分布阶扩散-波动方程的有限元解的误差估计

本文考虑了分布阶时间分数阶扩散波动方程,其中时间分数阶导数是在Caputo意义上定义的,其阶次$\alpha,\beta$分别属于（0,1）和（1,2）.文中提出了在计算上行之有效的数值方法来模拟分布阶时间分数阶扩散波动方程.在时间上,通过中点求积公式把分布阶项转换为多项的时间分数阶导数项,并且利用$L1$和$L2$公式来近似Caputo分数阶导数;空间上使用Galerkin有限元方法进行离散.给出了基于$H^1$范数的有限元解的稳定性和误差估计的详细证明,最后的数值算例结果说明了理论分析的正确性以及有效性.     

高维非齐次时间分数阶扩散-波动方程的解析解问题

分数阶微积分是专门研究任意阶积分和微分的数学性质及其应用的领域,是传统的整数阶微积分的推广,分数阶微分方程是含有非整数阶导数的方程.时间分数阶扩散-波动方程可以用来模拟由传统的扩散-波动方程演变而来的反常扩散方程.考虑在有限区间上高维非齐次时间分数阶扩散-波动方程的初边值问题.利用分离变量法,导出了高维非齐次时间分数阶扩散-波动方程初边值问题的基本解.     

分数阶微分方程的理论和数值方法研究

分数阶偏微分方程的研究有很长的历史,并在最近十多年得到快速发展.相比极为有限的理论成果,数值方法的研究成果已经相当丰富,几个国际研究团队对此作出了贡献.本文旨在对分数阶微分方程的理论与数值方法研究成果做个简要的评价,聚焦总结评述与高阶方法发展密切相关的研究.主要内容为讨论最基本的三类方程:时间分数阶扩散方程、空间分数阶扩散方程、以及时空分数阶扩散方程的理论进展和数值方法研究在最近十年取得的结果.我们还有针对性地选择一些算例,用以说明几个重要方法的精度和有效性.     

多项时间分数阶扩散方程类 Wilson非协调元的超收敛分析

基于L1离散格式,针对具有Caputo导数的二维多项时间分数阶扩散方程给出了类Wilson非协调有限元方法.首先证明其逼近格式的无条件稳定性.其次利用该单元的特殊性质和分数阶导数巧妙的处理技巧导出了超逼近结果,进一步地,借助插值后处理技术导出了超收敛估计.     

基于分数阶热传导方程激光加热瞬态温度场研究

基于分数阶Taylor(泰勒)级数展开原理,建立单相延迟一阶分数阶近似方程,获得分数阶热传导方程.针对短脉冲激光加热问题建立分数阶热传导方程组,并运用Laplace(拉普拉斯)变换方法进行求解,给出非Gauss(高斯)时间分布的激光内热源温度场解析解.针对具体算例数值研究温度波传播特性.结果表明热传播速度与分数阶阶次有关,分数阶阶次增加,热传播速度减小,温度变化幅度增加.分数阶方程可以用于描述介于扩散方程和热波方程间的热传输过程,且对热传播机制与分数阶热传导方程中分数阶项的关系做了深入剖析.     

分数阶Burgers方程的有限元计算(英文)

建立了一维和二维分数阶Burgers方程的有限元格式.时间分数阶导数使用L1方法离散,空间方向使用有限元方法离散.通过选择合适的基函数,将离散后的方程转化成一个非线性代数方程组,并应用牛顿迭代方法求解.数值实验显示出了方法的有效性.     

分数阶扩散方程的隐差分格式

根据移位的Grnwald方法,得到求解分数阶扩散方程的三类隐差分格式.利用分数阶von Neumann方法,证明了求解亚扩散方程的两类差分格式是无条件稳定的,而求解超扩散方程的差分格式是条件稳定的,同时也给出了相应差分格式的局部截断误差估计.最后,通过两个数值例子证实了所提出的差分格式的正确性和有效性.     

