Research On Semantic Information Petri Net Based On Ontology And Its Application In System Modeling

The Petri Net theory has been widely applied in many fields such as the system performance evaluation, communication protocols, distributed software systems, distributed databases, flexible manufacturing systems, discrete event systems, neural networks and decision-making model etc. As a formal modeling method, Petri Net provides many ways of performance evaluation and verification on system's concurrence behavior. But the ability of Petri net to describe the semantic information is very limited. Although all types of high-level Petri Net are designed to expand the expression ability. They are lacking of semantic information when comparing with the descriptive modeling method as UML and IDEF etc. Still they lack of all-round support for Object-Oriented mechanism. Further more Petri Net model needs support of specialized modeling software. And Petri Net can't be achieved automatic conversion and information sharing with other modeling methods. As a result of these problems, Petri net is unable to satisfy the demand of design method of the new generation system. Thus seriously constraint the development and application of Petri net in area of control and even distributed software systems and other systems.The advantage of UML and IDEF modeling methods is that they have enough semantic information. And the advantage of Petri net is that it has rigorous formal verification. The thesis is focusing on make use of both of their advantages. The thesis takes it basis aiming that enrich the Petri net with semantic information. It takes Object-Oriented real-time system as the main research content. And it integrates the basic thinking of domain Ontology and Ontology mapping algorithm into Petri net theory. So the thesis put forward the concept of Semantic Information Petri Net (SIPN) and the applied case. The SIPN combined concept of Object-Oriented and Petri net organically and take the Ontology as the model sharing bridge.The core content of this thesis is the definition of the semantic information Petri net and the method of model automatic conversion based of the ontology mapping. First, the thesis gives out the definition of semantic information Petri nets. The SIPN consist of three parts: UML model, model Ontology database and hierarchical Object-oriented Petri net. And the thesis also gives the process of system modeling based on SIPN. Modeling a general system with SIPN firstly the UML model should be established. Then convert the UML model to Hierarchal Object-Oriented Petri Net model automatically based on the use of Ontology library and Ontology mapping algorithm. And then analyze the accuracy and reliability of the system with reachability and boundedness of Petri Net theory. The model ontology database should be established by the domain experts in advance and should be improved during every modeling period. The thesis gives out the advised structure, establish method and steps of the model ontology. According to the advantage of ontology mapping methods available at present and the model ontology characteristic the thesis gives out the uniting semantic mapping algorithm as the basis method of concept mapping during the ontology mapping. At last the thesis takes SIPN as a description for the Autonomous Underwater Vehicle (AUV) control system and information interactive design model. Making it fully describe the information of hardware and software during the design stage of AUV. And the method can formally verify the concurrent conflict and full-featured function of the system during simulation and verification stage. So the method provides the means of verification in advance and optimization for the design.The thesis gives out a system modeling method which combined the Petri Net and Object-Oriented concept and Ontology together. It described the method form the definition and the details of algorithm to the application case. This is a general modeling method for object-oriented real-time system. That is, it can not only describe the system information in details, but also verify the performance of system formally and optimize the system after analysis. Also the thesis verified the feasibility and usefulness of SIPN by modeling the control systems of AUV and analyzing the reliability with it.