关于非齐次马氏链的Cesaro平均收敛性

首先证明非齐次马氏链一个Ｃｅｓａｒｏ平均收敛定理，它是Ｂｏｗｅｒｍａｎ等人一个结果的推广，本文利用这个收敛定理给出非齐次马氏链一元泛函的一个极限定理，并讨论这个极限定理在马氏决策过程和信息论中的应用     

关于非齐次马多链的Cesaro平均收敛性

首先证明了齐次马氏链一个Ｃｅｓａｒｏ平均收敛定理，它是Ｂｏｗｅｒｍａｎ等人的一个结果的推广，本文利用这个收敛定理给出非齐次马氏链一元泛函的一个极限定理，并讨论了这个极限定理在马氏决策过程和信息论中的应用。     

可列m重非齐次马氏链的遍历性与熵率

本文在研究可列m重非齐次马氏链各种遍历性定义的基础上,以转移概率引入可列m重非齐次马氏链绝对平均强遍历性的概念,通过设定可列m重非齐次马氏链满足这种强遍历的充分条件,得出可列m重非齐次马氏链泛函的一个极限定理,并应用此极限定理得到了可列m重非齐次马氏链熵率存在的一个定理。     

关于齐次树指标马氏链的广义熵遍历定理

本文主要研究有限状态齐次树指标Markov链的强大数定律和广义熵遍历定理．熵遍历定理研究的是信息论中信源的渐近均分割性,树指标Markov链是近年来概率论的研究方向之一．首先,参照非齐次Markov链广义熵密度概念,本文给出了树指标Markov链的广义熵密度的定义．然后,通过构造一组期望值为1的随机变量,利用Markov不等式和Borel-Cantelli引理,证明得到了定义在树指标Markov链上一类随机变量的延迟平均的强极限定理．最后,利用上述定理的推论,我们证明得到了Cayley树上有限状态Markov链状态出现次数的延迟平均的强大数定律和广义熵遍历定理．本文的结果是对一些已有结果的推广．     

二叉树上非齐次分支马氏链一类强极限定理

近年来树图或者树形网络等诸多复杂系统的结构性质与极限性质逐渐成为研究的热点问题,特别是在树指标马尔可夫链领域的研究中,国内外学者们取得了丰富的研究成果.二叉树上非齐次分支马尔可夫链作为一类特殊的树指标马尔可夫链,该模型的极限性质被国内外学者的广泛研讨并应用于生物动力学、信息论等诸多领域.本文致力于研究在有限状态空间空间取值的二叉树上非齐次分支马尔可夫链转移概率调和平均的极限性质以及该性质与树指标马尔可夫链模型之间的联系.首先在新的条件下,本文给出了在有限状态空间中取值的二叉树上非齐次分支马氏链的强极限定理,并进一步得到了其随机转移概率调和平均的强极限定理,最后借助于两类模型之间的等价关系以及平均值不等式,推广了树指标非齐次马氏链随机转移概率的极限定理。     

关于非齐次马氏链信源的一个编码定理

本文旨在将无记忆离散信源的编码定理推广至非齐次马尔科夫链情形,以扩展无记忆离散信源编码定理的适用范围．利用经典的波莱尔-坎特利引理,建立关于非齐次马尔科夫链延迟平均的强大数定理,应用独立随机信源逼近非齐次马氏信源,从而获得非齐次马氏信源的广义编码定理．最后运用得到的广义编码定理,给出分批数据假设检验问题中可容忍错误概率的最小值的计算方法．     

齐次树指标随机场的一类Shannon-McMillan偏差定理

本文通过齐次树指标任意随机场与马氏链场相比较,研究齐次树上任意随机场用不等式给出的一类Shannon-McMillan极限定理,即随机偏差定理,为进一步研究树图随机场的信息编码提供了理论基础.     

多重马氏链及其在气象上的应用例

本文给出运用多重马氏链的转移概率计算它的绝对概率的公式,并举例说明如何运用此公式于天气状态的子报.     