变系数分数阶对流扩散方程的一种算子矩阵方法

研究带Caputo分数阶导数的变系数对流扩散方程的数值解法.基于Chebyshev cardinal函数,推导Riemann-Liouville分数阶积分的一个有效算子矩阵,以之为基础,提出了变系数分数阶对流扩散方程的一种新的算子矩阵法.该方法将方程的求解转化成矩阵的代数运算,具有计算量小和易于编程等特点.给出数值算例并与一些现有的方法进行比较,结果表明该方法是收敛的且在计算精度上占有优势.     

带漂移的单侧正规化回火分数阶扩散方程的Crank-Nicolson拟紧格式

基于已有的针对单侧正规化回火分数阶扩散方程的三阶拟紧算法,将该算法的思想应用于带漂移的单侧正规化回火分数阶扩散方程的数值模拟,并结合Crank-Nicolson方法导出数值格式.证明了数值格式的稳定性与收敛性,且数值格式的时间收敛阶和空间收敛阶分别是二阶和三阶.通过数值试验验证了数值格式的有效性和理论结果.     

带初始奇异性的多项时间分数阶扩散方程一类全离散数值方法的误差分析

本文研究了带有初始奇异性的多项时间分数阶扩散方程的一种全离散数值方法.首先,基于L1公式在渐变网格下离散多项Caputo时间分数阶导数,构造了多项时间分数阶扩散方程的时间半离散格式,证明了时间格式通过选取合适的网格参数r,时间方向的误差可以达到最优的收敛阶2-α_1,其中α_1(0 α_11)为多项时间分数阶导数阶数的最大值.然后,空间采用谱方法进行离散,得到了全离散格式,证明了全离散格式的无条件稳定性和收敛性.为了降低计算量和储存量,对多项时间分数阶扩散方程又构造了时间方向的快速算法,同时证明了该格式的收敛性.数值算例验证了算法的有效性,显示了快速算法的高效性.     

基于IPAT模型的长株潭城市群经济增长与能源消耗的实证研究

基于IPAT方程,通过设定不同的情景来分析长株潭城市群2008年-2013年期间经济增长与资源消耗之间的定量关系．实证结果表明：GDP年增长率为12％,单位GDP能耗年下降4％这种组合是合理和适当的。符合长株潭城市群能源消耗的实际．以此为目标,能平稳推进长株潭城市群资源节约型社会建设进程,保持国民经济快速、健康、稳定的发展．     

由Smith标准形求解矩阵方程∑AiXBi=C

By using the Smith normal form of polynomial matrix and algebraic methods, this paper discusses the solvability for the linear matrix equation ∑AiXBi = C over a field, and obtains the explicit formulas of general solution or unique solution.     

A knowledge-based multi-role decision support system for ore blending cost optimization of blast furnaces

Literature illustrates the difficulties in obtaining the lowest-cost optimal solution to an ore blending problem for blast furnaces by using the traditional trial-and-error method in iron and steel enterprises. To solve this problem, we developed a cost optimization model which we have implemented in a multi-role-based decision support system (DSS). On the basis of analyzing the business flow and working process of ore blending, we propose an architecture of DSS which is built based on multi-roles. This DSS construction pre-processes the data for materials and elements, builds a general database, abstracts the related optimal operations research models and introduces the reasoning mechanism of an expert system. A non-linear model of ore blending for blast furnaces and its solutions are provided. A database, a model base and a knowledge base are integrated into the expert system-based multi-role DSS to meet the different demands of data, information and decision-making knowledge for the various roles of users. A comparison of the results for the DSS and the trial-and-error method is provided. The system has produced excellent economic benefits since it was implemented at the Xiangtan Iron & Steel Group Co. Ltd., China.     

