利用多级变量分型法作濮阳6～8月份暴雨48h预报

暴雨是一种中尺度现象.因其尺度小、突发性强、临界参数不明显,增加了预报的难度.形成暴雨的物理条件虽然只有两个.即强烈垂直上升运动和源源不断的水汽输送,但同时具备这两个条件的天气形势和影响系统确实是非常复杂的.常见的预报方法所用资料都是原始资料,而本方法所用二级变量、三级变量(二级变量是当天24h变高减去前一天24h变高;三级变量是当天二级变量减去前一天二级变量)是原始资料加工出来的.前者是各种系统静态和整体的反映,后者是局部的动态反映.

短期降水预报方法

为了提高短期预报的质量，在总结一些行之有效方法的基础上，建立了以三要素４８小时变量分型制作２４小时降水短期预报的方法．经过一年的实际使用，效果较好。

Oil–water two-phase flow pattern analysis with ERT based measurement and multivariate maximum Lyapunov exponent

Oil–water two-phase flow patterns in a horizontal pipe are analyzed with a 16-electrode electrical resistance tomography(ERT) system. The measurement data of the ERT are treated as a multivariate time-series, thus the information extracted from each electrode represents the local phase distribution and fraction change at that location. The multivariate maximum Lyapunov exponent(MMLE) is extracted from the 16-dimension time-series to demonstrate the change of flow pattern versus the superficial velocity ratio of oil to water. The correlation dimension of the multivariate time-series is further introduced to jointly characterize and finally separate the flow patterns with MMLE. The change of flow patterns with superficial oil velocity at different water superficial velocities is studied with MMLE and correlation dimension, respectively, and the flow pattern transition can also be characterized with these two features. The proposed MMLE and correlation dimension map could effectively separate the flow patterns, thus is an effective tool for flow pattern identification and transition analysis.

A Note on "Doubly Periodic Propagating Wave for (2+1)-Dimensional Breaking Soliton Equation"

In a recent paper [Commun. Theor. Phys. （Beijing, China） 49 （2008） 268], Huang et al. gave a general variable separation solution to the （2＋1）-dimensional breaking soliton equation via a special Biicldund transformation and the variable separation approach. In terms of the derived variable separation solution and by introducing Jacobi elliptic functions, they claimed that nonelastic types of interaction between Jacobi elliptic function waves are investigated both analytically and graphically. We show that some inappropriateness or errors exist in their paper, and say nothing of the conclusion that some nonelastic types of interaction between Jacobi elliptic function waves in the （2＋1）-dimensional breaking soliton equation have been found.

Solitons with Periodic Behavior in Korteweg-de Vries Type Models Related toSchrodinger System

The linear variable separation approach is successfully extended to(1+1)-dimensional Korteweg-de Vries (KdV) type models related to Schrodinger system. Somesignificant types of solitons such as compaction, peakon, and loop solutions with periodic behaviorare simultaneously derived from the (l+l)-dimensional soliton system by entrancing appropriatepiecewise smooth functions and multivalued functions.

Modeling Evolution of Weighted Clique Networks

We propose a weighted clique network evolution model, which expands continuously by the addition of a new clique （maximal complete sub-graph） at. each time step. And the cliques in the network overlap with each other. The structural expansion of the weighted clique network is combined with the edges＇ weight and vertices＇ strengths dynamical evolution. The model is based on a weight-driven dynamics and a weights＇ enhancement mechanism combining with the network growth. We study the network properties, which include the distribution of vertices＇ strength and the distribution o~ edges＇ weight, and find that both the distributions follow the scale-free distribution. At the same time, we also find that the relationship between strength and degree of a vertex are linear correlation during the growth of the network. On the basis of mean-field theory, we study the weighted network model and prove that both vertices＇ strength and edges＇ weight of this model follow the scale-free distribution. And we exploit an algorithm to forecast the network dynamics, which can be used to reckon the distributions and the corresponding scaling exponents. Furthermore, we observe that mean-field based theoretic results are consistent with the statistical data of the model, which denotes the theoretical result in this paper is effective.

Hybrid reliability model for fatigue reliability analysis of steel bridges

A kind of hybrid reliability model is presented to solve the fatigue reliability problems of steel bridges. The cumulative damage model is one kind of the models used in fatigue reliability analysis. The parameter characteristics of the model can be described as probabilistic and interval. The two-stage hybrid reliability model is given with a theoretical foundation and a solving algorithm to solve the hybrid reliability problems. The theoretical foundation is established by the consistency relationships of interval reliability model and probability reliability model with normally distributed variables in theory. The solving process is combined with the definition of interval reliability index and the probabilistic algorithm. With the consideration of the parameter characteristics of the S-N curve, the cumulative damage model with hybrid variables is given based on the standards from different countries. Lastly, a case of steel structure in the Neville Island Bridge is analyzed to verify the applicability of the hybrid reliability model in fatigue reliability analysis based on the AASHTO.

