CASIA OpenIR  > 毕业生  > 博士学位论文
基于模糊逻辑的软计算在混杂控制系统中的应用研究
任剑
Subtype工学博士
Thesis Advisor郑应平
1998-05-01
Degree Grantor中国科学院自动化研究所
Place of Conferral中国科学院自动化研究所
Degree Discipline控制理论与控制工程
Keyword模糊混杂控制系统 混杂控制系统 软计算 模糊逻辑 神经网络 遗传算法 模糊神经网络 Fuzzy Hybrid Control System Hybrid Control System Soft Computing Fuzzy Logic Neural Network Genetic Algorithm Fuzzy-neural Net
Abstract从信息角度出发,混杂控制系统提供了一种利用和处理混杂的定性和定量信 息以及混杂的量变和质变动态特性的体系结构和方法,强调控制与决策的高度统 一,是研究一大类复杂系统控制的一条有效途径,“研究混杂控制系统的最终目 标是实现复杂系统的智能控制”。而“作为智能系统建构和设计重要基础之一的 软计算,在增强系统的智能性方面也正发挥着越来越重要的作用”;因此,研究 软计算在混杂控制系统中的应用,将有助于解决混杂控制系统研究中现存的某些 问题,并加深对软计算的理解,同时也将给智能控制的研究以有益启发。 论文的主要工作和贡献是: · 首先,分别对软计算和混杂控制系统的研究和应用现状进行了慨述,指 出了各自存在的主要问题,并阐明了其方法论意义。 ·第二,根据混杂控制系统的特点,并参考一般动态系统和一般动态模糊 系统的定义,提出了一种适于混杂控制系统建模需要的模糊动态系统模型:主要 特色是:利用语言变量的原子语言值、复合语言值,以及模糊蕴涵关系和模糊规 则集等基本慨念,对所提模型作出了比较清楚的形式化描述、分析和解释。通过 将模糊动态系统模型同离散事件系统模型进行比较,表明其土要特点是:较强的 利用和处理混杂的定性和定量信息的能力;具备描述定性层次动态特性的条件, 适于处理有顺序要求或逻辑要求的问题;不但能处理精确的离散状态,还能处理 这些离散状态之间的过渡状态;通过语言变量的语法规则和语义规则的调整,乃 至模糊推理规则的调整,可具备一定的自适应能力。 ·第三,基于上面所提出的模糊动态系统,提出了一种新型的模糊混杂控 制系统模型,给出了其一般结构框架和形式化描述,并通过具体的建模实例,描 述了其建模过程的一般步骤;对比分析了它相对基于离散状态的混杂控制系统的 区别和优势,以及它相时常规模糊逻辑控制系统的区别和优势;另外对其典型应 用进行了分类。研究表明:由于能够对离散状态之间的过渡状态进行有效处理, 从而增强了其克服不确定性因素影响的能力,并有利于确定适当的决策和控制来 保证实现所期望的状态演化顺序和逻辑关系等定性层次的要求;分层次的建模和 形式化过程,有助于清楚地分解和表示PSC(Plan、Schedule、Control)型 任务的不同要求,并方便系统整体性能的分析和验证,进而有利于系统的逐步实 现,尤其是计算机控制的实现;利用模糊神经网络、遗传算法
Other AbstractHybrid Control Systems ( HCS ) are a kind of complex systems, which typically possess a hierarchical structure, characterized by continuous variable dynamics at the lowest level and logical decision-making at the highest. The final target of HCS research is to achieve the intelligent control for complex system. As one of the most important foundations for intelligent control . Soft-Computing ( SC ) is becoming more and more popular in the design and deployment of intelligent systems . The application of SC in HCS will be helpful to improve the intelligence of HCS and eventually lead to so-called hybrid intelligent systems. This thesis focuses on the application of fuzzy logic . neural network in the modeling, analysis and design of hierarchical HCS. The main works include : · First after summarizing the HCS and SC , we pointed out their methodological meanings. · Second. a kind of Fuzzy Dynamic System ( FDS ) model based on concept of general dynamic fuzzy system is proposed, and its formal description is presented. It is analyzed and explained by using atomic linguistic values, synthesized linguistic values, fuzzy rules and other fuzzy logic theories & methods. Its ability of utilizing hybrid qualitative and quantitative information is proved. Also its advantage over DEDS and the difference between them are presented. · Third . a new kind of Fuzzy Hybrid Control System ( FHCS ) model is proposed, and its general structure and formal description are presented. The modeling process is demonstrated through a case study. The difference between FHCS and common HCS that based on discrete states, and the advantage of FHCS over common HCS are discussed. The difference between FHCS and common Fuzzy Logic Control System ( FLCS ) and the advantage of FHCS over FLCS are presented, too. Moreover. the main application of FHCS are classified into 3 types. ~ Forth, we analyze the stability of FHCS in the continuos domain by using the common Lyapunov theory and Multiple Lyapunov Function( MLF ) theory, and 4 theorems for stability analysis are proposed. The continuity problem is discussed, and we prove that the continuous mappings from the measurement to control spaces are achieved in FHCS. and the validity of common FLCS is also explained from the view of FHCS . We also analyze the characteristics of the local control law and global control law in FHCS qualitatively at qualitative level. · Fifth, we proposed Direct Design Method ( DDM ) for the Class I application of FHCS and Indirect Design Method (IDM )for the Class II application of FHCS. DDM does not depend on the plant model, and the controller is obtained through online tuning A DDM based on FNN & Reinforcement Learning Algorithm is proposed. IDM does depend on the plant model, and can fully utilized the stability design methods for linear systems. 2 IDMs based on LQ method and pole placement method are presented. The validit
shelfnumXWLW454
Other Identifier454
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/5681
Collection毕业生_博士学位论文
Recommended Citation
GB/T 7714
任剑. 基于模糊逻辑的软计算在混杂控制系统中的应用研究[D]. 中国科学院自动化研究所. 中国科学院自动化研究所,1998.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[任剑]'s Articles
Baidu academic
Similar articles in Baidu academic
[任剑]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[任剑]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.