我国能源-环境-经济系统的区域间建模

本研究试图探讨用“自上而下”和“自下而上”相结合的建模方法,对我国二氧化硫和二氧化碳减排进行区域间建模.本研究以一个用于分析硫税的区域间波及影响、以及评价电力部门的脱硫技术的区域间可计算一般均衡模型为基础平台;从理论上探讨了如何对该模型进行扩展和改进.首先本研究对该模型的要点进行了阐述,强调了把能源技术模型融合在区域间可计算一般均衡模型中的建模方法.然后本研究对于实行环境税和排放权交易的环境经济效果进行了比较,用成本收益法分析了工业部门对减排技术的选择,并以前述模型为基础建立了区域间模型.本研究所提出的模型可以被用来从区域统筹的观点,对我国能源-环境-经济系统的可持续发展进行预测和分析.最后本研究提出应该从更新数据、研究环境政策和技术进步之间的关系等方面,进一步发展该模型.     

一种基于复杂性理论的装备作战仿真方法

采用什么样的计算机仿真方法来隐喻真实的作战系统是装备作战仿真研究的关键问题.从复杂性科学的研究角度,引入了复杂适应系统(CAS)理论及其技术体系,提炼了基于Agents/space的建模与仿真框架,说明了框架实现的关键技术——可计算模型、复杂性解决方案和仿真实现平台.进而利用该方法进行了典型装备作战仿真问题研究,包括:利用神经网络、三维连续空间可计算模型,并选用Mason平台实现了装甲装备战损规律仿真;利用三层元胞自动机、产生式系统可计算模型,并选用Repast平台实现了装备群对抗仿真.为基于复杂性理论开展装备作战仿真或具有类似特征问题的研究提供了一种新的试验途径.     

基于因子分析的学生数学建模能力结构研究

通过编制测试卷对深圳市龙岗区17327名初中生进行数学建模能力测试.根据测试数据,利用探索性因素分析,探索初中生数学建模能力的整体结构和组成因素,对于探索出的数学建模能力结构模型分别利用一阶和二阶验证性因素分析验证其合理存在性.研究表明,我国初中生数学建模能力包含模型建立能力,模型阐释能力,模型计算能力和模型评估能力四个组成因素,每个组成要素包含若干数学基本技能,该能力模型可以与现有数学建模能力理论形成三角互证.最后,根据初中生的数学建模能力结构模型,提出三个相应的教学策略,以期帮助学生提高数学建模能力.     

基于等效原理的波形钢腹板组合箱梁桥动力特性分析

为了简化波形钢腹板组合箱梁的三维有限元建模过程,通过刚度和位移等效的方法将三维波形钢腹板有限元模型简化为二维正交各向异性板的有限元模型.简化之后的有限元模型几何外形简单,单元和自由度的数量大幅降低,减少了计算时间,提高了计算效率.通过对比三维波形钢腹板组合箱梁有限元模型的自振频率计算结果、二维等效正交各向异性板组合箱梁有限元模型的自振频率计算结果以及实测频率值,发现三者吻合良好,验证了等效模型有限元建模方法的正确性和可靠性.研究结论可为波形钢腹板组合箱梁桥提供一种简单的有限元建模方法.     

论灰色理论的三数据建模

本文通过论证序列的GM(1,1)模型计算值与原始序列的初值无关,讨论仅有三个数据构成的序列的灰色建模问题,得出可进行灰色建模的最少数据数为3的结论.     

累加生成的改进和GM(1,1,t)灰色模型

根据卷积变换可提高变换序列光滑度的特性和累加生成的机理,对灰色建模中的序列生成方式和GM(1,1)模型加以改进,用线性序列对建模序列作卷积变换,建立带线性时间项的灰色模型GM(1,1,t),实例计算结果表明GM(1,1,t)模型的模拟精度较GM(1,1)模型有较大提高且适用范围更广.     

非参数局部多项式回归估计的最优子抽样算法

随着科学技术的发展,虽然人们提高了收集和处理数据的能力,但仍存在一些大数据集超出了现有计算机的计算能力.目前,抽取一部分样本来替代全样本进行建模计算是减轻计算负担的一种方法.大数据背景下线性模型的子抽样方法已经得到了相对成熟的研究,在减轻计算量方面获得了很大的优势.文章将线性模型下的子抽样方法推广到非参数回归模型,并推...     

基于模态综合法的大型结构模型修正技术

基于有限元模型修正理论和模态综合法,改进了子结构试验建模综合方法.在子结构模型修正阶段,应用固定交界面模态综合法,对子结构模型实施减缩,提高子结构模型的计算和修正效率,从而提高了子结构试验建模综合技术整体计算效率.在减缩模型的基础上,推导模型修正设计参数法的计算公式,详细列出此方法的执行步骤,并以某运载火箭模型为算例,结合商业软件ANSYS二次开发,验证了该文方法的有效性.     

数学建模视角下的教育量化研究

以对人教A版高中必修教材使用顺序研究的技术路线为例,即从概念到变量—从变量到数据—从数据到模型—从模型到计算结果—从计算结果到揭示、解释教育规律的模式,说明数学建模是量化研究的核心,量化研究的本质就是通过建立数学模型以达到解决相关教育问题的目的.     

