模糊定性建模与仿真集成实现

定性推理是使用定性模型 ,推理演绎物理系统以获取有用信息的技术 ,在系统诊断、功能行为仿真、产品设计与分析等领域得以广泛应用 .其面对的主要挑战是定性模型构造比较困难 ,仿真效率不高 ,得到有效解的仿真过程代价较大 .本文就此进行了研究 :利用模型片断库和面向对象的方法实现图形化建模环境 ,使得建模过程直观且易操作 ;采用模型定性仿真技术 ,拓展了传统定性仿真算法 ,提高了仿真效率 .文章最后给出软件实现和实例验证 .

An Integrated Application of Fuzzy Qualitative Modeling and Simulation

Qualitative reasoning is an inferring technique using qualitative model to derive new knowledge of and gain insight into a physical technical system in some specific issues. It has been applying in areas such as system diagnosis, system functional/behavior simulation, product design and synthesis. The main challenges are that the construction of qualitative model is difficult and its poor reasoning capability. The efficiency of those qualitative techniques is usually low and they can take a while to reach a valuable solution. This paper discusses the implementation of qualitative modeling and fuzzy qualitative simulation. We utilize model fragment library and object-oriented technique to make the modeling process more friendly and ease to use. The introduction of fuzzy sets makes it possible to solve the inherent defection of traditional qualitative reasoning encountered before. Furthermore, we provide ideas on software implementation and practical application.