Stabilization of explicit methods for hyperbolic partial differential equations

It is well known that explicit methods are subject to a restriction on the time step. This restriction is a drawback if the variation of the solution in time is so small that accuracy considerations would allow a larger time step. In this case, implicit methods are more appropriate because they do allow large time steps. However, in general, they require more storage and are more difficult to implement than explicit methods. In this paper we propose a technique by which it is possible to stabilize explicit methods for quasi-linear hyperbolic equations. The stabilization turns out to be so effective that explicit methods become a good alternative to unconditionally stable implicit methods.     

水坝瞬态热传导分析中的拟初始条件边界元法

使用一种时域边界元方法对混凝土水坝进行瞬态热传导分析。在对时间积分进行离散计算时,采用一种拟初始条件法,即在时间步迭代计算的过程中,将之前计算结果对当前时间步的影响都视作当前时间步的初始条件。在所取时间步长较小的情况下,这种处理方法容易导致数值结果不稳定,即每一步的计算误差会累计放大,最终导致计算崩溃。本文提出一种虚拟时刻方法以缓解这类数值不稳定现象,在该方法中,时间步长首先放大至合适尺度,计算某个虚拟时刻(往往在真实计算时刻之后)的温度和流量分布,再通过插值方法换算出真实时刻的温度和流量分布。在虚拟时刻点上的温度和流量计算过程中,边界已知温度或流量由真实时刻的温度或流量进行外插得到。本文简单证明了该方法在温度和流量随时间呈线性变化情况下的正确性,最后给出了两个分析实例,验证了该方法的准确性和稳定性。     

On the behavior of a three-dimensional fractional viscoelastic constitutive model

In this paper a three-dimensional isotropic fractional viscoelastic model is examined. It is shown that if different time scales for the volumetric and deviatoric components are assumed, the Poisson ratio is time varying function; in particular viscoelastic Poisson ratio may be obtained both increasing and decreasing with time. Moreover, it is shown that, from a theoretical point of view, one-dimensional fractional constitutive laws for normal stress and strain components are not correct to fit uniaxial experimental test, unless the time scale of deviatoric and volumetric are equal. Finally, the model is proved to satisfy correspondence principles also for the viscoelastic Poisson’s ratio and some issues about thermodynamic consistency of the model are addressed.     

Implicit time accurate simulation of unsteady flow

Implicit time integration was studied in the context of unsteady shock‐boundary layer interaction flow. With an explicit second‐order Runge–Kutta scheme, a reference solution to compare with the implicit second‐order Crank–Nicolson scheme was determined. The time step in the explicit scheme is restricted by both temporal accuracy as well as stability requirements, whereas in the A‐stable implicit scheme, the time step has to obey temporal resolution requirements and numerical convergence conditions. The non‐linear discrete equations for each time step are solved iteratively by adding a pseudo‐time derivative. The quasi‐Newton approach is adopted and the linear systems that arise are approximately solved with a symmetric block Gauss–Seidel solver. As a guiding principle for properly setting numerical time integration parameters that yield an efficient time accurate capturing of the solution, the global error caused by the temporal integration is compared with the error resulting from the spatial discretization. Focus is on the sensitivity of properties of the solution in relation to the time step. Numerical simulations show that the time step needed for acceptable accuracy can be considerably larger than the explicit stability time step; typical ratios range from 20 to 80. At large time steps, convergence problems that are closely related to a highly complex structure of the basins of attraction of the iterative method may occur. Copyright © 2001 John Wiley & Sons, Ltd.     

考虑主动时滞的汽车悬架系统控制特性研究

以汽车悬架系统为研究对象,采用理论和试验相结合的方法对考虑主动时滞的悬架系统控制特性进行分析。首先建立含时滞悬架系统动力学模型,分析系统的控制稳定性。理论和仿真结果均表明,采用传统二次型最优控制律不能保证含时滞系统的稳定性。系统时滞量存在稳定区间,时滞超出稳定区间时系统将失稳发散;为了保证控制系统的稳定性,采用状态变换法设计了含时滞系统的主动控制律,计算表明,该控制律可以保证系统稳定性。研究还发现,时滞量的变化会使系统振动幅值产生较大改变,为此在控制系统中引入主动时滞,研究主动时滞对系统振动特性的影响,计算表明,合理的主动时滞可以降低系统振动幅值;为验证结果的正确性,搭建了悬架时滞主动控制试验平台,通过对相同工况下仿真结果与试验结果进行对比,发现两者具有较好的一致性;而由于悬架受到的路面激励具有随机性,采用含时滞系统的主动控制律对路面随机激励下的悬架系统进行控制分析,发现当主动时滞为0.04 s时,车身加速度均方根值比无主动时滞降低了39.4%,说明主动时滞对悬架控制的有效性。本文研究对时滞主动控制的理论研究具有重要的促进作用。     

