Loss of Accuracy in Numerical Computations

Михаил М. Константинов, Петко Х. Петков - Разгледани са възможните катастрофални ефекти от неправилното използване на крайна машинна аритметика с плаваща точка. За съжаление, тази тема не винаги се разбира достатъчно добре от студентите по приложна и изчислителна математика, като положението в инженерните и икономическите специалности в никакъв случай не е по-добро. За преодоляване на този образователен пропуск тук сме разгледали главните виновници за загубата на точност при числените компютърни пресмятания. Надяваме се, че представените резултати ще помогнат на студентите и лекторите за по-добро разбиране и съответно за избягване на основните фактори, които могат да разрушат точността при компютърните числени пресмятания. Последното не е маловажно – числените катастрофи понякога стават истински, с големи щети и човешки жертви.

Numerical Simulation Of Sediment Transport And Bedmorphology Around A Hydraulic Structure On A River

Scour around hydraulic structures is a critical problem in hydraulic engineering. Under prediction of scour depth may lead to costly failures of the structure, while over prediction might result in unnecessary costs. Unfortunately, up-to-date empirical scour prediction formulas are based on laboratory experiments that are not always able to reproduce field conditions due to complicated geometry of rivers and temporal and spatial scales of a physical model. However, computational fluid dynamics (CFD) tools can perform using real field dimensions and operating conditions to predict sediment scour around hydraulic structures. In Korea, after completing the Four Major Rivers Restoration Project, several new weirs have been built across Han, Nakdong, Geum and Yeongsan Rivers. Consequently, sediment deposition and bed erosion around such structures have became a major issue in these four rivers. In this study, an application of an open source CFD software package, the TELEMAC-MASCARET, to simulate sediment transport and bed morphology around Gangjeong weir, which is the largest multipurpose weir built on Nakdong River. A real bathymetry of the river and a geometry of the weir have been implemented into the numerical model. The numerical simulation is carried out with a real hydrograph at the upstream boundary. The bedmorphology obtained from the numerical results has been validated against field observation data, and a maximum of simulated scour depth is compared with the results obtained by empirical formulas of Hoffmans. Agreement between numerical computations, observed data and empirical formulas is judged to be satisfactory on all major comparisons. The outcome of this study does not only point out the locations where deposition and erosion might take place depending on the weir gate operation, but also analyzes the mechanism of formation and evolution of scour holes after the weir gates.

Numerical simulation of the upward propagation of a flame in a vertical tube filled with a very lean mixture.

Upwardpropagation of a premixed flame in averticaltubefilled with a very leanmixture is simulated numerically using a single irreversible Arrhenius reaction model with infinitely high activation energy. In the absence of heat losses and preferential diffusion effects, a curved flame with stationary shape and velocity close to those of an open bubble ascending in the same tube is found for values of the fuel mass fraction above a certain minimum that increases with the radius of the tube, while the numerical computations cease to converge to a stationary solution below this minimum mass fraction. The vortical flow of the gas behind the flame and in its transport region is described for tubes of different radii. It is argued that this flow may become unstable when the fuel mass fraction is decreased, and that this instability, together with the flame stretch due to the strong curvature of the flame tip in narrow tubes, may be responsible for the minimum fuel mass fraction. Radiation losses and a Lewis number of the fuel slightly above unity decrease the final combustion temperature at the flame tip and increase the minimum fuel mass fraction, while a Lewis number slightly below unity has the opposite effect.

Numerical analysis of deflagration initation by a hot jet

Numerical simulations of axisymmetric reactive jets with one-step Arrhenius kinetics are used to investigate the problem of deflagration initiation in a premixed fuel–air mixture by the sudden discharge of a hot jet of its adiabatic reaction products. For the moderately large values of the jet Reynolds number considered in the computations, chemical reaction is seen to occur initially in the thin mixing layer that separates the hot products from the cold reactants. This mixing layer is wrapped around by the starting vortex, thereby enhancing mixing at the jet head, which is followed by an annular mixing layer that trails behind, connecting the leading vortex with the orifice rim. A successful deflagration is seen to develop for values of the orifice radius larger than a critical value a c in the order of the flame thickness of the planar deflagration δL. Introduction of appropriate scales provides the dimensionless formulation of the problem, with flame initiation characterised in terms of a critical Damköhler number Δc=(a d/δL)2, whose parametric dependence is investigated. The numerical computations reveal that, while the jet Reynolds number exerts a limited influence on the criticality conditions, the effect of the reactant diffusivity on ignition is much more pronounced, with the value of Δc increasing significantly with increasing Lewis numbers. The reactant diffusivity affects also the way ignition takes place, so that for reactants with the flame develops as a result of ignition in the annular mixing layer surrounding the developing jet stem, whereas for highly diffusive reactants with Lewis numbers sufficiently smaller than unity combustion is initiated in the mixed core formed around the starting vortex. The analysis provides increased understanding of deflagration initiation processes, including the effects of differential diffusion, and points to the need for further investigations corporating detailed chemistry models for specific fuel–air mixtures.