基于一般包含度的广义变精度粗糙集

本文将变精度粗糙集模型推广到基于一般包含度的广义变精度粗糙集,利用包含度的性质讨论广义变精度粗糙集的性质。给出广义齐次包含度的定义,验证了常用的概率型包含度等均为广义齐次包含度;给出基于该广义变精度粗糙集的知识约简的方法。不同的包含度代表不同的决策语义,因此借助一般包含度研究变精度粗糙集,能够使尽可能多的有用信息被提取、挖掘,克服了基本粗糙集模型中由于要求绝对精确的包含关系而使大量有用信息丢失,进一步推广了粗糙集模型,拓宽了粗糙集在数据挖掘、知识发现、模式识别及决策分析等领域中的应用。     

一种基于改进MCMC算法的模型修正方法

标准马尔可夫链蒙特卡罗(MCMC)算法不易收敛、拒绝率高,使其应用受到限制。在贝叶斯方法中引入最大熵值法来估计参数的后验概率密度函数最大值,进而将布谷鸟算法中新鸟巢更新的思想融入Metropolis-Hasting(MH)抽样算法得到改进的MH抽样算法,同时使用支持向量机(SVM)建立待修正参数与有限元模型输出之间的代理模型,以提高模型修正的计算效率。分别使用三自由度线性系统和平面桁架模型来验证本文方法的有效性,结果表明:修正后样本的马尔可夫链混合性能好,停滞概率低,修正后参数相对误差均小于2%。     

最大熵法可靠度理论在工程中的应用

基于信息论中最大熵的概念，探讨了最大熵法的可靠度理论在工程中的应用。讨论了样本的均值、标准差对收敛性的影响，以及样本数量与最大熵法阶数对概率密度函数、超越概率、可靠指标精度的影响。提出了结构分析中应采用四阶矩法，并给出应用该方法时精度控制的若干建议。     

某类系数与Fejér缺项级数有关的齐次和非齐次高阶线性微分方程亚纯解的增长性

Nevanlinna理论在复微分方程领域中具有广泛的应用,其中运用该理论研究复线性微分方程亚纯解的增长性和值分布与系数的增长性之间的关系是复微分方程领域中的重要论题．由于缺项级数具有一些特殊性质,当缺项级数作为方程系数时,这些性质即可发挥作用．因此,我们可结合缺项级数的定义和性质研究复线性微分方程亚纯解的性质．在本文中,我们运用Nevanlinna理论并结合Fejér缺项级数的定义和性质对一类齐次和非齐次高阶复线性微分方程进行了研究．当方程的某个系数与Fejér缺项级数有关而其余系数为整函数或亚纯函数时,得到了方程亚纯解的增长级的估计,推广并改进了前人已有结果．     

Wave Propagation in a Green–Naghdi Thermoelastic Solid with Diffusion

The Green and Naghdi theory of thermoelasticity is applied to study plane-wave propagation in an elastic solid with thermo-diffusion. The governing equations of an elastic solid with generalized thermo-diffusion are solved to show the existence of three coupled longitudinal waves and a shear vertical (SV) wave in a two-dimensional model of the solid with thermo-diffusion. The reflection of plane waves from a thermally insulated stress-free surface of an elastic solid with thermo-diffusion is also studied. A non-homogeneous system of four equations in reflection coefficients is obtained. The speeds of the plane waves are computed numerically and plotted against frequency for a particular range. The complex absolute values of the reflection coefficients of all reflected waves are computed numerically and plotted against the angle of incidence of the striking wave at the free surface. The effects of diffusion parameters are shown graphically for speeds and reflection coefficients of plane waves.     

基于LMD-MFE和DHMM的滚动轴承故障诊断算法

针对滚动轴承在恶劣环境影响下,其特征信息难以被有效提取出来的问题,提出一种基于局部均值分解(Local Mean Decomposition,LMD)和多尺度模糊熵(Multiscale Fuzzy Entropy,MFE)的滚动轴承故障诊断算法。首先,利用LMD对轴承振动信号进行分解,得到一系列乘积函数(Product function,PF)分量,并与经验模态分解(Empirical Mode Decomposition,EMD)进行对比,分析其优越性;然后对每一个分量分别提取MFE特征,同时与多尺度样本熵(Multiscale Sample Entropy,MSE)进行对比,分析MFE的优越性;最后结合各个轴承状态的类间平均距离对多个尺度因子下的熵值进行优选,筛选出可分性良好的敏感特征集,并输入到离散隐马尔科夫模型(Discrete Hidden Markov Models,DHMM)模式分类器中对轴承故障类型进行诊断识别。实验结果表明,所提出的基于LMD和MFE的轴承故障诊断算法能较好识别出多种轴承故障类型。     

