Downloads & Free Reading Options - Results

Thesis (M.A.)--Boston University, 1944

We study the numerical computation of an expectation of a bounded function with respect to a measure given by a non-normalized density on a convex body. We assume that the density is log-concave, satisfies a variability condition and is not too narrow. We consider general convex bodies or even the whole $\R^d$ and show that the integration problem satisfies a refined form of tractability. The main tools are the hit-and-run algorithm and an error bound of a multi run Markov chain Monte Carlo method.

In this paper we show two examples of numerical orbital integrations (Planar Circular Restricted Three Body Problem) in which even though the conservation of Jacobi's constant is near to 1 part in 10e8, the integration proves to be wrong. That is, for some particular cases the Jacobi's constant is insensitive to very important qualitative (not quantitative) changes in the integrated orbit which are produced by numerical errors. We provide specific recomendations to test and ensure that a numerical code is working properly, regardless the numerical method employed.

Bell System Technical Journal, 47: 4. April 1968 pp 511-527. Numerical Integration of Systems of Stiff Nonlinear Differential Equations. (Sandberg, I.W. ; Shichman, H.)

We present a fourth order convergent (2+1) numerical code to solve the Teukolsky equation in the time domain. Our approach is to rewrite the Teukolsky equation as a system of first order differential equations. In this way we get a system that has the form of an advection equation. This is used in combination with a series expansion of the solution in powers of time. To obtain a fourth order scheme we kept terms up to fourth derivative in time and use the advection-like system of differential equations to substitute the temporal derivatives by spatial derivatives. A local stability study leads to a Courant factor of 1.5 for the nonrotating case. This scheme is used to evolve gravitational perturbations in Schwarzschild and Kerr backgrounds. Our numerical method proved to be fourth order convergent in r* and theta directions. The correct power-law tail, ~1/t^{2\ell+3}, for general initial data, and ~1/t^{2\ell+4}, for time symmetric data, was found in the simulations where the duration in time of the tail phase was long enough. We verified that it is crucial to resolve accurately the angular dependence of the mode at late times in order to obtain these values of the exponents in the power-law decay. In other cases, when the decay was too fast and round-off error was reached before a tail was developed, the quasinormal modes frequencies provided a test to determine the validity of our code.

We present a fourth order convergent (2+1) numerical code to solve the Teukolsky equation in the time domain. Our approach is to rewrite the Teukolsky equation as a system of first order differential equations. In this way we get a system that has the form of an advection equation. This is used in combination with a series expansion of the solution in powers of time. To obtain a fourth order scheme we kept terms up to fourth derivative in time and use the advection-like system of differential equations to substitute the temporal derivatives by spatial derivatives. A local stability study leads to a Courant factor of 1.5 for the nonrotating case. This scheme is used to evolve gravitational perturbations in Schwarzschild and Kerr backgrounds. Our numerical method proved to be fourth order convergent in r* and theta directions. The correct power-law tail, ~1/t^{2\ell+3}, for general initial data, and ~1/t^{2\ell+4}, for time symmetric data, was found in the simulations where the duration in time of the tail phase was long enough. We verified that it is crucial to resolve accurately the angular dependence of the mode at late times in order to obtain these values of the exponents in the power-law decay. In other cases, when the decay was too fast and round-off error was reached before a tail was developed, the quasinormal modes frequencies provided a test to determine the validity of our code.

We present a fourth order convergent (2+1) numerical code to solve the Teukolsky equation in the time domain. Our approach is to rewrite the Teukolsky equation as a system of first order differential equations. In this way we get a system that has the form of an advection equation. This is used in combination with a series expansion of the solution in powers of time. To obtain a fourth order scheme we kept terms up to fourth derivative in time and use the advection-like system of differential equations to substitute the temporal derivatives by spatial derivatives. A local stability study leads to a Courant factor of 1.5 for the nonrotating case. This scheme is used to evolve gravitational perturbations in Schwarzschild and Kerr backgrounds. Our numerical method proved to be fourth order convergent in r* and theta directions. The correct power-law tail, ~1/t^{2\ell+3}, for general initial data, and ~1/t^{2\ell+4}, for time symmetric data, was found in the simulations where the duration in time of the tail phase was long enough. We verified that it is crucial to resolve accurately the angular dependence of the mode at late times in order to obtain these values of the exponents in the power-law decay. In other cases, when the decay was too fast and round-off error was reached before a tail was developed, the quasinormal modes frequencies provided a test to determine the validity of our code.