Exact and analytic-numerical solutions of lagging models of heat transfer in a semi-infinite medium

Different non-Fourier models of heat conduction have been considered in recent years, in a growing area of applications, to model microscale and ultrafast, transient, nonequilibrium responses in heat and mass transfer. In this work, using Fourier transforms, we obtain exact solutions for different lagging models of heat conduction in a semi-infinite domain, which allow the construction of analytic-numerical solutions with prescribed accuracy. Examples of numerical computations, comparing the properties of the models considered, are presented.

Difference schemes for numerical solutions of lagging models of heat conduction

Non-Fourier models of heat conduction are increasingly being considered in the modeling of microscale heat transfer in engineering and biomedical heat transfer problems. The dual-phase-lagging model, incorporating time lags in the heat flux and the temperature gradient, and some of its particular cases and approximations, result in heat conduction modeling equations in the form of delayed or hyperbolic partial differential equations. In this work, the application of difference schemes for the numerical solution of lagging models of heat conduction is considered. Numerical schemes for some DPL approximations are developed, characterizing their properties of convergence and stability. Examples of numerical computations are included.

Computer-Assisted Proofs and Symbolic Computations

We discuss some main points of computer-assisted proofs based on reliable numerical computations. Such so-called self-validating numerical methods in combination with exact symbolic manipulations result in very powerful mathematical software tools. These tools allow proving mathematical statements (existence of a fixed point, of a solution of an ODE, of a zero of a continuous function, of a global minimum within a given range, etc.) using a digital computer. To validate the assertions of the underlying theorems fast finite precision arithmetic is used. The results are absolutely rigorous. To demonstrate the power of reliable symbolic-numeric computations we investigate in some details the verification of very long periodic orbits of chaotic dynamical systems. The verification is done directly in Maple, e.g. using the Maple Power Tool intpakX or, more efficiently, using the C++ class library C-XSC.

Numerical Issues in Using MATLAB

Михаил Константинов, Весела Пашева, Петко Петков - Разгледани са някои числени проблеми при използването на компютърната система MATLAB в учебната дейност: пресмятане на тригонометрични функции, повдигане на матрица на степен, спектрален анализ на целочислени матрици от нисък ред и пресмятане на корените на алгебрични уравнения. Причините за възникналите числени трудности могат да се обяснят с особеностите на използваната двоичната аритметика с плаваща точка.

Comparison between Numerical and Simulation Methods for Age-dependent Branching Models with Immigration

2000 Mathematics Subject Classification: primary: 60J80, 60J85, secondary: 62M09, 92D40

Experimental and numerical studies of hydrodynamic interaction of two bodies in waves

This thesis focuses on experimental and numerical studies of the hydrodynamic interaction between two vessels in close proximity in waves. In the model tests, two identical box-like models with round corners were used. Regular waves with the same wave steepness and different wave frequencies were generated. Six degrees of freedom body motions and wave elevations between bodies were measured in a head sea condition. Three initial gap widths were examined. In the numerical computations, a panel-free method based seakeeping program, MAPS0, and a panel method based program, WAMIT, were used for the prediction of body motions and wave elevations. The computed body motions and wave elevations were compared with experimental data.

Analysis of small signal stability margins using genetic optimization

Power system small signal stability analysis aims to explore different small signal stability conditions and controls, namely: (1) exploring the power system security domains and boundaries in the space of power system parameters of interest, including load flow feasibility, saddle node and Hopf bifurcation ones; (2) finding the maximum and minimum damping conditions; and (3) determining control actions to provide and increase small signal stability. These problems are presented in this paper as different modifications of a general optimization to a minimum/maximum, depending on the initial guesses of variables and numerical methods used. In the considered problems, all the extreme points are of interest. Additionally, there are difficulties with finding the derivatives of the objective functions with respect to parameters. Numerical computations of derivatives in traditional optimization procedures are time consuming. In this paper, we propose a new black-box genetic optimization technique for comprehensive small signal stability analysis, which can effectively cope with highly nonlinear objective functions with multiple minima and maxima, and derivatives that can not be expressed analytically. The optimization result can then be used to provide such important information such as system optimal control decision making, assessment of the maximum network's transmission capacity, etc. (C) 1998 Elsevier Science S.A. All rights reserved.

