Predicting growth and interaction of multiple cracks in structural systems using Dynamic Bayesian Networks

Digital twins have the potential to improve future health prognosis and operational safety of engineering structures. However, few studies attempted validate such with experimental data. In this work, two system-level digital twin models for structural capacity degradation are evaluated data from a four-crack fatigue specimen. Each model is based on dynamic Bayesian network, but differ in complexity crack-to-crack interaction model. Using measured crack lengths experiment, ability each predict evolution size assessed. For specimen, high level must be captured realistically trace system. The suitability network type problem discussed.