Search

We consider a one-dimensional first-order nonlinear wave equation, the so-called forward Maxwell equation (FME), which applies to a few-cycle optical pulse propagating along a preferred direction in a nonlinear medium, e.g., ultrashort pulses in nonlinear fibers. The model is a good approximation to the standard second-order wave equation under assumption of weak nonlinearity and spatial homogeneity in the propagation direction. We compare FME to the commonly accepted generalized nonlinear Schrodinger equation, which quantifies the envelope of a quickly oscillating wave field based on the slowly varying envelope approximation. In our numerical example, we demonstrate that FME, in contrast to the envelope model, reveals new spectral lines when applied to few-cycle pulses. We analyze and compare pseudo-spectral numerical schemes employing symmetric splitting for both models. Finally, we adopt these schemes to a parallel computation and discuss scalability of the parallelization.

A dynamical model is employed to determine some peculiar properties of the scission neutrons, i.e., those emitted at the rupture of the neck between the fragments during low energy nuclear fission. Thus, by a specially adapted procedure, we calculate the time-dependent decay rate. This, defines the character of the emission process: pulsed or exponentially. From the time dependence of the survival probability, we deduced the “half-life” of the scission neutron emission. The angular distribution of the scission neutrons with respect to the fission axis is calculated as well. This is obtained for sets of neutron wave functions defined by a given quantum number [Formula: see text] (projection of the total angular momentum on the symmetry axis). A strong dependence on [Formula: see text] was found, namely, wave functions with different [Formula: see text] have different most probable emission angles: from emission along the fission axis (1/2) to emission perpendicular to the fission axis (9/2). This result leads to a new interpretation of the measured angular distribution of the prompt fission neutrons. Finally, an extension of the dynamical model (i.e., with a time-dependent potential also after scission) is presented.

International audience; In this work, a numerical model is developed for a rapid computation of eddy currents in composite materials , adaptable for both carbon fiber reinforced polymers (CFRPs) for NDT applications and multifilamentary high temperature superconductive (HTS) tapes for AC loss evaluation. The proposed model is based on an integro-differential formulation in terms of the electric vector potential in the frequency domain. The high anisotropy and the nonlinearity of the considered materials are easily handled in the frequency domain.

In this paper a four stages twelfth algebraic order symmetric two-step method with vanished phase-lag and its first, second, third, fourth and fifth derivatives is developed for the first time in the literature. For the new proposed method: (1) we will study the phase-lag analysis, (2) we will present the development of the new method, (3) the local truncation error (LTE) analysis will be studied. The analysis is based on a test problem which is the radial time independent Schrödinger equation, (4) the stability and the interval of periodicity analysis will be presented, (5) stepsize control technique will also be presented, (6) the examination of the accuracy and computational cost of the proposed algorithm which is based on the approximation of the Schrödinger equation.

High-level algebra-algorithmic software tools for automated design of parallel code in the OpenMP environment are developed for the purpose of both producing efficient parallel code and increasing the performance of program developers. Application of the tools is illustrated with an example of a problem in atmosphere circulation modeling, represented as a service belonging to an Internet portal providing meteorological forecasting services. Results of execution of the parallel weather forecasting program on multiprocessor platforms are given.

We present and analyze different splitting algorithms for numerical solution of the both classical and generalized nonlinear Schrödinger equations describing propagation of wave packets with special emphasis on applications to nonlinear fiber-optics. The considered generalizations take into account the higher-order corrections of the linear differential dispersion operator as well as the saturation of nonlinearity and the self-steepening of the field envelope function. For stabilization of the pseudo-spectral splitting schemes for generalized Schrödinger equations a regularization based on the approximation of the derivatives by the low number of Fourier modes is proposed. To illustrate the theoretically predicted performance of these schemes several numerical experiments have been done.

For the simulation of the propagation of optical waves in open wave guiding structures of integrated optics the parabolic approximation of the scalar wave equation is commonly used. This approach is commonly termed the beam propagation method (BPM). It is of paramount importance to have well-performing transparent boundary conditions applied on the boundaries of the finite computational window, to enable the superfluous portion of the propagating wave to radiate away from the wave guiding structure. Three different formulations (continuous, semi-discrete and fully-discrete) of the non-local transparent boundary conditions are described and compared here.

In this work we present a new numerical technique for solving periodic structure problems. This new approach possesses several advantages. First, it allows for a fast evaluation of the Robin-to-Robin operator for periodic array problems. Secondly, this computational method can also be used for bi-periodic structure problems with local defects. Our strategy is an improvement of the recently developed recursive doubling process by Yuan and Lu. In this paper we consider several problems, such as the exterior elliptic problems with strong coercivity, the time-dependent Schrödinger equation and finally the Helmholtz equation with damping.

We present a numerical study aimed at assessing the validity of certain approximations, chiefly the fluid and quasi-neutral ones, generally used in the theoretical and computational study of fusion plasmas. The impact of numerical artifacts has been minimised by using the very precise weighted essentially non-oscillatory (WENO) method, which can be applied to both kinetic and fluid simulations. This is the first application of WENO schemes to Eulerian simulations of Vlasov plasmas. The fluid modelling of plasmas appears adequate even when collisions are negligible, provided certain relationships hold between the characteristic speed of the phenomenon under consideration and the thermal velocities of the species which make up the plasma. The breakdown of this behaviour is probably caused by non-linear Landau damping; it appears linked with filamentation, which suggests that the magnitude of the departure from Maxwellian equilibrium may not be the one and only measurement of non-fluid behaviour. As for quasi-neutrality, our study suggests that the generally accepted criteria should be used with great care, according to the degree of neutrality expected.

Polarization properties of apertureless-type scanning near-field optical microscopy (a-SNOM) were measured experimentally and were also analyzed using a finite-difference time-domain (FDTD) simulation. Our study reveals that the polarization properties in the a-SNOM are maintained and the a-SNOM works as a wave plate expressed by a Jones matrix. The measured signals obtained by the lock-in detection technique could be decomposed into signals scattered from near-field region and background signals reflected by tip and sample. Polarization images measured by a-SNOM with an angle resolution of 1° are shown. FDTD analysis also reveals the polarization properties of light in the area between a tip and a sample are p-polarization in most of cases.

International audience; In this review article we discuss different techniques to solve numerically the time-dependent Schrödinger equation on unbounded domains. We present and compare several approaches to implement the classical transparent boundary condition into finite difference and finite element discretizations. We present in detail the approaches of the authors and describe briefly alternative ideas pointing out the relations between these works. We conclude with several numerical examples from different application areas to compare the presented techniques. We mainly focus on the one-dimensional problem but also touch upon the situation in two space dimensions and the cubic nonlinear case.