Zur Systematik instationärer Transportvorgänge

In order to find a universal scheme to compute transient transport problems, 18 transient processes originating in the fields of chemical engineering and fluid dynamics are investigated. They all have vanishing fluxes as time increases and can be classified into two groups. The first group represents all those problems which are characterized by a resistance that builds up and depends on the quantity transferred up to the time t. In the simplest cases they are described by only one characteristic number. If the resistance is proportional to that quantity the known relationships with squareroot of time are obtained. The driving gradients that diminish with time characterize the second group. If they can be set proportional to the quantity that is stilllto be transferred till the equilibrium state is reached (flux=0) one yields the known exponential functions with at least two dimensionless parameters. Also, for some more complicated problems the governing equations are given.     

Criteria for the Existence and Lower Bounds of Principal Eigenvalues of Time Periodic Nonlocal Dispersal Operators and Applications

The current paper is concerned with the spectral theory, in particular, the principal eigenvalue theory, of nonlocal dispersal operators with time periodic dependence, and its applications. Nonlocal and random dispersal operators are widely used to model diffusion systems in applied sciences and share many properties. There are also some essential differences between nonlocal and random dispersal operators, for example, a smooth random dispersal operator always has a principal eigenvalue, but a smooth nonlocal dispersal operator may not have a principal eigenvalue. In this paper, we first establish criteria for the existence of principal eigenvalues of time periodic nonlocal dispersal operators with Dirichlet type, Neumann type, or periodic type boundary conditions. It is shown that a time periodic nonlocal dispersal operator possesses a principal eigenvalue provided that the nonlocal dispersal distance is sufficiently small, or the time average of the underlying media satisfies some vanishing condition with respect to the space variable at a maximum point or is nearly globally homogeneous with respect to the space variable. Next we obtain lower bounds of the principal spectrum points of time periodic nonlocal dispersal operators in terms of the corresponding time averaged problems. Finally we discuss the applications of the established principal eigenvalue theory to time periodic Fisher or KPP type equations with nonlocal dispersal and prove that such equations are of monostable feature, that is, if the trivial solution is linearly unstable, then there is a unique time periodic positive solution which is globally asymptotically stable.     

Existence,Uniqueness, and Stability of Generalized Traveling Waves in Time Dependent Monostable Equations

The current paper is devoted to the study of traveling wave solutions of spatially homogeneous monostable reaction diffusion equations with ergodic or recurrent time dependence, which includes periodic and almost periodic time dependence as special cases. Such an equation has two spatially homogeneous and time recurrent solutions with one of them being stable and the other being unstable. Traveling wave solutions are a type of entire solutions connecting the two spatially homogeneous and time recurrent solutions. Recently, the author of the current paper proved that a spatially homogeneous time almost periodic monostable equation has a spreading speed in any given direction. This result can be easily extended to monostable equations with recurrent time dependence. In this paper, we introduce generalized traveling wave solutions for time recurrent monostable equations and show the existence of such solutions in any given direction with average propagating speed greater than or equal to the spreading speed in that direction and non-existence of such solutions of slower average propagating speed. We also show the uniqueness and stability of generalized traveling wave solutions in any given direction with average propagating speed greater than the spreading speed in that direction. Moreover, we show that a generalized traveling wave solution in a given direction with average propagating speed greater than the spreading speed in that direction is unique ergodic in the sense that its wave profile and wave speed are unique ergodic, and if the time dependence of the monostable equation is almost periodic, it is almost periodic in the sense that its wave profile and wave speed are almost periodic.     

Computing brine transport in porous media with an adaptive-grid method

An adaptive-grid finite-difference method is applied to a model for non-isothermal, coupled flow and transport of brine in porous media. In the vicinity of rock salt formations the salt concentration in the fluid becomes large, giving rise to disparate scales in the salt concentrations profiles. A typical situation one encounters is that of a sharp freshwater-saltwater interface that moves in time. In such situations adaptive-grid methods are more effective than standard fixed-grid methods, since they refine the space grid locally and, hence, provide for substantial reduction in the number of grid points, memory use and CPU time. The adaptive-grid method of this paper is a static, local uniform grid refinement method. Its main feature is that it integrates on nested sequences of locally uniformly refined Cartesian space grids, which are automatically adjusted in time to follow rapid spatial transitions. Variable time steps are used to cope with rapid temporal transitions, including a fast march to possible steady-state solutions. For time stepping, the implicit, second-order BDF scheme is used. Two specific example problems are numerically illustrated. The main physical properties involved here are advection and dispersion and in case of dominant advection sharp freshwater-saltwater interfaces arise.     

