Dynamic knowledge inference and learning under adaptive fuzzy Petri net framework

Since knowledge in expert system is vague and modified frequently, expert s ystems are fuzzy and dynamic systems. It is very important to design a dyna mic knowledge inference framework which is adjustable according to knowledg e variation as human cognition and thinking. Aiming at this object, a gener alized fuzzy Petri net model is proposed in this paper, it is called adapti ve fuzzy Petri net (AFPN). AFPN not only takes the descriptive advantages o f fuzzy Petri net, but also has learning ability like neural network. Just as other fuzzy Petri net (FPN) models, AFPN can be used for knowledge repre sentation and reasoning, but AFPN has one important advantage: it is suitab le for dynamic knowledge, i.e., the weights of AFPN are ajustable, Based on AFPN transition firing rule, a modified back propagation learning algorith m is developed to assure the convergence of the weights.