SEARCH · Searcharxiv
Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
In principle, all the natural systems such as biological, ecological and economical systems are structure-variable systems (in which some environment parameters are not fixed). In this Letter we show that data sequences from many structure-variable systems are short-term predictable. We also argue how to characterize the data sequences from structure-variable systems.
We consider the problem of reconstructing a function given its values on a set of points with finite density. We prove that with probability one, the values of an almost periodic function on a random array of points (with finite density) completely determine the function. We also give some properties of the associated Blaschke product.
In this paper, we study the two dimensional lattice Boltzmann BGK model (LBGK) by analytically solving a simple flow in a 2~-D channel. The flow is driven by the movement of upper boundary with vertical injection fluid at the porous boundaries. The velocity profile is shown to satisfy a second-order finite-difference form of the simplified incompressible Navier-Stokes equation. With the analysis, different boundary conditions can be studied theoretically. A momentum exchange principle is also revealed at the boundaries. A general boundary condition for any given velocity boundary is proposed based on the analysis.
A method for estimating theoretical predictability of time series is presented, based on information-theoretic functionals---redundancies and surrogate data technique. The redundancy, designed for a chosen model and a prediction horizon, evaluates amount of information between a model input (e.g., lagged versions of the series) and a model output (i.e., a series lagged by the prediction horizon from the model input) in number of bits. This value, however, is influenced by a method and precision of redundancy estimation and therefore it is a) normalized by maximum possible redundancy (given by the precision used), and b) compared to the redundancies obtained from two types of the surrogate data in order to obtain reliable classification of a series as either unpredictable or predictable. The type of predictability (linear or nonlinear) and its level can be further evaluated. The method is demonstrated using a numerically generated time series as well as high-frequency foreign exchange data and the theoretical predictability is compared to performance of a nonlinear predictor.
Pressure (density) and velocity boundary conditions inside a flow domain are studied for 2-D and 3-D lattice Boltzmann BGK models (LBGK) and new method to specify these conditions are proposed. These conditions are consistent with the boundary condition we proposed in a previous paper using an idea of bounce-back of non-equilibrium distribution. These conditions give excellent results for the regular LBGK models, and were shown to be second-order accurate by numerical examples. When they are used together with the improved incompressible LBGK model proposed by zou et al. the simulation results recover the analytical solution of the plane Poiseuille flow driven by pressure (density) difference with machine accuracy.
Conventional computers are ill suited to run CA models, and so discourage their development. By creating a hardware platform that makes a broad range of new CA algorithms practical for real applications, we hope to whet the appetite of researchers for the astronomical computing power that can be harnessed in microphysics in a CA format.
A general method for testing nonlinearity in time series is described and applied to measurements of different pressure data inside the draft tube surge of a real Francis turbine. Comparing the current original time series to an ensemble of surrogates time series, suitably constructed to mimic the linear properties of the original one, we was able to distinguish a linear stochastic from a nonlinear deterministic behaviour and, moreover, to quantify the degree of nonlinearity present in the related dynamics. The problem of detecting nonlinear structure in real data is quite complicated by the influence of various contaminations, like broadband noise and/or long coherence times. These difficulties have been overcame using the combination of a suitable nonlinear filtering technique and a qualitative redundancy statistic analysis. The above investigations allow a quantitative characterization of different dynamical regimes of motion of gas cavities inside real turbines and, moreover, allow to support the reliability of some related mathematical modelizations.
The method of surrogate data is a tool to test whether data were generated by some class of model. Tests based on the periodogram have been proposed to decide if linear systems driven by Gaussian noise could have generated a sample time series. We show that this procedure based on the periodogram, in general, misspecifies the test statistic. Based on the theory of linear systems we suggest an alternative procedure to obtain the correct distribution of the test statistic and discuss problems of this approach.
In order to investigate the evolutionary process of many deterministic Dynamical systems with unfixed parameter, a set of dynamical models with parameter changing continuously and the accumulation of this change might be large is introduced and discussed. The boundary crises and the period-doubling bifurcations are found suppressed and scaling properties for these phenomena are exploited. Due to this suppression, the period-doubling bifurcations seem no longer continuous. We further discuss the possible applications of these models.