Asymptotic stability,Onsager fluxes and reaction kinetics

This paper constitutes a first step in the derivation of thermodynamics directly from the dynamics of physical systems. The existence of an asymptotically stable equilibrium point is used to construct a family of admissible entropy functions. These functions have nonnegative entropy production and assume an absolute maximum at the equilibrium point. A nonlinear generalization of the Onsager theory is then used to obtain a one-to-one correspondence between entropy production functions and the governing system of autonomous rate equations. The theory is applied to well stirred chemically reacting systems with constant temperature and pressure. This allows the derivation of chemical potentials, Gibbs' potentials and enthalpy for such systems. The rate equations for reaction kinetics and for classical thermodynamic reaction theory are obtained. Classic thermodynamic reaction theory is shown to give maximum entropy for constant enthalpy at the equilibrium point, while reaction kinetics gives this result only to within quadratic terms in the departure from equilibrium.     

Dependability analysis of systems modeled by non-homogeneous Markov chains

The case of time non-homogeneous Markov systems in discrete time is studied in this article. In order to have measures adapted to this kind of systems, some reliability and performability measures are formulated, such as reliability, availability, maintainability and different time variables including new indicators more dedicated to electrical systems like instantaneous expected load curtailed and the expected energy not supplied on a time interval. The previous indicators are also formulated in the case of cyclic chains where asymptotic results can be obtained. The interest of taking into account hazard rate time variation, is to get more accurate and more instructive indicators but also be able to access new performability indicators that cannot be obtained by classical methods. To illustrate this, an example from an Electricité De France electrical substation is solved.     

Storage Conditions for Stability Testing of Pharmaceuticals in Hot and Humid Regions

ABSTRACT

A review of the methodology for determination of the storage conditions for stability testing according to Schumacher/Grimm is presented in this paper. The purpose is to provide scientific information useful for the definition of storage conditions for stability testing of pharmaceuticals suitable to the region where the product will be dispensed. Special attention is given to stability testing in the new markets located in developing countries with very hot and humid climates. Finally, storage conditions for stability testing in the Brazilian regions were derived and examined comparatively with the guidelines of the world health organization (WHO) and regulatory bodies. The storage conditions were derived from the calculated values of the mean kinetic temperature and the relative humidity (RH). These parameters were estimated from daily values of dry and dew point temperatures of all Brazilian capitals from 1998 to 2002; collected in the morning (9 a.m.), in the afternoon (3 p.m.), and at night (9 p.m.). The Brazilian Center of Weather Forecast and Climatic Studies of the National Institute of Spatial Research (CPTEC/INPE) kindly furnished these data. Significant differences of the mean kinetic temperature (MKT) and relative humidity (RH) for Brazilian regions were observed. These results indicate the existence of a high climatic diversity between the Brazilian regions, making challenging the definition of a single storage condition for the stability testing. Some regions present RH values higher than 80%, giving support to the concerns of the WHO, indicating the necessity of revision of existing guidelines for stability testing mainly for very hot and humid regions.     

Storage conditions for stability testing of pharmaceuticals in hot and humid regions

A review of the methodology for determination of the storage conditions for stability testing according to Schumacher/Grimm is presented in this paper. The purpose is to provide scientific information useful for the definition of storage conditions for stability testing of pharmaceuticals suitable to the region where the product will be dispensed. Special attention is given to stability testing in the new markets located in developing countries with very hot and humid climates. Finally, storage conditions for stability testing in the Brazilian regions were derived and examined comparatively with the guidelines of the world health organization (WHO) and regulatory bodies. The storage conditions were derived from the calculated values of the mean kinetic temperature and the relative humidity (RH). These parameters were estimated from daily values of dry and dew point temperatures of all Brazilian capitals from 1998 to 2002; collected in the morning (9 a.m.), in the afternoon (3 p.m.), and at night (9 p.m.). The Brazilian Center of Weather Forecast and Climatic Studies of the National Institute of Spatial Research (CPTEC/INPE) kindly furnished these data. Significant differences of the mean kinetic temperature (MKT) and relative humidity (RH) for Brazilian regions were observed. These results indicate the existence of a high climatic diversity between the Brazilian regions, making challenging the definition of a single storage condition for the stability testing. Some regions present RH values higher than 80%, giving support to the concerns of the WHO, indicating the necessity of revision of existing guidelines for stability testing mainly for very hot and humid regions.