连续损伤力学基临界奇异指数与破坏时间预测

响应量在临近破坏时呈现出临界幂律奇异性加速特征,是一种被广泛证实的灾变破坏前兆,并被火山、滑坡和岩石破坏实验等后验预测结果证实为一种对破坏时间进行短临期预测的可行方法.但是,奇异性指数测量值的较大分散性导致了对其具体取值的争议和预测效果的不确定性.因此,理解奇异性指数取值特征及其内在物理控制因素,成为了一个核心问题.本文基于连续介质损伤力学和材料时间相关失效特征,构建了刻画损伤加速发展通向破坏过程的力学模型.导出了恒名义应力蠕变加载和控制名义应力随时间线性增大两种典型加载方式下,损伤和应变率加速发展通向破坏的临界幂律奇异性前兆特征.阐明了临界幂律奇异性指数取值依赖于材料损伤与承受真应力之间的非线性关系这一内在物理根源,表明了实际测量中奇异性指数的分散性不完全归结于测量数据误差,而是有着内在物理控制因素.针对破坏前奇异性指数的不确定性,建议了在未知奇异性指数条件下预测破坏时间的方法,并基于花岗岩脆性蠕变破坏实验进行了验证和说明.      

Transient bending waves in plates studied by hologram interferometry

Propagating bending waves are studied in plates made of aluminum and wood. The waves are generated by the impact of a ballistic pendulum. Hologram interferometry, with a double pulsed ruby laser as the light source, is used to record the out of plane motion of the waves. Elliptic-like fringes visualize differences in wave speed for different directions in the anisotropic plate and circular ones are obtained for the isotropic plate. The experimental data for the isotropic plate compare favorably with analytical results derived from the Kirchhoff-plate equation with a point impact of finite duration. A similarity variable is found when starting conditions are modeled as a Dirac pulse in space and time, that brings new understanding to the importance of specific parameters for wave propagation in plates. A formal solution is obtained for a point force with an arbitrary time dependence. For times much larger than the contact time, the plate deflection is shown to be identical to that from a Dirac pulse applied at the mean contact time. A method for determining material parameters, and the mean contact time, from the interferograms is hence developed.     

时变参数时滞减振控制研究

时滞动力吸振器对谐波激励有着良好的减振控制效果,但对随机激励的减振控制效果却并不明显,具体表现为时滞动力吸振器对随机激励的减振控制效果与被动吸振器几乎相同。针对上述问题,本文提出了一种新的时变参数时滞减振控制方法。在原有时滞减振控制方法的基础上,首先将时滞增益系数由定值形式变为时间函数形式,然后通过时变优化得到多组时滞控制参数并使其以一定时间周期循环作用于减振控制过程,通过这种方法进一步改善了时滞动力吸振器减振性能。本文最后以二自由度时滞动力吸振器减振模型为例,以主系统的振动响应为仿真对象,运用精细积分法求解了具有时变时滞参数的时滞动力学方程,以此得到了在谐波激励和随机激励作用下主系统振动的时域仿真结果。研究结果表明,在时变参数时滞动力吸振器的控制下,主系统无论是受谐波激励作用还是受随机激励作用,其振动位移、振动速度和振动加速度均比在定值参数时滞动力吸振器控制下时有大幅的减少,时滞动力吸振器的减振性能有了明显的改善。      

On the comparison of two different solutions in the form of series of the governing equation of an unsteady flow of a second grade fluid

In this paper, two different solutions in the form of series of the governing equation of unsteady flow of a second grade fluid are considered. These are series expansions with respect to inverse power of time and a perturbation expansion. Two illustrative examples are given. One of them is the unsteady flow of a second grade fluid over a plane wall suddenly set in motion and the other is the diffusion of a line vortex in a fluid of second grade. It is a remarkable fact that the expression of the series expansion with respect to inverse power of time is exactly in the same form as that of the perturbation expansion. Thus, it is possible to replace a series expansion with respect to inverse power of time with a perturbation expansion.     

The effect of two distinct fast time scales in the rotating,stratified Boussinesq equations: variations from quasi-geostrophy

Inspired by the use of fast singular limits in time-parallel numerical methods for a single fast frequency, we consider the limiting, nonlinear dynamics for a system of partial differential equations when two fast, distinct time scales are present. First-order slow equations are derived via the method of multiple time scales when the two small parameters are related by a rational power. We find that the resultant system depends only on the relationship of the two fast time scales, i.e. which fast time is fastest? Using the theory of cancellation of fast oscillations, we show that with the appropriate assumptions on the nonlinear operator of the full system, this reduced slow system is exactly that which the solution will converge to if each asymptotic limit is considered sequentially. The same result is also obtained via the method of renormalization. The specific example of the rotating, stratified Boussinesq equations is explored in detail, indicating that the most common distinguished limit of this system—quasi-geostrophy, is not the only limiting asymptotic system.     