In view of the tremendous computing power jump of modern RISC processors the interest in parallel computing seems to be thinning out. Why use a complicated system of parallel processors, if the problem can be solved by a single powerful micro-chip? It is a general law, however, that exponential growth will always end by some kind of a saturation, and then parallelism will again become a hot topic. We try to prepare ourselves for this eventuality. The MPPC project started in 1990 in the heydeys of parallelism and produced four ASTRA machines (presented at CHEP'92) with 4k processors (which are expandable to 16k) based on `yesterday's chip-technology' (chip presented at CHEP'91). These machines now provide excellent test-beds for algorithmic developments in a complete, real environment. We are developing for example fast-pattern recognition algorithms which could be used in high-energy physics experiments at the LHC (planned to be operational after 2004 at CERN) for triggering and data reduction. The basic feature of our ASP (Associative String Processor) approach is to use extremely simple (thus very cheap) processor elements but in huge quantities (up to millions of processors) connected together by a very simple string-like communication chain. In this paper we present powerful algorithms based on this architecture indicating the performance perspectives if the hardware (i.e. chip fabrication) quality reaches `present or even future technology levels'.
The problem of hypothesis testing is examined from both the historical and Bayesian points of view in the case that sampling is from an underlying joint probability distribution and the hypotheses tested for are those of independence and dependence of the underlying distribution. Exact results for the Bayesian method are provided. Asymptotic Bayesian results and historical method test quantities are compared, and historical method quantities are interpreted in terms of clearly defined Bayesian quantities. The asymptotic Bayesian test relies upon a statistic that is primarily mutual information.
We study an interface-capturing two-phase fluid model in which the interfacial tension is modelled as a volumetric stress. Since these stresses are obtainable from a Van der Waals-Cahn-Hilliard free energy, the model is, to a certain degree, thermodynamically realistic. Thermal fluctuations are not considered presently for reasons of simplicity. The utility of the model lies in its momentum-conservative representation of surface tension and the simplicity of its numerical implementation resulting from the volumetric modelling of the interfacial dynamics. After validation of the model in two spatial dimensions, two prototypical applications---instability of an initially high-Reynolds-number liquid jet in the gaseous phase and spinodal decomposition in a liquid-gas system--- are presented.
The method of generating of high-dimensional oscillations on the basis of summing of low-dimensional chaotic signals of noncoupled dynamical systems is investigated. It is shown that the correlation dimension of attractor reconstructed from such oscillations is equal to the sum of dimensions of the original signals. The possibilities for obtaining of homogeneous high-dimensional chaotic attractor are numerically demonstrated with the original signals of R{ö}ssler system.
Numerical simulations describing plunging breakers including the splash-up phenomenon are presented. The motion is governed by the classical, incompressible, two-dimensional Navier-Stokes equation. The numerical modelling of this two-phase flow is based on a piecewise linear version of the volume of fluid method. Preliminary results concerning the time evolution of the liquid-gas interface and the vorticity field are given for short waves, showing how an initial steep wave undergoes breaking and successive splash-up cycles.
A simple extension of the Lattice Boltzmann equation is proposed, which permits to handle reactive flow dynamics in the limit of fast chemistry at virtually no extra-cost with respect to the purely hydrodynamic scheme.
The growing need for a better understanding of nonlinear processes in plasma physics has in the last decades stimulated the development of new and more advanced data analysis techniques. This review lists some of the basic properties one may wish to infer from a data set and then presents appropriate analysis techniques with some recent applications. The emphasis is put on the investigation of nonlinear wave phenomena and turbulence in space plasmas.
From the striped coats of zebras to the ripples in windblown sand, the natural world abounds with locally banded patterns. Such patterns have been of great interest throughout history, and, in the last twenty years, scientists in a wide variety of fields have been studying the patterns formed in well-controlled experiments that yield enormous quantities of high-precision data. These experiments involving phenomena as diverse as chemical reactions in shallow layers, thermal convection in horizontal fluid layers, periodically shaken layers of sand, and the growth of slime mold colonies often display patterns that appear qualitatively similar. Methods are needed to characterize in a reasonable amount of time the differences and similarities in patterns that develop in different systems, as well as in patterns formed in one system for different experimental conditions. In this Letter, we introduce a novel, fast method for determining local pattern properties such as wavenumber, orientation, and curvature as a function of position for locally striped patterns.
We show that predicting the HPP or FHP III lattice gas for finite time is equivalent to calculating the output of an arbitrary Boolean circuit, and is therefore P-complete: that is, it is just as hard as any other problem solvable by a serial computer in polynomial time. It is widely believed in computer science that there are inherently sequential problems, for which parallel processing gives no significant speedup. Unless this is false, it is impossible even with highly parallel processing to predict lattice gases much faster than by explicit simulation. More precisely, we cannot predict t time-steps of a lattice gas in parallel computation time O(log^k t) for any k, or O(t^α) for α< 1/2, unless the class P is equal to the class NC or SP respectively.