Detailed Analysis of Quantum Phase Transitions Within the u(2) Algebra

We analyze in detail the quantum phase transitions that arise in models based on the u（2） algebraic description for bosonic systems with two types of scalar bosons. First we discuss the quantum phase transition that occurs in hamiltonians that admix the two dynamical symmetry chains u（2） u（1） and u（2） so（2） by diagonalizing the problem exactly in the u（1） basis. Then we apply the coherent state formalism to determine the energy functioned. Finally we show that a quantum phase transition of a different nature, but displaying similar characteristics, may arise also within a single chain just by including higher order terms in the hamiltonian.

IT Influence on Organizational Structure： Empirical Studies Among Polish Organizations

The paper describes information technologies （IT） role in organization---especially its influence on organizational structure. Article concerns the importance of analyzing IT acceptance, while describing IT in organization and points out that inadequate variable choice may influence validity of IT analysis. First part of the article describes both variables analyzed in presented research--IT dissemination and IT acceptance. It also presents how in theory IT can influence organizational structure. The main part of the article describes empirical studies conducted in order to verify if the influence of IT on the organizational structure exists. First, the main goal and methodology of the empirical studies are presented. Variables used to assess IT and organizational structure in organizations are discussed. Then, there is a description of research results--statistical correlation between analyzed variables and regression models is shown. Conclusion of the article is that IT can influence organizational structure, but the most important factor ensuring this influence is the actual use of IT by employees of the organization--their access to IT is not enough.

Finite Difference Scheme for Solving Parabolic Partial Differential Equations with Random Variable Input under Mean Square Sense

This study deal with seven points finite difference method to find the approximation solutions in the area of mean square calculus solutions for linear random parabolic partial differential equations. Several numerical examples are presented to show the ability and efficiency of this method.

Benford＇s Law and Invariances

Benford＇s law is logarithmic law for distribution of leading digits formulated by P[D=d]= log（1＋1/d） where d is leading digit or group of digits. It＇s named by Frank Albert Benford （1938） who formulated mathematical model of this probability. Befbre him, the same observation was made by Simon Newcomb. This law has changed usual preasumption of equal probability of each digit on each position in number.The main characteristic properties of this law are base, scale, sum, inverse and product invariance. Base invariance means that logarithmic law is valid for any base. Inverse invariance means that logarithmic law for leading digits holds for inverse values in sample. Multiplication invariance means that if random variable X follows Benford＇s law and Y is arbitrary random variable with continuous density then XY follows Benford＇s law too. Sum invariance means that sums of significand are the same for any leading digit or group of digits. In this text method of testing sum invariance property is proposed.

Calculation on model of weakening hydrogen bond of thick oil of petroleum

The process of use catalyst or functional material that contains iron ion to weaken -O-H-O- hydrogen bond of the thick oil to reduce viscidity or crack, in aspects of the ion charge. covalent bond order, total energy and the average distance of Fe-O. is studied with density function theory and discrete variational method （DFT-DVM）, one of the first principle methods. With the decrease of the distance of Fe-O. the charge of Fe ion increases, the charge of hydrogen ion decreases, and hydrogen bond is weakened. There are obvious and more stable effects to use the catalyst that contains multiple metal ions or increase the catalyst amount in weakening hydrogen bond of the thick oil. This theoretic work is helpful to exploit and process the thick oil of petroleum and maybe overcome the crisis of petroleum energy is approaching to us.

Urban landscape classification using Chinese advanced high-resolution satellite imagery and an object-oriented multi-variable model

The Chinese ZY-1 02C satellite is one of the most advanced high-resolution earth observation systems designed for terrestrial resource monitoring. Its capability for comprehensive landscape classification, especially in urban areas, has been under constant study. In view of the limited spectral resolution of the ZY-1 02C satellite （three bands）, and the complexity and hetero- geneity across urban environments, we attempt to test its performance of urban landscape classification by combining a multi- variable model with an object-oriented approach. The multiple variables including spectral reflection, texture, spatial autocorre- lation, impervious surface fraction, vegetation, and geometry indexes were first calculated and selected using forward stepwise linear discriminant analysis and applied in the following object-oriented classification process. Comprehensive accuracy as- sessment which adopts traditional error matrices with stratified random samples and polygon area consistency （PAC） indexes was then conducted to examine the real area agreement between a classified polygon and its references. Results indicated an overall classification accuracy of 92.63% and a kappa statistic of 0.9124. Furthermore, the proposed PAC index showed that more than 82% of all polygons were correctly classified. Misclassification occurred mostly between residential area and barren/farmland. The presented method and the Chinese ZY-1 02C satellite imagery are robust and effective for urban landscape classification.