Shape Recovery of Viscoelastic Beams After Stowage

The deployment of viscoelastic structures that have been held stowed for a given time duration can be formulated as a viscoelastic boundary value problem in which the prescribed condition switches from constant displacement to constant traction. This paper presents closed-form expressions for the load relaxation and shape recovery of a linear viscoelastic beam subject to such time-varying constraints. It is shown that a viscoelastic beam recovers to its original shape asymptotically over time. The analytical solutions are employed to investigate the effect of temperature and stowage time on the time required to achieve recovery with a specified precision. Based on the time-temperature equivalence principle, the relationship between recovery time and holding duration is concisely presented on a single plot. It is found that recovery time increases with holding duration but with a diminishing effect.     

Taylor-Galerkin-based spectral element methods for convection-diffusion problems

Several explicit Taylor-Galerkin-based time integration schemes are proposed for the solution of both linear and non-linear convection problems with divergence-free velocity. These schemes are based on second-order Taylor series of the time derivative. The spatial discretization is performed by a high-order Galerkin spectral element method. For convection-diffusion problems an operator-splitting technique is given that decouples the treatment of the convective and diffusive terms. Both problems are then solved using a suitable time scheme. The Taylor-Galerkin methods and the operator-splitting scheme are tested numerically for both convection and convection-diffusion problems.     

Temperature distribution in a storage tank

In this work the temperature distribution in a vertical storage tank fed at its top with warmer fluid at constant flow rate is investigated. The flow rate is considered to be so small that laminar flow conditions are established. The temperature of the admitted fluid is allowed to vary with time where step and ramp time functions are used in this work. A rigorous analytical solution is obtained which describes the temperature as function of vertical depth and time. Graphical representation of the temperature distribution is given for selected values of the flow parameter and the time constant. Special reference is made in selecting these numerical values to solar water heating systems. The results are discussed and also limitations are recalled.     

A new approach for computing the steady state fluid–structure interaction response of periodic problems

A special type of fluid–structure interaction (FSI) problems are problems with periodic boundary conditions like in turbomachinery. The steady state FSI response of these problems is usually calculated with similar techniques as used for transient FSI analyses. This means that, when the fluid and structure problem are not simultaneously solved with a monolithic approach, the problem is partitioned into a fluid and structural part and that each time step coupling iterations are performed to account for strong interactions between the two sub-domains. This paper shows that a time-partitioned FSI computation can be very inefficient to compute the steady state FSI response of periodic problems. A new approach is introduced in which coupling iterations are performed on periodic level instead of per time step. The convergence behaviour can be significantly improved by implementing existing partitioned solution methods as used for time step coupling (TSC) algorithms in the time periodic coupling (TPC) framework. The new algorithm has been evaluated by comparing the convergence behaviour to TSC algorithms. It is shown that the number of fluid–structure evaluations can be considerably reduced when a TPC algorithm is applied instead of a TSC. One of the most appealing advantages of the TPC approach is that the structural problem can be solved in the frequency domain resulting in a very efficient algorithm for computing steady state FSI responses.     

On the time development of dispersion in electroosmotic flow through a rectangular channel

This is an analytical study on the time development of hydrodynamic dispersion of an inert species in electroosmotic flow through a rectangular channel.The objective is to determine how the channel side walls may affectthe dispersion coefficient at different instants of time.Tothis end,the generalized dispersion model,which is valid forshort and long times,is employed in the present study.Analytical expressions are derived for the convection and dispersion coefficients as functions of time,the aspect ratio ofthe channel,and the Debye-Hu¨ckel parameter representingthe thickness of the electric double layer.For transport ina channel of large aspect ratio,the dispersion may undergoseveral stages of transience.The initial,fast time development is controlled by molecular diffusion across the narrowchannel height,while the later,slower time development isgoverned by diffusion across the wider channel breadth.Fora sufficiently large aspect ratio,there can be an interludebetween these two periods during which the coefficient isnearly steady,signifying the resemblance of the transportto that in a parallel-plate channel.Given a sufficiently longtime,the dispersion coefficient will reach a fully-developedsteady value that may be several times higher than that without the side wall effects.The time scales for these periods oftransience are identified in this paper.     

On the flows produced by sudden application of a constant pressure gradient or by impulsive motion of a boundary

Some properties of the time-dependent Navier-Stokes equations are discussed for flows impulsively started from rest by sudden application of a constant pressure gradient or by the impulsive motion of a boundary. Five illustrative examples are given. They are: unsteady flow in a circular cylinder moving parallel to its length, starting flow in a circular pipe, unsteady flow in a rotating cylinder, starting flow in a rectangular channel moving parallel to its length and unsteady flow in a channel of rectangular cross-section. It is found that the expressions of the quantities such as velocity, flux and skin friction are in series forms which may be rapidly convergent for large values of the time but slowly convergent for small values of the time or vice versa. It is shown that if their expressions can be found for one of large values of the time or small values of the time, these expressions can be used for the other.