| Petri nets are a modeling and analyzing tool of nets. Its main characteristics include the fallowing items: parallelism, uncertainty, asynchronous, step-by-step description ability and analyzing ability, It could be expressed by graph directly; it also has the profound background foundation of mathematics. Petri net system can describe the order and parallelism of events precisely. It is a powerful tool to study asynchronous concurrent systems. The important information of the simulated system and its behavior can be revealed through the network model of the actual system. The information can be used to conduct behavior estimates and analysis of the actual system analysis. However, some of the system with fuzzy behavior can not be described very clearly by the traditional Petri nets. Therefore, considering ambiguity of states and transitions, this thesis combines classic Petri nets with fuzzy sets, emphasizing the system behavior of fuzzy Petri nets, to adapt to requirements of various fuzzy systems. Fuzzy Petri nets Petri nets expand the knowledge representation and knowledge acquisition method.First of all, the state of art of Petri nets is reviewed, concerning the current theory and applications of Petri nets, and as well the research purpose and significance of studying Petri nets is described. Next, a new method to find all minimal T-invariant for a net based dynamic analysis method of Petri net is presented. The basic idea of this method is to find the T-invariant of a random net by solving the full rank state equation and getting the reachable vector. Some examples are given to illustrate the effectiveness and superiority of the algorithm. Secondly, a new kind of fuzzy dynamic Petri net model (FDPN) is proposed to address the vague information. The Fuzzy Petri nets model with cycles are firstly studied and formal definition of FDPN and its relation matrix are given. Based on the relation matrix, fuzzy dynamic effective algorithm is proposed to study the dynamic process of FDPN. Then the reasoning algorithm of this model is proposed. The efficiency is verified by examples in the thesis. Thirdly, FDPN language was first proposed in this paper and the features of FDPN such as aliveness and deadlock-free are described through FDPN language. FDPN language reflects the order of FDPN behavior to some extent. For a simple uncertainty system, we can directly use the FDPN modeling, but for design and analysis of the complexity fuzzy system, composition is an important method. Finally, the synchronization composition of FDPN is described. The thesis proposes the analysis of aliveness based on FDPN language through the reachability analysis. The characteristics of synchronism FDPN language are obtained. This paper originally remodels the fuzzy Petri net and introduces the FDPN language of it to study the behavior features of the model. Further, FDPN language is used in synchronization composition analysis, which approaches the reality clearly in the network description. Finally, a comprehensive summary to the thesis is given and the drawbacks and future research directions are pointed out. |
PDF Full Text Request