随机化交通灯的二维元胞自动机交通模型

元胞自动机交通模型以简单的规则反映交通系统中的多种因素，可以分析各种交通现象，且可在计算机上方便、高效地运作·Ｂｉｈａｍ－Ｍｉｄｄｌｅｔｏｎ－Ｌｅｖｉｎｅ模型（ＢＭＬ模型）实现了二维交通问题的元胞自动机模型的模拟研究·本文对ＢＭＬ模型作了改进，解除了该模型中关于交通灯同步变化的限制·在新模型中，每个路口的交通灯可以自由选定起始工作时间和变化节奏，于是可以更全面、准确地反映交通灯对交通系统性能的影响·本文还对新模型中出现的若干新效应作了解释·     

Three-dimensional integrable models and associated tangle invariants

In this paper we show that the Boltzmann weights of the three-dimensional Baxter-Bazhanov model give representations of the braid group if some suitable spectral limits are taken. In the trigonometric case we classify all possible spectral limits which produce braid group representations. Furthermore, we prove that for some of them we get cyclotomic invariants of links and for others we obtain tangle invariants generalizing the cyclotomic ones.     

The Modeling of Velocity Enhancement in Polymer Flooding

In single-phase polymer flooding experiments it has repeatedly been observed that the average velocity of the polymer molecules is higher than the average velocity of the water molecules. This effect is incorporated in many conventional Enhanced Oil Recovery (EOR) simulators by the introduction of a constant velocity enhancement factor. In this paper we show that, in absence of dispersion, a constant enhancement factor in the mathematical model for two-phase polymer flow (Buckley--Leverett displacement) leads to ill-posedness of the model equations. We propose a saturation dependent enhancement factor, derived from a model based on percolation concepts, for which this problem does not occur.     

On a Question of Phillips

In [5] Phillips proved that one can obtain the additive group of any nonstandard model *? of the ring ? of integers by using a linear mod 1 function h : F ?, where F is the α-dimensional vector space over ? when α is the cardinality of *?. In this connection it arises the question whether there are linear mod 1 functions which are neither addition nor quasi-linear. We prove that this is the case.     

The Nonexistence of Procedures with Bounded Performance Characteristics in Certain Parametric Inference Problems

This paper gives a condition which implies the nonexistence of parametric statistical procedures with bounded risk or error performance characteristics. Many examples for which such a condition is satisfied are considered.     

产品定级中挑选专家的数学模型

在工农业产品评定级别中，专家组的水平是至关重要的。文中给出一个挑选专家组成员的数学模型，可以检验专家对参评产品的决策水平，为评价所评结论的可信度提供参考依据。     

Shell model interaction between proton holes and between proton holes and neutron particles around 208Pb

An effective residual interaction between particles and holes for shell model calculations around ²⁰⁸Pb, derived from the interaction between free nucleons, is compared with the measured properties of proton-hole neutron states in ²⁰⁸Tl and the interaction between proton holes is adjusted to newly measured level energies in ²⁰⁶Hg. These interaction elements are particularly relevant for neutron-rich nuclei. The adjustment of two mixing elements reproduces the known γ-decay data in ²⁰⁸Tl. Received: 2 April 2002 / Accepted: 2 May 2002     

Linear and Nonlinear Optical Properties Of Racemic (±)2-(α-Methylbenzylamino)-5-Nitropyridine Single Crystals

Linear and nonlinear optical properties of racemic (±)2-(α-methylbenzylamino)-5-nitropyridine ((±)MBANP) single crystals have been comprehensively investigated and compared with those of the enantiomorph (–)2-(α-methylbenzylamino)-5-nitropyridine ((–)MBANP) crystals. (±)MBANP crystal exhibits very high chemical and physical stability, but relatively small nonlinear optical coefficients (d₃₁ = 6.8 pm/V, d₃₂ = 4.7 pm/V, d₃₃ = 0.84 pm/V). A comparison between the nonlinear optical coefficients of (±)MBANP and (–)MBANP demonstrates the validity of the oriented-gas model in molecular crystals that neglects all the contributions from intermolecular interaction.     

无偏的岭回归迭代算法

本文探讨线性模型的无偏的岭回归迭代算法,这种算法保持最小二乘法的性质,当存在较为严重的共线性时,它能给出较为精确的参数及其协差阵的估计值;当存在严格的共线性时,给出参数及其协差阵的无穷多解中的一个,这个解由初值决定。文章还给出了算法的收敛性及一些其它性质的证明。     

Assessing the Error Probability of the Model Selection Test

The asymptotic error probability of Linhart's model selection test isevaluated, and compared with the nominal significance level. We examine thecase where the expected discrepancies of the candidate models from the truemodel are asymptotically equal. The local alternatives method is employed inthe limiting operation of the asymptotic evaluation. Although the errorprobability under the null hypothesis is actually shown to be equal to orless than the level for most situations, intolerable violations of the errorcontrol are observed for nested models: It is often erroneously concludedthat the smaller model is significantly better than the larger model. Toprevent this violation, a modification of Linhart's test statistic isproposed. The effectiveness of the proposed test is confirmed throughtheoretical analysis and numerical simulations.