Uncertainty often recurs in structural system characterization as well as in choosing the mechanical model and in calibrating it. When analyzing a structure founded in cohesionless soils, the uncertainty in system modeling comes from soil inherent variability, site conditions, construction tolerance, and failure mechanisms. In this research, a Fuzzy-Neural Network method to predict the behavior of structures built on complex cohesionless soils is proposed. The method is based on an ArtificialNeural Network (ANN) for modeling the soil-foundation interaction. Its learning process analyzes over 200 records of building foundations, tanks, and embankments settlements on sand and gravel. Once validated, ANN is introduced in the soil-foundation-sovrastructure interaction model. Using fuzzy sets to define vague and ambiguous variables, the Fuzzy-Neural Network method predicts the system behavior and quantifies the uncertainty of its response. A numerical example shows the method effectiveness in the case of uncertainty in soil parameters and gives suggestions for successive applications.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.