A fuzzy Petri net-based expert system and its application to damageassessment of bridges

摘要

In this paper, a fuzzy Petri net approach to modeling fuzzynrule-based reasoning is proposed to bring together the possibilisticnentailment and the fuzzy reasoning to handle uncertain and impreciseninformation. The three key components in our fuzzy rule-basednreasoning-fuzzy propositions, truth-qualified fuzzy rules, andntruth-qualified fuzzy facts-can be formulated as fuzzy places, uncertainntransitions, and uncertain fuzzy tokens, respectively. Four types ofnuncertain transitions-inference, aggregation, duplication, andnaggregation-duplication transitions-are introduced to fulfil thenmechanism of fuzzy rule-based reasoning. A framework of integratednexpert systems based on our fuzzy Petri net, called fuzzy Petrinnet-based expert system (FPNES), is implemented in Java. Major featuresnof FPNES include knowledge representation through the use ofnhierarchical fuzzy Petri nets, a reasoning mechanism based on fuzzynPetri nets, and transformation of modularized fuzzy rule bases intonhierarchical fuzzy Petri nets. An application to the damage assessmentnof the Da-Shi bridge in Taiwan is used as an illustrative example ofnFPNES