Advanced Self-adaptation Learning and Inference Techniques for Fuzzy Petri Net Expert System Units

Abstract

In a complicated expert reasoning system, it is inefficient for commonly fuzzy production rules to depict the vague and modified knowledge. Fuzzy Petri nets are more accurate for dynamic knowledge proposition in describing expert knowledge. However, the bad learning ability of fuzzy Petri net constrains its application in dynamic knowledge expert system. In this paper, an advanced self-adaptation learning way based on error back-propagation is proposed to train parameters of fuzzy production rules in fuzzy Petri net. In order to enhance reasoning and learning efficiency, fuzzy Petri net is transformed into hierarchy model and continuous functions are built to approximate transition firing and fuzzy reasoning. Simulation results show that the designed advanced learning way can make rule parameters arrive at optimization rapidly. These techniques used in this paper are quite effective and can be applied to most practical Petri net models and fuzzy expert systems.