase compressible flows

A novel finite-volume interface (contact) capturing method is presented for simulation of multi-component compressible flows with high density ratios and strong shocks. In addition, the materials on the two sides of interfaces can have significantly different equations of state. Material boundaries are identified through an interface function, which is solved in concert with the governing equations on the same mesh. For long simulations, the method relies on an interface compression technique that constrains the thickness of the diffused interface to a few grid cells throughout the simulation. This is done in the spirit of shock-capturing schemes, for which numerical dissipation effectively preserves a sharp but mesh-representable shock profile. For contact capturing, the formulation is modified so that interface representations remain sharp like captured shocks, countering their tendency to diffuse via the same numerical diffusion needed for shock-capturing. Special techniques for accurate and robust computation of interface normals and derivatives of the interface function are developed. The interface compression method is coupled to a shock-capturing compressible flow solver in a way that avoids the spurious oscillations that typically develop at material boundaries. Convergence to weak solutions of the governing equations is proved for the new contact capturing approach. Comparisons with exact Riemann problems for model one-dimensional multi-material flows show that the interface compression technique is accurate. The method employs Cartesian product stencils and, therefore, there is no inherent obstacles in multiple dimensions. Examples of two- and three-dimensional flows are also presented, including a demonstration with significantly disparate equations of state: a shock induced collapse of three-dimensional van der Waal’s bubbles (air) in a stiffened equation of state liquid (water) adjacent to a Mie-Grüneisen equation of state wall (copper).     

Compact Accurately Boundary-Adjusting high-REsolution Technique for fluid dynamics

A novel high-resolution numerical method is presented for one-dimensional hyperbolic problems based on the extension of the original Upwind Leapfrog scheme to quasi-linear conservation laws. The method is second-order accurate on non-uniform grids in space and time, has a very small dispersion error and computational stencil defined within one space–time cell. For shock-capturing, the scheme is equipped with a conservative non-linear correction procedure which is directly based on the maximum principle. Plentiful numerical examples are provided for linear advection, quasi-linear scalar hyperbolic conservation laws and gas dynamics and comparisons with other computational methods in the literature are discussed.     

Shock-Capturing and Front-Tracking Methods for Granular Avalanches

Shock formations are observed in granular avalanches when supercritical flow merges into a region of subcritical flow. In this paper we employ a shock-capturing numerical scheme for the one-dimensional Savage–Hutter theory of granular flow to describe this phenomenon. A Lagrangian moving mesh scheme applied to the nonconservative form of the equations reproduces smooth solutions of these free boundary problems very well, but fails when shocks are formed. A nonoscillatory central (NOC) difference scheme with TVD limiter or WENO cell reconstruction for the conservative equations is therefore introduced. For the avalanche free boundary problems it must be combined with a front-tracking method, developed here, to properly describe the margin evolution. It is found that this NOC scheme combined with the front-tracking module reproduces both the shock wave and the smooth solution accurately. A piecewise quadratic WENO reconstruction improves the smoothness of the solution near local extrema. The schemes are checked against exact solutions for (1) an upward moving shock wave, (2) the motion of a parabolic cap down an inclined plane, and (3) the motion of a parabolic cap down a curved slope ending in a flat run-out region, where a shock is formed as the avalanche comes to a halt.     

Particle acceleration at astrophysical shocks: A theory of cosmic ray origin

The theory of first order Fermi acceleration at collisionless astrophysical shock fronts is reviewed. Observations suggest that shock waves in different astrophysical environments accelerate cosmic rays efficiently. In the first order process, high energy particles diffuse through Alfvén waves that scatter them and couple them to the background plasma. These particles gain energy, on the average, every time they cross the schock front and bounce off approaching scattering centers. Calculations demonstrate that the distribution function transmitted by a plane shock is roughly a power law in momentum with slope similar to that inferred in galactic cosmic ray sources. The generation of the scattering Alfvén waves by the streaming cosmic rays is described and it is argued that the wave amplitude is probably non-linear within sufficiently strong astrophysical shocks. Hydromagnetic scattering can operate on the thermal particles as well, possibly establishing the shock structure. This suggests a model of strong collisionless shocks in which high energy particles are inevitably produced very efficiently. Observable consequences of this model, together with its limitations and some alternatives, are described. Cosmic ray origin and astrophysical shocks can no longer be considered separately.     