An indexed-based object recognition system using geometric invariance techniques has been designed, and used to recognize buildings in an image of a military site and for recognizing curved planar objects including gasless. New invariants and indexing techniques for polyhedral and curved objects with repetition or bilateral symmetry and objects with the imaged outline of a surface of revolution have been developed. A method to distinguish projectively equivalent but Euclidean distinct objects in an uncalibrated view has been investigated. A group-theoretic framework for relating quasi-invariants to invariants has been formulated. Computing invariants can be formulated as an algebraic manipulation problem involving variable elimination and solving nonlinear polynomial equations. Based on Dixon's formulation of resultants, new methods for eliminating variables have been developed and implemented. These methods are much faster and superior than other elimination techniques. A branch and prune approach for numerically solving polynomial equations has been developed. A simple algorithm for separating invariant relations among object and image features to compute invariants of object features has been designed. These algorithms can serve as a basis for building an invariant work-bench that would enable researchers to experiment with geometric configurations and investigate their geometric invariants.

An indexed-based object recognition system using geometric invariance techniques has been designed, and used to recognize buildings in an image of a military site and for recognizing curved planar objects including gasless. New invariants and indexing techniques for polyhedral and curved objects with repetition or bilateral symmetry and objects with the imaged outline of a surface of revolution have been developed. A method to distinguish projectively equivalent but Euclidean distinct objects in an uncalibrated view has been investigated. A group-theoretic framework for relating quasi-invariants to invariants has been formulated. Computing invariants can be formulated as an algebraic manipulation problem involving variable elimination and solving nonlinear polynomial equations. Based on Dixon's formulation of resultants, new methods for eliminating variables have been developed and implemented. These methods are much faster and superior than other elimination techniques. A branch and prune approach for numerically solving polynomial equations has been developed. A simple algorithm for separating invariant relations among object and image features to compute invariants of object features has been designed. These algorithms can serve as a basis for building an invariant work-bench that would enable researchers to experiment with geometric configurations and investigate their geometric invariants.

An indexed-based object recognition system using geometric invariance techniques has been designed, and used to recognize buildings in an image of a military site and for recognizing curved planar objects including gasless. New invariants and indexing techniques for polyhedral and curved objects with repetition or bilateral symmetry and objects with the imaged outline of a surface of revolution have been developed. A method to distinguish projectively equivalent but Euclidean distinct objects in an uncalibrated view has been investigated. A group-theoretic framework for relating quasi-invariants to invariants has been formulated. Computing invariants can be formulated as an algebraic manipulation problem involving variable elimination and solving nonlinear polynomial equations. Based on Dixon's formulation of resultants, new methods for eliminating variables have been developed and implemented. These methods are much faster and superior than other elimination techniques. A branch and prune approach for numerically solving polynomial equations has been developed. A simple algorithm for separating invariant relations among object and image features to compute invariants of object features has been designed. These algorithms can serve as a basis for building an invariant work-bench that would enable researchers to experiment with geometric configurations and investigate their geometric invariants.

Journal of Research of the National Bureau of Standards

Classical asymptotic analysis of ordinary differential equations derives approximate solutions that are numerically stable. However, the analysis also leads to tedious expansions in powers of the relevant parameter for a particular problem. The expansions are replaced with integrals that can be evaluated by numerical integration. The resulting numerical solutions retain the linear independence that is the main advantage of asymptotic solutions. Examples, including the Falkner-Skan equation from laminar boundary layer theory, illustrate the method of asymptotic analysis with numerical integration.