拟稳态Maxwell方程的可计算建模和应用

介绍拟稳态Maxwell方程在电气工程领域的可计算建模及应用。对于含导电材料的电磁设备,拟稳态Maxwell方程是描述电流密度分布和欧姆损耗的常用模型,在电机、大型变压器等电气工程设备和集成电路等微电子技术领域有广泛应用。本文以国际计算电磁学会公布的TEAM Workshop Problem 7和21基准族问题为例,阐述拟稳态Maxwell方程的可计算建模和自适应有限元计算。     

Initial segments of computable linear orders with additional computable predicates

We study computable linear orders with computable neighborhood and block predicates. In particular, it is proved that there exists a computable linear order with a computable neighborhood predicate, having a Π₁⁰ -initial segment which is isomorphic to no computable order with a computable neighborhood predicate. On the other hand, every Σ₁⁰ -initial segment of such an order has a computable copy enjoying a computable neighborhood predicate. Similar results are stated for computable linear orders with a computable block predicate replacing a neighborhood relation. Moreover, using the results obtained, we give a simpler proof for the Coles–Downey–Khoussainov theorem on the existence of a computable linear order with Π₂⁰ -initial segment, not having a computable copy.     

A posteriori estimation of the linearization error for strongly monotone nonlinear operators

We investigate the a posteriori estimation of the modeling (or linearization) error which arises when a nonlinear problem is replaced by a linear model. Using the context of strongly monotone operators, we construct a computable upper estimator for this error, and also provide an estimator that gives a lower bound. Several numerical results illustrating our theory are provided.     

图灵机计算实数函数的稳定性

定义在全体实数上的可计算函数是一个很重要的概念．在这以前定义可计算的实数函数有两个途径．第一个途径是首先要定义可计算实数的指标．想要确定实数函数y=f（x）是不是可以计算就要看是否存在一个自然数的（部分）递归函数将可计算实数x的指标对应到可计算实数y的指标．这样一来对实数函数的研究依赖于对自然数函数的研究．第二个定义可计算的实数函数的途径是以逼近为基础的．一个实数函数是可以计算的如果它既是序列可计算的同时也是一致连续的．用这个途径来定义可计算实数函数使用的条件过强以至于很多有用的实数函数成为不可计算的实数函数．例如“〈”和“=”的命题函数就是不可以计算的因为它们是不连续的命题函数．本文讨论了图灵机的稳定性并且给出了一个基于稳定图灵机的可计算实数函数的定义．我们的定义不需要用到自然数的（部分）递归函数．根据我们的定义很多常用实数函数特别是一些不连续的常用实数函数都是可以计算的．用我们的定义来讨论可计算实数函数的性质比原来的定义要方便得多．     

Computable dimension for ordered fields

The computable dimension of a structure counts the number of computable copies up to computable isomorphism. In this paper, we consider the possible computable dimensions for various classes of computable ordered fields. We show that computable ordered fields with finite transcendence degree are computably stable, and thus have computable dimension 1. We then build computable ordered fields of infinite transcendence degree which have infinite computable dimension, but also such fields which are computably categorical. Finally, we show that 1 is the only possible finite computable dimension for any computable archimedean field.     

Bounded iteration and unary functions

The set of unary functions of complexity classes defined by using bounded primitive recursion is inductively characterized by means of bounded iteration. Elementary unary functions, linear space computable unary functions and polynomial space computable unary functions are then inductively characterized using only composition and bounded iteration. (© 2004 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Boolean Algebras with Finite Families of Computable Automorphisms

We prove that each computable Boolean algebra has a computable presentation in which for every computable family of automorphisms the set of atoms moved by at least one of its members is finite. This implies that each computable atomic Boolean algebra has a computable presentation in which its every computable family of automorphisms is finite. The priority argument is not used in the proof.     

Some constructions on the class of groups of computable automorphisms

We show that the computable inverse limit of a computable family of computable groups, the computable wreath product of a group of computable automorphisms and a computable group, as well as the commutant and center of every computable group can be realized as groups of computable automorphisms of suitable computable models.     

When series of computable functions with varying domains are computable

In this paper we study the behavior of computable series of computable partial functions with varying domains (but each domain containing all computable points), and prove that the sum of the series exists and is computable exactly on the intersection of the domains when a certain computable Cauchyness criterion is met. In our point‐free approach, we name points via nested sequences of basic open sets, and thus our functions from a topological space into the reals are generated by functions from basic open sets to basic open sets. The construction of a function that produces the sum of a series requires working with an infinite array of pairs of basic open sets, and reconciling the varying domains. We introduce a general technique for using such an array to produce a set function that generates a well‐defined point function and apply the technique to a series to establish our main result. Finally, we use the main finding to construct a computable, and thus continuous, function whose domain is of Lebesgue measure zero and which is nonextendible to a continuous function whose domain properly includes the original domain. (We had established existence of such functions with domains of measure less than ε for any , in an earlier paper.)     

Universal computable enumerations of finite classes of families of total functions

In the paper we introduce the notion of a computable enumeration of a class of families. We prove a criteria for the existence of universal computable enumerations of finite classes of computable families of total functions. In particular, we show that there is a finite computable class of families of total functions without universal computable enumerations.