Replica Symmetry Breaking in Short-Range Spin Glasses: Theoretical Foundations and Numerical Evidences

We discuss replica symmetry breaking (RSB) in spin glasses. We update work in this area, from both the analytical and numerical points of view. We give particular attention to the difficulties stressed by Newman and Stein concerning the problem of constructing pure states in spin glass systems. We mainly discuss what happens in finite-dimensional, realistic spin glasses. Together with a detailed review of some of the most important features, facts, data, and phenomena, we present some new theoretical ideas and numerical results. We discuss among others the basic idea of the RSB theory, correlation functions, interfaces, overlaps, pure states, random field, and the dynamical approach. We present new numerical results for the behaviors of coupled replicas and about the numerical verification of sum rules, and we review some of the available numerical results that we consider of larger importance (for example, the determination of the phase transition point, the correlation functions, the window overlaps, and the dynamical behavior of the system).     

一个磁流体力学中心型数值通量的耗散控制器

提出一种针对磁流体力学模拟中中心型数值通量的耗散控制器.由于磁流体力学方程组较为复杂,中心型数值通量不需要复杂的特征分解,因而中心型数值格式更适合磁流体力学的数值模拟.但是稳定单调的中心型数值通量比迎风型数值通量耗散大,对此,提出一种控制中心型数值通量数值耗散的控制器.将这个耗散控制器应用于以局部Lax-Feiedrichs格式通量为例的中心型数值通量.两个基准解的数值算例证明方法有效.     

Loop space Hamiltonians and numerical methods for large-N gauge theories

We consider large-N gauge theories in the hamiltonian, collective field approach. We derive an alternative collective representation which leads to significant reduction when translation invariance is invoked. It allows for a simplified computer simulation of loop rearrangements and the development of numerical techniques in the hamiltonian, loop space formalism. We proceed to give numerical evidence for validity of our representation and outline a general numerical approach for solving large-N QCD in terms of gauge-invariant Wilson loop variables.     

Wave propagation in locally perturbed periodic media (case with absorption): Numerical aspects

We are interested in the numerical simulation of wave propagation in media which are a local perturbation of an infinite periodic one. The question of finding artificial boundary conditions to reduce the actual numerical computations to a neighborhood of the perturbation via a DtN operator was already developed in [1] at the continuous level. We deal in this article with the numerical aspects associated to the discretization of the problem. In particular, we describe the construction of discrete DtN operators that relies on the numerical solution of local cell problems, non stationary Ricatti equations and the discretization of non standard integral equations in Floquet variables.     

Implicit schemes for real-time lattice gauge theory

We develop new gauge-covariant implicit numerical schemes for classical real-time lattice gauge theory. A new semi-implicit scheme is used to cure a numerical instability encountered in three-dimensional classical Yang-Mills simulations of heavy-ion collisions by allowing for wave propagation along one lattice direction free of numerical dispersion. We show that the scheme is gauge covariant and that the Gauss constraint is conserved even for large time steps.     

Loop-space hamiltonians and numerical methods for large-N gauge theories (II)

We generalize our numerical loop-space methods to consider the full large-N Yang-Mills problem. First we study the weak-coupling phase in loop space and derive the equation obeyed by spacelike Wilson loops. We then consider the problem of topological reduction and derive the effective potential for topologically distinct loops. In connection with numerical computations we discuss the issues of truncation of the system. We present an initial numerical result for a simplest truncation of the full (2+1)-dimensional Yang-Mills theory.     

Bootstrapping Calabi–Yau quantum mechanics

Recently, a novel bootstrap method for numerical calculations in matrix models and quantum mechanical systems was proposed. We apply the method to certain quantum mechanical systems derived from some well-known local toric Calabi–Yau geometries, where the exact quantization conditions have been conjecturally related to topological string theory. We find that the bootstrap method provides a promising alternative for the precision numerical calculations of the energy eigenvalues. An improvement in our approach is to use a larger set of two-dimensional operators instead of one-dimensional ones. We also apply our improved bootstrap methods to some non-relativistic models in the recent literature and demonstrate better numerical accuracies.     

An accurate and robust numerical method for micromagnetics simulations

We propose a new robust, accurate, and fast numerical method for solving the Landau–Lifshitz equation which describes the relaxation process of the magnetization distribution in ferromagnetic material. The proposed numerical method is second-order accurate in both space and time. The approach uses the nonlinear multigrid method for handling the nonlinearities at each time step. We perform numerical experiments to show the efficiency and accuracy of the new algorithm on two- and three-dimensional space. The numerical results show excellent agreements with exact analytical solutions, the second-order accuracy in both space and time, and the energy conservation or dissipation property.     

Construction of grating profiles yielding prescribed diffraction efficiencies

We present a numerical iteration procedure for the (re)construction of perfectly conducting gratings that yield prescribed (or measured) diffraction efficiencies. Prior knowledge of the approximate grating profile is not required. The method is not limited to shallow, smooth profiles. We give numerical results for a profile synthesis and a reconstruction problem. We discuss the question of nonuniqueness.     

Numerical solution of the Fokker-Planck equation with variable coefficients

We study the numerical solution of the Fokker-Planck equation. This equation gives a good approximation to the radiative transport equation when scattering is peaked sharply in the forward direction which is the case for light propagation in tissues, for example. We derive first the numerical solution for the problem with constant coefficients. This numerical solution is constructed as an expansion in plane wave solutions. Then we extend that result to take into account coefficients that vary spatially. This extension leads to a coupled system of initial and final value problems. We solve this system iteratively. Numerical results show the utility of this method.     