Operational unification of finite difference methods for numerical integration of ordinary differential equations

AN EXPOSITION OF PROCEDURES IS PRESENTED FOR APPROXIMATION OF PARTICULAR SOLUTIONS TO LINEAR AND NONLINEAR DIFFERENTIAL EQUATIONS (OR SYSTEMS OF EQUATIONS) BY MEANS OF DIFFERENCE EQUATIONS. THE RESULTS OF THE CLASSICAL AND MODIFIED PROCEDURES ARE COMPARED FOR SPECIFIC EXAMPLES. QUESTIONS OF CONVERGENCE, STABILITY (RELATIVE AND ABSOLUTE), CONSISTENCY, AND OPTIMALITY ARE TREATED.

The numerical integration technique to utilized in the reduction and analysis of data gathered in the USAF Aeroballistic Research Facility located at Eglin Air Force Base is described. The method of Chapman and Kirk has been developed into a system of digital computer programs which may be easily and routinely used in the analysis of data derived from spark range firing. The equations of motion for a missile/projectile in free-flight are derived in this report utilizing the six degrees of freedom. The parametric equations, philosophy, and methods of implementation are also derived and discussed. Brief descriptions of the computer program involved are also given.

The objective of the present research program is the development of fast numerical multidimensional Fourier techniques applied to a wide range of wave modeling and analysis problems. Important progress made in the past year has been the development of new algorithms for multidimensional Fourier analysis. These algorithms are the key to future hyperfast applications of the method.

In contrast to the prevailing view in the literature, it is shown that even extremely stiff sets of ordinary differential equations may be solved efficiently by explicit methods if limiting algebraic solutions are used to stabilize the numerical integration. The stabilizing algebra differs essentially for systems well-removed from equilibrium and those near equilibrium. Explicit asymptotic and quasi-steady-state methods that are appropriate when the system is only weakly equilibrated are examined first. These methods are then extended to the case of close approach to equilibrium through a new implementation of partial equilibrium approximations. Using stringent tests with astrophysical thermonuclear networks, evidence is provided that these methods can deal with the stiffest networks, even in the approach to equilibrium, with accuracy and integration timestepping comparable to that of implicit methods. Because explicit methods can execute a timestep faster and scale more favorably with network size than implicit algorithms, our results suggest that algebraically-stabilized explicit methods might enable integration of larger reaction networks coupled to fluid dynamics than has been feasible previously for a variety of disciplines.

Simulations of biophysical systems inevitably include steps that correspond to time integrations of ordinary differential equations. These equations are often related to enzyme action in the synthesis and destruction of molecular species, and in the regulation of transport of molecules into and out of the cell or cellular compartments. Enzyme action is almost invariably modeled with the quasi-steady-state Michaelis-Menten formula or its close relative, the Hill formula: this description leads to systems of equations that may be stiff and hard to integrate, and poses unusual computational challenges in simulations where a smooth evolution is interrupted by the discrete events that mark the cells' lives. This is the case of a numerical model (Virtual Biophysics Lab - VBL) that we are developing to simulate the growth of three-dimensional tumor cell aggregates (spheroids). The program must be robust and stable, and must be able to accept frequent changes in the underlying theoretical model: here we study the applicability of known integration methods to this unusual context and we describe the results of numerical tests in situations similar to those found in actual simulations.

In this paper we develop a family of three 8-step methods, optimized for the numerical integration of oscillatory ordinary differential equations. We have nullified the phase-lag of the methods and the first r derivatives, where r=1,2,3. We show that with this new technique, the method gains efficiency with each derivative of the phase-lag nullified. This is the case for the integration of both the Schrodinger equation and the N-body problem. A local truncation error analysis is performed, which, for the case of the Schrodinger equation, also shows the connection of the error and the energy, revealing the importance of the zero phase-lag derivatives. Also the stability analysis shows that the methods with more derivatives vanished, have a bigger interval of periodicity.