New Similarity Reduction Solutions for the(2+1)-Dimensional Nizhnik-Novikov-Veselov Equation

In this paper,some new formal similarity reduction solutions for the(2+1)-dimensional Nizhnik-Novikov-Veselov equation are derived.Firstly,we derive the similarity reduction of the NNV equation with the optimal system of the admitted one-dimensional subalgebras.Secondly,by analyzing the reduced equation,three types of similarity solutions are derived,such as multi-soliton like solutions,variable separations solutions,and KdV type solutions.

Variable Selection of Varying Dispersion Student-t Regression Models

The Student-t regression model is a useful extension of the normal model,which can be used for statistical modeling of data sets involving errors with heavy tails and/or outliers and provides robust estimation of means and regression coefficients.In this paper,the varying dispersion Student-t regression model is discussed,in which both the mean and the dispersion depend upon explanatory variables.The problem of interest is simultaneously select significant variables both in mean and dispersion model.A unified procedure which can simultaneously select significant variable is given.With appropriate selection of the tuning parameters,the consistency and the oracle property of the regularized estimators are established.Both the simulation study and two real data examples are used to illustrate the proposed methodologies.

An Improved Dynamic Hysteretic Model for Soils

A dynamic hysteretic constitutive model for soil dynamics, Ramberg-Osgood model, is introduced and improved in the paper. Since the model is inherently 1D and is assumed to apply to shear components only, other components of the deviatorie stress and strain and their relations in 3D case could not be fully described. Two parameters, the equivalent shear stress and the equivalent shear strain, are defined to reasonably establish relations between each of stress and strain components respectively. The constitutive equations of the initial Ramberg-Osgood model are extended to generalize the theory into multidimensional cases. Difficulties of the definition of load reversal in 3D are also addressed and solved. The improved constitutive model for soil dynamics is verified by comparisons with different soil dynamic testing data covering both sands and clays. Results show that the dynamic nonlinear hysteretie behaviors of soils can be well predicted with the improved constitutive model.

Symmetries and solutions to geometrical flows

In this paper,we investigate group-invariant solutions to the hyperbolic geometric flow on Riemann surfaces,which include solutions of separation variables,traveling wave solutions,self-similar solutions and radial solutions.In the proceeding of reduction,there are elliptic,hyperbolic and mixed types of equations.For the first kind of equation,some exact solutions are found;while for the last two kinds,with implicit solutions found,we furthermore investigate whether there will be a global solution or blowing up.Referring to the work of Kong et al.(2009),the results come out perfectly.

Effects of binary interactions on the color evolution of M33

In this work, predictions of the spectral energy distribution from populations of single and binary stars are incorporated into a galactic chemical and color evolution model to explore the significance of the effects of the binary interactions on the color evolution of M33. We first constructed a model without binary interactions, and the model is able to reproduce most of the available observational constraints on the distribution of stellar parameters. We then run simulations with the same set of model parameters but with binary interactions considered. By comparing the results for the populations with and without binary interactions, we find that the inclusion of binary interactions makes the surface brightness greater (~0.1 mag arcsec 2) in FUV-band but smaller (~0.7 mag arcsec 2) in K-band, while it results in the FUV K color bluer (~0.8 mag). To reproduce the observations, a model that considers the binary interactions should make more gas fall onto the disk in the early time of the galaxy evolution, or increase the total stellar mass, or both.

LIFE DISTRIBUTION OF SERIES UNDER THE SUCCESSIVE DAMAGE MODEL

We analyse further the reliability behaviour of series and parallel systems in the successive damage model initiated by Downton. The results are compared with those obtained for other models with different bivariate distributions.

BAYESIAN IMAGE SUPERRESOLUTION AND HIDDEN VARIABLE MODELING

Superresolution is an image processing technique that estimates an original high-resolutionimage from its low-resolution and degraded observations.In superresolution tasks,there have beenproblems regarding the computational cost for the estimation of high-dimensional variables.Theseproblems are now being overcome by the recent development of fast computers and the developmentof powerful computational techniques such as variational Bayesian approximation.This paper reviewsa Bayesian treatment of the superresolution problem and presents its extensions based on hierarchicalmodeling by employing hidden variables.