细胞型膜系统设计方法研究
发布时间：2018-08-13 20:26
【摘要】：膜计算(Membrane Computing, MC)作为计算机科学的一个重要研究领域,旨在研究从生命细胞的结构与功能中以及从组织和器官等细胞群的协作中抽象出的计算模型,其计算模型被称为膜系统或P系统。膜系统是一类具有分布式、极大并行性的计算模型,大量研究表明,许多简单膜系统在理论上具有与图灵机同等的计算能力。 膜计算是当前非常活跃的一个研究领域,研究者们用数学、形式语言等工具来进行膜计算的理论研究,提出了各种P系统模型并取得很多研究成果。目前已提出的P系统模型都是由研究者们精心设计而成,其设计过程都较为复杂,既需要扎实的专业知识,又需要仔细反复推导论证,费时费力。对于膜系统设计方法研究,目前相关工作还非常少,非常需要进一步研究。本文以细胞型膜系统作为研究对象,探讨采用量子进化算法和P-Lingua软件实现膜系统自动优化设计的方法。 论文主要工作及研究成果如下： 1、概述膜计算基本理论,简要介绍细胞型膜系统仿真软件,对膜系统仿真软件P-Lingua进行详细介绍,并通过实例说明细胞型膜系统在P-Lingua软件中仿真实现方法。 2、详细描述和分析细胞型膜系统设计问题,给出简单细胞型膜系统优化设计方法,具体介绍了其设计思路、采用的量子进化算法、膜系统的编码和评价方式。将此方法应用于计算四平方膜系统的设计,获得比遗传算法更好的实验结果。在此基础上,对此方法进行拓展,用于计算任意自然数平方膜系统设计,并通过实验验证该方法的有效性。 3、通过改进评价方法和在量子进化算法中加入局部扰动策略,实现膜系统优化设计方法的改进。运用改进后的设计方法,在同样的给定条件下,只通过改变评价函数,分别实现任意自然数平方P系统、代数式P系统、语言产生P系统的设计。此外,与采用传统的量子进化算法、随机搜索算法的设计方法分别进行对比实验,结果表明改进后的设计方法能获得更好的效果。 4、给出细胞型膜系统的踪迹设计法,具体介绍了膜系统踪迹的定义、设计思路、膜系统的编码和评价方式。将此方法应用于计算任意数平方P系统的设计,结果表明膜系统踪迹设计法是可行并且有效的。 本文工作得到国家自然科学基金(61170016)和教育部新世纪优秀人才支持计划项目(NCET-11-0715)和教育部留学回国人员科研启动基金项目的共同资助。
[Abstract]:As an important research field in computer science, membrane computing (Membrane Computing, MC) is designed to study computational models abstracted from the structure and function of living cells and from the collaboration of cell groups such as tissues and organs. Its computational model is called membrane system or P system. Membrane system is a kind of computing model with distributed and maximal parallelism. A large number of studies show that many simple membrane systems have the same computing power as Turing machines in theory. Membrane computing is a very active research field at present. Researchers use mathematics formal language and other tools to carry out theoretical research on membrane computing. A variety of P system models have been proposed and many research results have been obtained. At present, all the proposed P system models are carefully designed by researchers, and the design process is complicated, which requires not only solid professional knowledge, but also careful and repeated deduction and demonstration, which is time-consuming and laborious. The research of membrane system design method is still very few, and it needs further research. In this paper, we take the cellular membrane system as the research object, and discuss the method of realizing the automatic optimal design of the membrane system by using the quantum evolutionary algorithm and P-Lingua software. The main work and research results are as follows: 1. The basic theory of membrane computing is summarized. The simulation software of cellular membrane system is briefly introduced, and the simulation software P-Lingua is introduced in detail. The simulation method of cell membrane system in P-Lingua software is illustrated. 2. The design problem of cell membrane system is described and analyzed in detail, the optimization design method of simple cell membrane system is given, and the design idea is introduced in detail. The quantum evolutionary algorithm is used to encode and evaluate the membrane system. The proposed method is applied to the design of a four-square-film system, and the experimental results are better than that of the genetic algorithm. On this basis, the method is extended to calculate the design of arbitrary natural number square membrane system. The effectiveness of the method is verified by experiments. 3. The optimization design method of membrane system is improved by improving the evaluation method and adding the local perturbation strategy to the quantum evolutionary algorithm. By using the improved design method and under the same given conditions, the design of any natural number square P system, algebraic P system and language generating P system is realized by changing the evaluation function respectively. In addition, compared with the traditional quantum evolutionary algorithm and random search algorithm, the results show that the improved design method can achieve better results. 4. The trace design method of cellular membrane system is given. The definition, design idea, coding and evaluation of membrane system are introduced in detail. The method is applied to the design of arbitrary number square P system. The results show that the membrane system trace design method is feasible and effective. The work of this paper is supported by the National Natural Science Foundation of China (61170016), the New Century talents support Program (NCET-11-0715) of the Ministry of Education and the Scientific Research Start-up Fund of returned students of the Ministry of Education.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP38