A finite element model of a rotor bearing system was analyzed to determine the stability limits of the forward, backward, and centered Euler; Runge-Kutta; Milne; and Adams numerical integration techniques. The analysis concludes that the highest frequency mode determines the maximum time step for a stable solution. Thus, the number of mass elements should be minimized. Increasing the damping can sometimes cause numerical instability. For a uniform shaft, with 10 mass elements, operating at approximately the first critical speed, the maximum time step for the Runge-Kutta, Milne, and Adams methods is that which corresponds to approximately 1 degree of shaft movement. This is independent of rotor dimensions.

We implement and investigate the numerical properties of a new family of integrators connecting both variants of the symplectic Euler schemes, and including an alternative to the classical symplectic mid-point scheme, with some additional terms. This family is derived from a new method, introduced in a previous study, for generating symplectic integrators based on the concept of special symplectic manifold. The use of symplectic rotations and a particular type of projection keeps the whole procedure within the symplectic framework. We show that it is possible to define a set of parameters that control the additional terms providing a way of "tuning" these new symplectic schemes. We test the "tuned" symplectic integrators with the perturbed pendulum and we compare its behavior with an explicit scheme for perturbed systems. Remarkably, for the given examples, the error in the energy integral can be reduced considerably. There is a natural geometrical explanation, sketched at the end of this paper. This is the subject of a parallel article where a finer analysis is performed. Numerical results obtained in this paper open a new point of view on symplectic integrators and Hamiltonian error.

Analytical Proof of the Taylor Equation Including Taylor's Constant Sγ Which Previously Required Numerical Integration, with Applications Backup of http://vixra.org/abs/1003.0259 dated 28 Mar 2010 Authors: Nigel B. Cook British mathematician Sir Geoffrey I. Taylor in secret work for British civil defence in 1941 (declassified in 1950 and published in the Proceedings of the Royal Society, vol. 201A, pp. 159-186), derived the strong shock solution equation, namely distance, (equation) , where (equation) is the ambient (pre-shock) atmospheric density, t is time after explosion, E is the energy released and Sg is Taylor's calculated function of g, requiring a complex step-wise numerical integration. We present a proof of the equation (equation), implying that Taylor's so-called constant (equation), not requiring any complex integration. This is useful for close-in shock waves from nuclear explosions and supernovae explosions. We further obtain the general arrival time of the shock wave (equation), by noting two asymptotic solutions; namely, at very great distances, the blast decays into a sound wave so the arrival time t approaches the ratio of distance to sound velocity (equation), while at very close-in distances the strong shock equation previously derived becomes accurate, and there is also an easily included effect at intermediate distances due to the expansion of the hot air in reducing shock front arrival times. The errors of method made by Taylor for nuclear test explosions in air were also made by Russian mathematician Leonid I. Sedov who applied similar cumbersome numerical integrations in a 1946 paper (published in the Journal of Applied Mathematics and Mechanics, vol. 10, pp. 241-50). Please see paper for equations in abstract. See also: The effects of the atomic bomb on Hiroshima, Japan (the secret U.S. Strategic Bombing Survey report 92, Pacific Theatre) located at: http://archive.org/details/TheEffectsOfTheAtomicBombOnHiroshima

14

In the calculation of cross sections for infrared-safe observables in high energy collisions at next-to-leading order, one approach is to perform all of the integrations, including the virtual loop integration numerically. One would use a subtraction scheme that removes infrared and collinear divergences from the integrand in a style similar to that used for real emission graphs. Then one would perform the loop integration by Monte Carlo integration along with the integrations over final state momenta. In this paper, we have explored how one can perform the numerical integration. We have studied the N-photon scattering amplitude with a massless electron loop in order to have a case with a singular integrand that is not, however, so singular as to require the subtractions. We report results for N = 4, N = 5 with left-handed couplings, and N=6.

We explore the combination of deterministic and Monte Carlo methods to facilitate efficient automatic numerical computation of multidimensional integrals with singular integrands. Two adaptive algorithms are presented that employ recursion and are runtime and memory optimised, respectively. SINGINT, a C implementation of the algorithms, is introduced and its utilisation in the calculation of particle scattering amplitudes is exemplified.

