A Possibilistic Petri Net Model for Diagnosing Cracks in RC Structures
Kevin F. R. Liu · Computer-Aided Civil and Infrastructure Engineering · 2003
Abstract: This article proposes a possibilistic Petri net (PPN) model to diagnose cracks in reinforced concrete structures. PPN integrates Petri nets with possibilistic reasoning and maintains the advantages of both formalisms. Two major features of PPN include the possibilistic token to carry information to describe an object and its corresponding possibility and necessity measures, and four types of possibilistic transitions—inference, duplication, aggregation, and aggregation-duplication. A reasoning algorithm, based on possibilistic Petri nets, is also developed to execute PPN. The PPN model and the reasoning algorithm are further used to diagnose possible causes of cracking in RC structures, and three basic cases are considered to demonstrate the applicability of the approach. The reliability of the results is increased by explaining the diagnostic process through the movement of tokens. Moreover, the confidence level associated with each possible cause of concrete cracking can be used to determine the relevance of the diagnosis.