Two-phase flow model for energetic proton beam induced pressure waves in mercury target systems in the planned European Spallation Source

Two-phase flow calculations are presented to investigate the thermo-hydraulical effects of the interaction between 2 ms long 1.3 GeV proton pulses with a closed mercury loop which can be considered as a model system of the target of the planned European Spallation Source (ESS) facility. The two-fluid model consists of six first-order partial differential equations that present one dimensional mass, momentum and energy balances for mercury vapor and liquid phases are capable to describe quick transients like cavitation effects or shock waves. The absorption of the proton beam is represented as instantaneous heat source in the energy balance equations. Densities and internal energies of the mercury liquid-vapor system is calculated from the van der Waals equation, but general method how to obtain such properties using arbitrary equation of state is also presented. A second order accurate high-resolution shock-capturing numerical scheme is applied with different kind of limiters in the numerical calculations. Our analysis show that even 75 degree temperature heat shocks cannot cause considerable cavitation effects in mercury.     

铁-铍介质界面上激波折射现象的数值研究

用数值方法研究铁-铍介质界面上的激波折射现象.运用激波极曲线理论分析不同强度的激波从正规折射过渡到非正规折射的临界角变化.运用一个具有二阶精度和波传播性质的激波捕捉法,数值求解激波折射运动的流体力学方程组.对正规折射,数值结果与激波极曲线理论一致;对非正规折射,不同强度的激波大都存在前驱的折射激波,并且入射激波的强度不同、入射角度不同,激波折射的图像也不同.     

A shock-detecting sensor for filtering of high-order compact finite difference schemes

A new shock-detecting sensor for properly switching between a second-order and a higher-order filter is developed and assessed. The sensor is designed based on an order analysis. The nonlinear filter with the proposed sensor ensures damping of the high-frequency waves in smooth regions and at the same time removes the Gibbs oscillations around the discontinuities when using high-order compact finite difference schemes. In addition, a suitable scaling is proposed to have dissipation proportional to the shock strength and also to minimize the effects of the second-order filter on the very small scales. Several numerical experiments are carried out and the accuracy of the nonlinear filter with the proposed sensor is examined. In addition, some comparisons with other filters and sensors are made.     

Shock interaction computations on unstructured,two-dimensional grids using a shock-fitting technique

A new shock-fitting technique has been recently proposed and implemented by the authors in conjunction with an unstructured shock-capturing solver. In the present paper, the attention is addressed towards the computation of shock–shock and shock–wall interactions by means of this novel computational technique.     

Vocal tract resonances tracking by auxiliary vector particle filters

An auxiliary vector particle filter was proposed to present the vocal tract resonances (VTRs) tracking.It uses particle filter based on a version of state-space model that describes the characteristics of speech signal.The speech model consists of a target-guided dynamic function and a non-linear prediction mapping from resonance frequencies and bandwidths to LPC cepstra(LPCC).There are two characteristics in the proposed method.First,particle filtering technique is put forth to solve the non-linear problem of speech model.Second,an auxiliary vector,embedded in the state function of speech model,is applied to incorporate the most current observations and to generate the proposal distribution of particle filter.The experimental results show that this method is able to track the VTRs of continuous speech utterance efficiently with a small number of particles and able to solve the problem of spurious peaks and merging peaks.     

A Multi-Speed Compressible Lattice-Boltzmann Model

An extension of the lattice Boltzmann BGK method to compressible flows is presented that combines three novel additions: (1) particles move density and energy weights in multiple velocity bins (11 for 1-D flow) to nearby cell centers. (2) the equilibrium distribution remains an unexpanded Maxwellian; and (3) transport and relaxation to equilibrium are performed implicitly at each node. These advances allow for the parallel modeling of high Mach number shocks and high Reynolds number flows, while avoiding advective numerical diffusion, the need for Riemann solvers, and non-linear limiters. A 1D shock tube application is shown. Generalization to higher dimensions and multi-materials are discussed.     

辅助变量粒子滤波的声道共振特性轨迹跟踪

提出了一种基于辅助变量粒子滤波技术的连续语音声道共振特性(VTRs)轨迹跟踪方法。该方法基于描述语音信号特征的状态空间模型,采用粒子滤波技术跟踪VTRs的轨迹。语音模型由具有目标导向特性的动态方程和VTRs至倒谱系数(LPCC)的非线性映射构成。该方法有两个特点:首先,采用粒子滤波技术来处理语音模型的非线性问题;其次,在语音模型的状态方程中嵌入辅助变量用于标示VTRs在频域中的分布信息,并为粒子滤波过程中的粒子抽取提供目标导向。实验结果表明,该方法只需少量粒子即可正确跟踪连续语音的VTRs轨迹,而且可以在跟踪过程中避免虚假峰和合并峰的干扰。     