A new scheme for numerical integration of motion for classical systems composed of rigid polyatomic molecules is proposed. The scheme is based on a matrix representation of the rotational degrees of freedom. The equations of motion are integrated within the Verlet framework in velocity form. It is shown that, contrary to previous methods, in the approach introduced the rigidity of molecules can be conserved automatically without any additional transformations. A comparison of various techniques with respect to numerical stability is made.

Taking a novel approach, this paper discuss the structure of compact stars, an important topic in theoretical astrophysics. Adopting the Newtonian gravitation, we solve the hydrostatic equilibrium equation by imposing a simple parametrization for the mass density inside the star. The solutions of the equilibrium equation are carried out without numerical integration, with the aim of determining few global properties of white dwarfs and neutron stars. The global properties of compact stars are thus provided by simple algebraic relationships. The model is intended as an introductory approach to the study of compact stars at the undergraduate or graduate levels.

We study numerical integration on the unit sphere $\mathbb{S}^2 \subset \mathbb{R}^3$ using equal weight quadrature rules, where the weights are such that constant functions are integrated exactly. The quadrature points are constructed by lifting a $(0,m,2)$-net given in the unit square $[0,1]^2$ to the sphere $\mathbb{S}^2$ by means of an area preserving map. A similar approach has previously been suggested by Cui and Freeden [SIAM J. Sci. Comput. 18 (1997), no. 2]. We prove three results. The first one is that the construction is (almost) optimal with respect to discrepancies based on spherical rectangles. Further we prove that the point set is asymptotically uniformly distributed on $\mathbb{S}^2$. And finally, we prove an upper bound on the spherical cap $L_2$-discrepancy of order $N^{-1/2} (\log N)^{1/2}$ (where $N$ denotes the number of points). This slightly improves upon the bound on the spherical cap $L_2$-discrepancy of the construction by Lubotzky, Phillips and Sarnak [Comm. Pure Appl. Math. 39 (1986), 149--186]. Numerical results suggest that the $(0,m,2)$-nets lifted to the sphere $\mathbb{S}^2$ have spherical cap $L_2$-discrepancy converging with the optimal order of $N^{-3/4}$.

A numerical study of hypersonic forebody/inlet integration problem is presented in the form of the view-graphs. The following topics are covered: physical/chemical modeling; solution procedure; flow conditions; mass flow rate at inlet face; heating and skin friction loads; 3-D forebogy/inlet integration model; and sensitivity studies.

"Certain Integration Formulae Useful in Numerical Computation" is an article from The American Mathematical Monthly, Volume 19 . View more articles from The American Mathematical Monthly . View this article on JSTOR . View this article's JSTOR metadata . You may also retrieve all of this items metadata in JSON at the following URL: https://archive.org/metadata/jstor-2972443

The efficiency and accuracy of several algorithms recently developed for the efficient numerical integration of stiff ordinary differential equations are compared. The methods examined include two general-purpose codes, EPISODE and LSODE, and three codes (CHEMEQ, CREK1D, and GCKP84) developed specifically to integrate chemical kinetic rate equations. The codes are applied to two test problems drawn from combustion kinetics. The comparisons show that BODE is the fastest code currently available for the integration of combustion kinetic rate equations. An important finding is that an interactive solution-of the algebraic energy conservation equation to compute the temperature does not result in significant errors. In addition, this method can be more efficient than evaluating the temperature by integrating its time derivative. Significant reductions in computational work are realized by updating the rate constants (k = ATN exp(-E/RT)) only when the temperature change exceeds an amount T that is problem dependent. An approximate expression for the automatic evaluation of T is derived and is shown to result in increased efficiency.

We study numerically classical 1-dimensional Hamiltonian lattices involving inter-particle long range interactions that decay with distance like 1/r^alpha, for alpha>=0. We demonstrate that although such systems are generally characterized by strong chaos, they exhibit an unexpectedly organized behavior when the exponent alpha

We study numerically classical 1-dimensional Hamiltonian lattices involving inter-particle long range interactions that decay with distance like 1/r^alpha, for alpha>=0. We demonstrate that although such systems are generally characterized by strong chaos, they exhibit an unexpectedly organized behavior when the exponent alpha

NAS 1.60:1092 Digitized from IA1177307-10-0084 .

