Petri nets modeling for dynamic fuzzy anticipatory systems

Anticipatory systems are very dynamic because of the continuous contributions in their behavior of associated anticipatory models (including human experts) and changes of objectives. Therefore it would be very useful to have modeling approaches capable to adjust the model parameters according to the system dynamics. Aiming at this objective, a knowledge base model is proposed, implemented as a generalized fuzzy Petri net model. This model called AFPNM (Anticipatory Fuzzy Petri Net Model), has both the features of a fuzzy Petri net and the learning ability of a neural network. After being trained, an AFPNM can be used as a model for dynamic knowledge representation and inference related to the anticipatory system.