Exact and numerical solutions to the perturbed sine-Gordon equation

We present numerical and analytic solutions to the perturbed sine-Gordon equation, which models long Josephson tunnel junctions. We make comparisons between numerical results and results obtained from perturbational methods. We present unstable, analytic kink solutions to the equation and further a solution, which is an array of kinks, corresponding to a solution, where the current through the junction is larger than the critical current.     

球坐标动量空间下相对论Vlasov方程的数值算法

针对相对论Vlasov方程动量区间跨度大、难以计算的困难,将相对论Vlasov方程在球坐标动量空间中进行数值求解.对相对论Vlasov方程球坐标动量空间构造4阶非分裂守恒型数值格式.数值模拟相对论Landau阻尼问题并与解析理论进行比较,验证数值模型和算法的有效性.对激光等离子体相互作用进行初步模拟分析,表明通过采用球坐标下的动量空间,可在相对较少动量网格情形下,获得与粒子模拟可相互验证的结果.     

Power laws and crossovers in off-critical surface-directed spinodal decomposition

We study the dynamics of phase separation in binary mixtures near a surface with a preferential attraction for one of the components of the mixture. We obtain detailed numerical results for a range of mixture compositions. In the case where the minority component is attracted to the surface, wetting layer growth is characterized by a crossover from a surface-potential-dependent growth law to a universal law. We formulate a simple phenomenological model to explain our numerical results.     

Phase-field models for moving boundary problems: Controlling metastability and anisotropy

We introduce a variant of a recently proposed method of rotated lattices for numerical treatment of moving boundary problems. The usual lattice introduced for numerical computation of phase-field models gives rise to unphysical metastable states and anisotropy. In the present case we rotate and shift the lattice by random angles and fractions of a lattice constant. We show that a twelve point interpolation formula is adequate to keep numerical interpolation errors sufficiently localized. This removes the unphysical metastabilities and makes the model fully isotropic. This is demonstrated by a few example-calculations for dendritic pattern formation.     

An anti-diffusive numerical scheme for the simulation of interfaces between compressible fluids by means of a five-equation model

We propose a discretization method of a five-equation model with isobaric closure for the simulation of interfaces between compressible fluids. This numerical solver is a Lagrange–Remap scheme that aims at controlling the numerical diffusion of the interface between both fluids. This method does not involve any interface reconstruction procedure. The solver is equipped with built-in stability and consistency properties and is conservative with respect to mass, momentum, total energy and partial masses. This numerical scheme works with a very broad range of equations of state, including tabulated laws. Properties that ensure a good treatment of the Riemann invariants across the interface are proven. As a consequence, the numerical method does not create spurious pressure oscillations at the interface. We show one-dimensional and two-dimensional classic numerical tests. The results are compared with the approximate solutions obtained with the classic upwind Lagrange–Remap approach, and with experimental and previously published results of a reference test case.     

Interplay between experimental and numerical approaches in the fluid dynamo problem

After years of purely analytical and numerical investigations, the dynamo fluid problem has advanced to a phase of experimental study. We present an outline of the numerical steps that have accompanied the Von Kármán Sodium (VKS) experiment at Cadarache for the past ten years. We show how various numerical studies contributed progressively to the optimization of the experimental facility. The recent success of the VKS2 experiment of September 2006 in achieving dynamo action has prompted an extension of the numerical code. Modeling of the electromotive force induced in the volume of the impellers shows that an axial dipole is excited, as observed in the experiment. We infer from these results that the observed value of the critical magnetic Reynolds number may be linked to the soft iron of the impellers and not to turbulence which occurs for any choice of materials. We conclude with proposals for further lines of numerical development. To cite this article: J. Léorat, C. Nore, C. R. Physique 9 (2008).     

An unconditionally gradient stable numerical method for solving the Allen-Cahn equation

We consider an unconditionally gradient stable scheme for solving the Allen-Cahn equation representing a model for anti-phase domain coarsening in a binary mixture. The continuous problem has a decreasing total energy. We show the same property for the corresponding discrete problem by using eigenvalues of the Hessian matrix of the energy functional. We also show the pointwise boundedness of the numerical solution for the Allen-Cahn equation. We describe various numerical experiments we performed to study properties of the Allen-Cahn equation.     

Multifractality in elastic percolation

We present numerical results on the distribution of forces in the central-force percolation model at threshold in two dimensions. We conjecture a relation between the multifractal spectrum of scalar and vector percolation that we test for central-foce percolation. This relation is in excellent agreement with our numerical data.     

Gaussian models for the distribution of Brownian particles in tilted periodic potentials The limit of high barriers

We present two Gaussian approximations for the time-dependent probability density function (PDF) of an overdamped Brownian particle moving in a tilted periodic potential. We assume high potential barriers in comparison with the noise intensity. The accuracy of the proposed approximated expressions for the time-dependent PDF is checked with numerical simulations of the Langevin dynamics. We found a quite good agreement between theoretical and numerical results at all times.     

Asymptotic distribution of global errors in the numerical computations of dynamical systems

We propose an analysis of the effects introduced by finite-accuracy and round-off arithmetic on numerical computations of discrete dynamical systems. Our method, which uses the statistical tool of the decay of fidelity, computes the error by directly comparing the numerical orbit with the exact one (or, more precisely, with another numerical orbit computed with a much higher accuracy). Furthermore, as a model of the effects of round-off arithmetic on the map, we also consider a random perturbation of the exact orbit with an additive noise, for which exact results can be obtained for some prototype maps. We investigate the decay laws of fidelity and their relationship with the error probability distribution for regular and chaotic maps, for both additive and numerical noise. In particular, for regular maps we find an exponential decay for additive noise, and a power-law decay for numerical noise. For chaotic maps, numerical noise is equivalent to additive noise, and our method is suitable for identifying a threshold for the reliability of numerical results, i.e., the number of iterations below which global errors can be ignored. This threshold grows linearly with the number of bits used to represent real numbers.