Computation of the linear elastic properties of random porous materials with a wide variety of microstructure

A finite-element method is used to study the elastic properties of random three-dimensional porous materials with highly interconnected pores. We show that Young's modulus, E, is practically independent of Poisson's ratio of the solid phase, nu(s), over the entire solid fraction range, and Poisson's ratio, nu, becomes independent of nu(s) as the percolation threshold is approached. We represent this behaviour of nu in a flow diagram. This interesting but approximate behaviour is very similar to the exactly known behaviour in two-dimensional porous materials. In addition, the behaviour of nu versus nu(s) appears to imply that information in the dilute porosity limit can affect behaviour in the percolation threshold limit. We summarize the finite-element results in terms of simple structure-property relations, instead of tables of data, to make it easier to apply the computational results. Without using accurate numerical computations, one is limited to various effective medium theories and rigorous approximations like bounds and expansions. The accuracy of these equations is unknown for general porous media. To verify a particular theory it is important to check that it predicts both isotropic elastic moduli, i.e. prediction of Young's modulus alone is necessary but not sufficient. The subtleties of Poisson's ratio behaviour actually provide a very effective method for showing differences between the theories and demonstrating their ranges of validity. We find that for moderate- to high-porosity materials, none of the analytical theories is accurate and, at present, numerical techniques must be relied upon.

Modelling of Lamb waves in composite laminated plates excited by interdigital transducers

The technique of permanently attaching interdigital transducers (IDT) to either flat or curved structural surfaces to excite single Lamb wave mode has demonstrated great potential for quantitative non-destructive evaluation and smart materials design, In this paper, the acoustic wave field in a composite laminated plate excited by an IDT is investigated. On the basis of discrete layer theory and a multiple integral transform method, an analytical-numerical approach is developed to evaluate the surface velocity response of the plate due to the IDTs excitation. In this approach, the frequency spectrum and wave number spectrum of the output of IDT are obtained directly. The corresponding time domain results are calculated by applying a standard inverse fast Fourier transformation technique. Numerical examples are presented to validate the developed method and show the ability of mode selection and isolation. A new effective way of transfer function estimation and interpretation is presented by considering the input wave number spectrum in addition to the commonly used input frequency spectrum. The new approach enables the simple physical evaluation of the influences of IDT geometrical features such as electrode finger widths and overall dimension and excitation signal properties on the input-output characteristics of IDT. Finally, considering the convenience of Mindlin plate wave theory in numerical computations as well as theoretical analysis, the validity is examined of using this approximate theory to design IDT for the excitation of the first and second anti-symmetric Lamb modes. (C) 2002 Elsevier Science Ltd. All rights reserved.

Three-dimensional supersonic flow over a spike-nosed body of revolution

The unsteady, viscous, supersonic flow over a spike-nosed body of revolution is numerically investigated by solving the Navier-Stokes equations. The time-accurate computations are performed employing an implicit algorithm based on the second-order time-accurate LU-SGS scheme with the incorporation of a subiteration procedure to maintain time accuracy. The characteristics of the flow field for a Mach number of 3.0, Reynolds number of 7.87 x 10(6)/m, and angles of attack of 5 and 10 degrees are described. Self-sustained asymmetric shock wave oscillations were observed in the numerical computations for these angles of attack. The main characteristic of the flow field, as well as its influence on drag coefficient is discussed.

Représentation et détection des images et des surfaces déformables

La représentation d'une surface, son lissage et son utilisation pour l'identification, la comparaison, la classification, et l'étude des variations de volume, de courbure ou de topologie sont omniprésentes dans l'aire de la numérisation. Parmi les méthodes mathématiques, nous avons retenu les transformations difféomorphiques d'un pattern de référence. Il y a un grand intérêt théorique et numérique à approcher un difféomorphisme arbitraire par des difféomorphismes engendrés par des champs de vitesses. Sur le plan théorique la question est : "est-ce que le sous-groupe de difféomorphismes engendrés par des champs de vitesses est dense dans le groupe plus large de Micheletti pour la métrique de Courant ?" Malgré quelques progrès réalisés ici, cette question demeure ouverte. Les pistes empruntées ont alors convergé vers le sous-groupe de Azencott et de Trouvé et sa métrique dans le cadre de l'imagerie. Elle correspond à une notion de géodésique entre deux difféomorphismes dans leur sous-groupe. L'optimisation est utilisée pour obtenir un système d'équations état adjoint caractérisant la solution optimale du problème d'identification à partir des observations. Cette approche est adaptée à l'identification de surfaces obtenues par un numériseur tel que, par exemple, le scan d'un visage. Ce problème est beaucoup plus difficile que celui d'imagerie. On doit alors introduire un système de référence courbe et une surface à facettes pour les calculs. On donne la formulation du problème d'identification et du calcul du changement de volume par rapport à un scan de référence.