Non-linear filtering of ultrasonic signals using neural networks

Structure noise from inhomogeneous micro-structures makes the detection of flaws present in highly scattering materials difficult. Several techniques have been applied to improve the signal-to-noise ratio (SNR) in order to make flaw detection easier. Linear filtering does not provide good results because both structure noise and flaw signal concentrate energy in the same frequency band. Non-linear filtering can be used to reduce the structure noise of ultrasonic signals. Therefore, neural networks are applied in this work for this purpose. In order to use neural networks for non-linear filtering, dynamic structures must be applied. The easiest way to implement a neural network with the capability of processing temporal patterns is to consider them spatial ones, applying the signal into a tapped delay line of finite extension, that is the input of a static neural network (for example, a multi-layer perceptron). In this work, a dynamic neural network has been built to filter ultrasonic signals with structure noise, and has been trained with the real-time back-propagation algorithm, using as inputs 3000 synthetic ultrasonic signals of 896 samples each. Target signals for training are the same as the ones used as inputs but without noise. The neural network is trained in order to generate as output the target signal when the noisy input one is applied. For testing the performance of the non-linear filter, a new set of 500 noisy signals has been used. The SNR improvement is about 6 dB average. The results show that this non-linear filtering method is quite useful as pre-processing stage in flaw detection systems.     

Filtering skill for turbulent signals for a suite of nonlinear and linear extended Kalman filters

The filtering skill for turbulent signals from nature is often limited by errors due to utilizing an imperfect forecast model. In particular, real-time filtering and prediction when very limited or no a posteriori analysis is possible (e.g. spread of pollutants, storm surges, tsunami detection, etc.) introduces a number of additional challenges to the problem. Here, a suite of filters implementing stochastic parameter estimation for mitigating model error through additive and multiplicative bias correction is examined on a nonlinear, exactly solvable, stochastic test model mimicking turbulent signals in regimes ranging from configurations with strongly intermittent, transient instabilities associated with positive finite-time Lyapunov exponents to laminar behavior. Stochastic Parameterization Extended Kalman Filter (SPEKF), used as a benchmark here, involves exact formulas for propagating the mean and covariance of the augmented forecast model including the unresolved parameters. The remaining filters use the same nonlinear forecast model but they introduce model error through different moment closure approximations and/or linear tangent approximation used for computing the second-order statistics of the augmented stochastic forecast model. A comprehensive study of filter performance is carried out in the presence of various moment closure errors which are enhanced by additional model errors due to incorrect parameters inducing additive and multiplicative stochastic biases. The estimation skill of the unresolved stochastic parameters is also discussed and it is shown that the linear tangent filter, despite its popularity, is completely unreliable in many turbulent regimes for both parameter estimation and filtering; moreover, regimes of filter divergence for the linear tangent filter are identified. The results presented here provide useful guidelines for filtering turbulent, high-dimensional, spatially extended systems with more general model errors, as well as for designing more skillful methods for superparameterization of unresolved intermittent processes in complex multi-scale models. They also provide unambiguous benchmarks for the capabilities of linear and nonlinear extended Kalman filters using incorrect statistics on an exactly solvable test bed with rich and realistic dynamics.     

Three-dimensional reactive shock bifurcations

We model interactions of a premixed flame with incident and reflected shocks in a rectangular shock tube using three-dimensional (3D) reactive Navier–Stokes numerical simulations. Shock-flame interactions occur in the presence of boundary layers that cause the reflected shock to bifurcate and form a reactive shock bifurcation (RSB), which contains a flame in the recirculation zone behind the oblique shock. The recirculation zone acts as a flame holder thus attaching the flame to the shock in the vicinity of the wall, and providing a mechanism for a detonationless supersonic flame spread. The accelerated burning induced by an RSB, and Mach stems that may result from RSB–RSB interactions, promote hot-spot formation, and eventually accelerate deflagration-to-detonation transition. Schlieren-type images generated from the simulation results show that the 3D structure of an RSB may not always be easily recognized in experiments if the RSB is attached to the surface of the observation window. The main 3D effect observed in the simulations is caused by the presence of the second no-slip wall in a 3D rectangular channel. Two RSBs that form at adjacent walls interact with each other and produce an oblique Mach stem between two oblique shocks. The oblique Mach stems then interacts with a central Mach stem that forms near symmetry plane, and this interaction creates a hot-spot that leads to a detonation initiation.