Temporal Knowledge Reasoning Algorithm Based On Petri Nets

With the development of science and technology, especially information technology, the scope of our research gets larger and larger, the content is much richer and the uncertain factors emerge in endlessly. The present research includes these fields such as qualitative and quantitative behaviors, certainty and uncertainty issues and hybrid issues of multi-system etc. Time is temporal knowledge with uncertainty, which is important in many application systems. The present representations of temporal knowledge use point of time and time interval. The usual representation of single point of time is impractical and not integrated and time interval is difficult to do quantitative analysis. The corresponding temporal knowledge reasoning is not accurate or clear. The process of reasoning algorithm based on time Petri net is more complex.Based on existed Petri net theories, the paper gives prominence to uncertainty in system modeling and constructs a kind of representation of temporal uncertainty and system analysis method for complex systems. Combining Petri net theory and fuzzy set theory, we do quantitative analysis for temporal uncertainty based on system modeling. The paper defines a fuzzy time Petri net (FTPN) adopting four fuzzy set theoretic functions of time called fuzzy timestamp, fuzzy enabling time, fuzzy occurrence time and fuzzy delay, to deal with temporal uncertainty of some complex systems. According to fuzzy time functions a temporal knowledge reasoning algorithm is presented and the calculation methods of latest, earliest and Min are discussed in this paper. The train operation instances show that the method can efficiently validate train operation plan, represent temporal knowledge and reason. Compared with existed methods, this method has some outstanding characteristics such as quantitative analysis, simple computation, system simplifying and convenience for system integrating.Fuzzy time Petri net can represent temporal uncertainty knowledge in the process of modeling many complex systems. The algorithm based on fuzzy time Petri net can do quantitative analysis for temporal uncertainty and can be applied to intelligent traffic, communication control of computers etc. In addition, FTPN model can integrate with agent technology and various Petri net tools and it is significant for research of various intelligent expert systems.