本文编号：2182088
[Abstract]:As an important research field in computer science, membrane computing (Membrane Computing, MC) is designed to study computational models abstracted from the structure and function of living cells and from the collaboration of cell groups such as tissues and organs. Its computational model is called membrane system or P system. Membrane system is a kind of computing model with distributed and maximal parallelism. A large number of studies show that many simple membrane systems have the same computing power as Turing machines in theory. Membrane computing is a very active research field at present. Researchers use mathematics formal language and other tools to carry out theoretical research on membrane computing. A variety of P system models have been proposed and many research results have been obtained. At present, all the proposed P system models are carefully designed by researchers, and the design process is complicated, which requires not only solid professional knowledge, but also careful and repeated deduction and demonstration, which is time-consuming and laborious. The research of membrane system design method is still very few, and it needs further research. In this paper, we take the cellular membrane system as the research object, and discuss the method of realizing the automatic optimal design of the membrane system by using the quantum evolutionary algorithm and P-Lingua software. The main work and research results are as follows: 1. The basic theory of membrane computing is summarized. The simulation software of cellular membrane system is briefly introduced, and the simulation software P-Lingua is introduced in detail. The simulation method of cell membrane system in P-Lingua software is illustrated. 2. The design problem of cell membrane system is described and analyzed in detail, the optimization design method of simple cell membrane system is given, and the design idea is introduced in detail. The quantum evolutionary algorithm is used to encode and evaluate the membrane system. The proposed method is applied to the design of a four-square-film system, and the experimental results are better than that of the genetic algorithm. On this basis, the method is extended to calculate the design of arbitrary natural number square membrane system. The effectiveness of the method is verified by experiments. 3. The optimization design method of membrane system is improved by improving the evaluation method and adding the local perturbation strategy to the quantum evolutionary algorithm. By using the improved design method and under the same given conditions, the design of any natural number square P system, algebraic P system and language generating P system is realized by changing the evaluation function respectively. In addition, compared with the traditional quantum evolutionary algorithm and random search algorithm, the results show that the improved design method can achieve better results. 4. The trace design method of cellular membrane system is given. The definition, design idea, coding and evaluation of membrane system are introduced in detail. The method is applied to the design of arbitrary number square P system. The results show that the membrane system trace design method is feasible and effective. The work of this paper is supported by the National Natural Science Foundation of China (61170016), the New Century talents support Program (NCET-11-0715) of the Ministry of Education and the Scientific Research Start-up Fund of returned students of the Ministry of Education.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP38
【参考文献】
相关期刊论文 前4条
1 张宗飞;;一种改进型量子遗传算法[J];计算机工程;2010年06期
2 潘林强;张兴义;曾湘祥;汪隽;;脉冲神经膜计算系统的研究进展及展望(英文)[J];计算机学报;2008年12期
3 张兴义;曾湘祥;潘林强;罗斌;;脉冲神经膜系统求解任意两个自然数的乘积[J];计算机学报;2009年12期
4 张葛祥;潘林强;;自然计算的新分支——膜计算[J];计算机学报;2010年02期
相关硕士学位论文 前3条
1 周芬;粒子群膜算法及其应用研究[D];西南交通大学;2011年
2 陈俊伟;基于膜计算优化方法的控制系统设计研究[D];浙江大学;2008年
3 濮阳;生物过程的形式化建模及仿真[D];上海交通大学;2007年
，本文编号：2182088
本文链接：https://www.wllwen.com/kejilunwen/jisuanjikexuelunwen/2182088.html