We consider the radiation transfer problem in the discrete-ordinate, plane-parallel approach. We introduce two benchmark problems with exact known solutions and show that for strongly non-homogeneous media the homogeneous layers approximation can lead to errors of 10% in the estimation of the intensity. We propose and validate a general purpose numerical method that transforming the two-boundary problem into an initial boundary problem, using an adaptative step integration and an interpolation of the local optical properties, can improve the accuracy of the solution up to two orders of magnitude. This is furthermore of interest for practical applications, such as atmospheric radiation transfer, where the scattering and absorbing properties of the media vary strongly with height and are only known, based on measurements or models, at certain discrete points.

39 p. 28 cm

39 p. 28 cm

39 p. 28 cm

39 p. 28 cm

39 p. 28 cm

Mathematics 128A, 001 - Spring 2014 Creative Commons 3.0: Attribution-NonCommercial-NoDerivs

We provide a comprehensive survey of splitting and composition methods for the numerical integration of ordinary differential equations (ODEs). Splitting methods constitute an appropriate choice when the vector field associated with the ODE can be decomposed into several pieces and each of them is integrable. This class of integrators are explicit, simple to implement and preserve structural properties of the system. In consequence, they are specially useful in geometric numerical integration. In addition, the numerical solution obtained by splitting schemes can be seen as the exact solution to a perturbed system of ODEs possessing the same geometric properties as the original system. This backward error interpretation has direct implications for the qualitative behavior of the numerical solution as well as for the error propagation along time. Closely connected with splitting integrators are composition methods. We analyze the order conditions required by a method to achieve a given order and summarize the different families of schemes one can find in the literature. Finally, we illustrate the main features of splitting and composition methods on several numerical examples arising from applications.

This paper presents an analytical model and a geometric numerical integrator for a tethered spacecraft model that is composed of two rigid bodies connected by an elastic tether. This model includes important dynamic characteristics of tethered spacecraft in orbit, namely the nonlinear coupling between deformations of tether, rotational dynamics of rigid bodies, a reeling mechanism, and orbital dynamics. A geometric numerical integrator, referred to as a Lie group variational integrator, is developed to numerically preserve the Hamiltonian structure of the presented model and its Lie group configuration manifold. The structure-preserving properties are particularly useful for studying complex dynamics of a tethered spacecraft over a long period of time. These properties are illustrated by numerical simulations.

We present a comparison of different numerical techniques for the integration of variational equations. The methods presented can be applied to any autonomous Hamiltonian system whose kinetic energy is quadratic in the generalized momenta, and whose potential is a function of the generalized positions. We apply the various techniques to the well-known H\'enon-Heiles system, and use the Smaller Alignment Index (SALI) method of chaos detection to evaluate the percentage of its chaotic orbits. The accuracy and the speed of the integration schemes in evaluating this percentage are used to investigate the numerical efficiency of the various techniques.

We prove the curse of dimensionality in the worst case setting for numerical integration for a number of classes of smooth $d$-variate functions. Roughly speaking, we consider different bounds for the derivatives of $f \in C^k(D_d)$ and ask whether the curse of dimensionality holds for the respective classes of functions. We always assume that $D_d \subset \R^d$ has volume one and consider various values of $k$ including the case $k=\infty$ which corresponds to infinitely many differentiable functions. We obtain necessary and sufficient conditions, and in some cases a full characterization for the curse of dimensionality. For infinitely many differentiable functions we prove the curse if the bounds on the successive derivatives are appropriately large. The proof technique is based on a volume estimate of a neighborhood of the convex hull of $n$ points which decays exponentially fast if $n$ is small relative to $d$. For $k=\infty$, we also also study conditions for quasi-polynomial, weak and uniform weak tractability.