Splitting up value: A critical review of residual income theories

This paper deals with the notion of residual income, which may be defined as the surplus profit that residues after a capital charge (opportunity cost) has been covered. While the origins of the notion trace back to the 19th century, in-depth theoretical investigations and widespread real-life applications are relatively recent and concern an interdisciplinary field connecting management accounting, corporate finance and financial mathematics (Peasnell, 1981, 1982; Peccati, 1987, 1989, 1991; Stewart, 1991; Ohlson, 1995; Arnold and Davies, 2000; Young and O’Byrne, 2001; Martin, Petty and Rich, 2003). This paper presents both a historical outline of its birth and development and an overview of the main recent contributions regarding capital budgeting decisions, production and sales decisions, implementation of optimal portfolios, forecasts of asset prices and calculation of intrinsic values. A most recent theory, the systemic-value-added approach (also named lost-capital paradigm), provides a different definition of residual income, consistent with arbitrage theory. Enfolded in Keynes’s (1936) notion of user cost and forerun by Pressacco and Stucchi (1997), the theory has been formally introduced in Magni (2000a,b,c; 2001a,b; 2003), where its properties are thoroughly investigated as well as its relations with the standard theory; two different lost-capital metrics have been considered, for value-based management purposes, by Drukarczyk and Schueler (2000) and Young and O’Byrne (2001). This work illustrates the main properties of the two theories and their relations, and provides a minimal guide to construction of performance metrics in the two approaches.     

关于群中乘积元素的阶

讨论群中两个元素a,b的阶不相等时其乘积ab的阶的一类计算问题.设ㄧaㄧ=m,ㄧ bㄧ=n,若(m,n)=1,且存在k∈N使a=bk,则有ㄧabㄧ=mn/d1d2,其中d1=(m,k+1),d2=(n,k+1).若m≠n,ab=ba,且(m,n)ㄧm/(m,n),或(m,n)ㄧn/(m,n),则有ㄧabㄧ=[m,n].     

泛系分析:数学·方法论·相对论·辩证思维

基于对广义系统、泛导、泛对称、泛箱原理、泛系相对性提出了新的分析模式与新的相对数学化的简化强化的定义,发展了泛系方法论与泛系相对论的框架与理法,具体内容涉及:相对普适的数学化的泛系,200类对偶,对偶转化,泛对称转化,泛系辩证,八畴方法,泛系数学方法,广义量化,逼近转化原则,泛等价定理,供求分析,思维实验,广义灰色系统,等等.     

The permanence and extinction of a nonlinear growth rate single-species non-autonomous dispersal models with time delays

In this paper, we consider the effect of diffusion on the permanence and extinction of a non-autonomous nonlinear growth rate single-species dispersal model with time delays. Firstly, the sufficient conditions of the permanence and extinction of the species are established, which shows if the growth rate and dispersal coefficients is suitable, the species is permanent, on the contrary, it is extinction. Secondly, an interesting result is established, that is, if only the species in some patches even in one patch is permanent, then it is also permanent in other patches. Finally, some examples together with their numerical simulations show the feasibility of our main results.     

Generalized symmetry of graphs — A survey

Symmetry of graphs has been extensively studied over the past fifty years by using automorphisms of graphs and group theory which have played and still play an important role for graph theory, and promising and interesting results have been obtained, see for examples, [L.W. Beineke, R.J. Wilson, Topics in Algebraic Graph Theory, Cambridge University Press, London, 2004; N. Biggs, Algebraic Graph Theory, Cambridge University Press, London, 1993; C. Godsil, C. Royle, Algebraic graph theory, Springer-Verlag, London, 2001; G. Hahn, G. Sabidussi, Graph Symmetry: Algebraic Methods and Application, in: NATO ASI Series C, vol. 497, Kluwer Academic Publishers, Dordrecht, 1997]. We introduced generalized symmetry of graphs and investigated it by using endomorphisms of graphs and semigroup theory. In this paper, we will survey some results we have achieved in recent years. The paper consists of the following sections.

1. Introduction

2. End-regular graphs

3. End-transitive graphs

4. Unretractive graphs

5. Graphs and their endomorphism monoids.