Some Properties of Fractals Generated by Linear Cellular Automata

Fractals and cellular automata are both significant areas of research in nonlinear analysis. This paper studies a class of fractals generated by cellular automata. The patterns produced by cellular automata give a special sequence of sets in Euclidean space. The corresponding limit set is shown to be a fractal and the dimension is independent of the choice of the finite initial seed. As opposed to previous works, the fractals here do not depend on the time parameter.     

Visualization of Gene Mutation Complicated Pattern of Hepatitis B Virus Based on Cellular Automata

Hepatitis B virus shows instantaneous and high rate mutations in biological experiments, some sorts of which affect the efficiency of virus replication greatly through enhancing or depressing the viral replication, while others have no influence at all. Taking advantage of prominent features of cellular automata, we simulate the effect of hepatitis B virus gene mutation on its replication efficiency. The computer simulation results demonstrate the feasibility of our novel model by comparing with the results of biological experiments.     

Protein Subcellular Localization Prediction and Genomic Polymorphism Analysis of the SARS Coronaviru

The cause of severe acute respiratory syndrome (SARS) has been identified as a new coronavirus (CoV). Several sequences of the complete genome of SARS-CoV have been determined. The subcellular localization (SubLocation) of annotated open-reading frames of the SARS-CoV genome was predicted using a support vector machine. Several gene products were predicted to locate in the Golgi body and cell nucleus. The SubLocation information was combined with predicted transmembrane information to develop a model of the viral life cycle. The results show that this information can be used to predict the functions of genes and even the virus pathogenesis. In addition, the entire SARS viral genome sequences currentlyav ailable in GenBank were compared to identify the sequence variations among different isolates. Some variations in the Hong Kong strains may be related to the special clinical manifestations and provide clues for understanding the relationship between gene functions and evolution. These variations reflect the evolution of the SARS virus in human populations and may help development of a vaccine.     

一种一维细胞自动机的分类

讨论了基本细胞自动机的一种扩展模型,通过二元矩阵展现了局部规则与全局演化的关系.利用矩阵的秩对这类细胞自动机进行了计算机实验分类.     

基于策略演化的股票市场元胞自动机模型

将基于Moore型邻居的投资者策略演化行为及策略演化效用函数引入投资者预期驱动的股票市场元胞自动机模型,对演化动力系统做了动力学分析,并研究了策略模仿行为对价格的影响.理论分析及计算机仿真结果表明:股票价格呈现基本面价值收敛,验证了动力系统的稳定性;引入基于Moore型邻居的策略演化规则后,投资者的策略选择出现明显的羊群行为,并且导致价格波动异常,价格发现效率降低;收益率峰度与投资者策略模仿强度有关.     

基于CA的城市土地利用演变人工神经网络模拟

基于CA原理,利用学习矢量量化神经网络从不同时相遥感数据中挖掘土地利用演变的内在规律,自动找到土地利用元胞的转换规则,并以该规则反演和预测土地利用格局.在上海市区典型边缘带的应用显示,挖掘出的元胞转换规则,与同期上海城市发展状况相吻合,表明该模型可以满足土地利用演变模拟预测的要求,大大缩短了建立CA转换规则所需时间.若能增加社会、经济因素的影响,减少网络训练时的信息浪费,将进一步优化模拟效果.     

Sub-classes and evolution stability of Wolfram's classes in the total-rule cellular automata

Cellularautomata(CA)aremathematicalmodelsforcomplexnaturalsystemscontaininglargernumbersofsimpleidenticalcomponentswithlocalinteractions.TheCAconsistofalatticeofsites,eachwithafinitesetofpossiblevalues.Thesitevaluesevolvesynchronouslyineverytimestep[1～4].Onecellularautomatonconsistsoffourparts:theevolutionrule,theformofgeometricalneighborhood,theinitialconfigurationandtheclock.Whentheclockshowsthattimegoesforwardforonestep,configurationoftheCAupdatesonce.Generally,theevolutionrulesofCAare…     

焊接接头微观组织演化的可视化模拟

本文主要针对焊接接头微观组织演化建模及其可视化方法进行研究。采用元胞自动机模型,通过对焊接接头的微观组织晶粒生长进行模拟来分析和研究微观组织的形态学和动力学特征;通过微观组织图像人工生成技术,直观、精确地展现出微观组织演化的可视化效果。实验结果表明,经过回归分析得到的晶粒生长指数同主流的Burke经验公式相符,使用元胞自动机算法对微观组织演化进行的可视化模拟得到了令人满意的效果。     

禁止字方法在元胞自动机中的一个应用

给出了使用禁止字理论、计算机搜索和符号动力学研究初等元胞自动机演化语言的一种方法，并且使用该方法得到了27号初等元胞自动机演化语言在数学上的精确描述．     

元胞自动机在环境科学中的应用

阐述了元胞自动机(Cellular Automata)的基本原理、方法,总结了CA的甚本特点和适用范围,并又探讨了其在环境科学和生态学中的应用．     

元胞自动机中斑图复制的识别算法

给出了元胞自动机中斑图复制识别的一种算法。通过定义每个元胞的特征数,将两个斑图统一放入一个区域内比较它们的特征数,从而判定是否出现复制现象。数值例子表明,该算法是有效的。     

基于元胞自动机的乘客登机策略

登机策略的研究以最小化乘客登机时间为目标。该文对登机策略选择问题进行仿真建模和评价分析。分析乘客登机过程和特点,利用双曲线模型描述乘客放置行李时的过道冲突,提出了过道冲突模型和座位冲突模型。根据元胞自动机仿真原理,建立了乘客登机模拟流程。对7类21种登机策略的模拟结果表明:有序的策略并不一定比无序策略更能提高登机效率;相反,乘客自由登机可能是一种最简单且行之有效的办法。     

多态概率元胞自动机森林火灾模型及应用

通过对概率元胞自动机模型的改进,分析了在既不能生长又不能燃烧的空地存在或森林砍伐形成的多态邻居条件下,森林中树木占有率的变化。模拟结果表明在适度留一些既不能生长又不能燃烧的空地或适度砍伐的条件下,不但不会减少森林中树木的比率反而有利于树木的生长。模拟结果与真实情况比较吻合。