Higher order digital nets have recently been recognized as one of the most promising branches of quasi-Monte Carlo methods. The notable feature of higher order digital nets is that they can exploit the smoothness of a function for numerical integration and achieve an improved convergence rate of the integration error for smooth functions. One prominent construction of such nets is based on digitally interlacing the components of the classical digital net whose number of components is a multiple $d$ of the dimension. In this study we consider the weighted unanchored Sobolev spaces of smoothness $\alpha \ge 2$, and derive an upper bound on the mean square worst-case error for digitally shifted higher order digital nets. Employing our obtained bound as a quality criterion, we prove that the component-by-component construction can be made efficient use of to obtain good polynomial lattice point sets that are used for interlaced components. Through this approach we are able to get some tractability results under certain conditions on the weights. Fast construction using the fast Fourier transform requires the construction cost of order $O(sdN \log N)$ operations using O(N) memory, where $N$ is the number of points and $s$ is the dimension, which is a significant reduction in the construction cost as compared to higher order polynomial lattice rules. Numerical experiments confirm that the performance of our constructed point sets often outperforms in terms of our introduced quality criterion the performances of higher order digital nets with Sobol' sequences and Niederreiter-Xing sequences used as interlaced components, indicating the usefulness of our algorithm.

The integration of the equations of motion in gravitational dynamical systems -- either in our Solar System or for extra-solar planetary system -- being non integrable in the global case, is usually performed by means of numerical integration. Among the different numerical techniques available for solving ordinary differential equations, the numerical integration using Lie series has shown some advantages. In its original form (Hanslmeier 1984), it was limited to the N-body problem where only gravitational interactions are taken into account. We present in this paper a generalisation of the method by deriving an expression of the Lie-terms when other major forces are considered. As a matter of fact, previous studies had been made but only for objects moving under gravitational attraction. If other perturbations are added, the Lie integrator has to be re-built. In the present work we consider two cases involving position and position-velocity dependent perturbations: relativistic acceleration in the framework of General Relativity and a simplified force for the Yarkovsky effect. A general iteration procedure is applied to derive the Lie series to any order and precision. We then give an application to the integration of the equation of motions for typical Near-Earth objects and planet Mercury.

We find estimates for the error in replacing an integral $\int f d\mu$ with respect to a fractal measure $\mu$ with a discrete sum $\sum_{x \in E} w(x) f(x)$ over a given sample set $E$ with weights $w$. Our model is the classical Koksma-Hlawka theorem for integrals over rectangles, where the error is estimated by a product of a discrepancy that depends only on the geometry of the sample set and weights, and variance that depends only on the smoothness of $f$. We deal with p.c.f self-similar fractals, on which Kigami has constructed notions of energy and Laplacian. We develop generic results where we take the variance to be either the energy of $f$ or the $L^1$ norm of $\Delta f$, and we show how to find the corresponding discrepancies for each variance. We work out the details for a number of interesting examples of sample sets for the Sierpi\'{n}ski gasket, both for the standard self-similar measure and energy measures, and for other fractals.

We examine the effect of numerical integration on the convergence of high order pyramidal finite element methods. Rational functions are indispensable to the construction of pyramidal interpolants so the conventional treatment of numerical integration, which requires that the finite element approximation space is piecewise polynomial, cannot be applied. We develop an analysis that allows the finite element approximation space to include rational functions and show that despite this complication, conventional rules of thumb can still be used to select appropriate quadrature methods on pyramids. Along the way, we present a new family of high order pyramidal finite elements for each of the spaces of the de Rham complex.

We examine the effect of numerical integration on the convergence of high order pyramidal finite element methods. Rational functions are indispensable to the construction of pyramidal interpolants so the conventional treatment of numerical integration, which requires that the finite element approximation space is piecewise polynomial, cannot be applied. We develop an analysis that allows the finite element approximation space to include rational functions and show that despite this complication, conventional rules of thumb can still be used to select appropriate quadrature methods on pyramids. Along the way, we present a new family of high order pyramidal finite elements for each of the spaces of the de Rham complex.