A general framework for data-driven uncertainty quantification under complex input dependencies using vine copulas

• Published in: Probabilistic engineering mechanics Vol. 55; pp. 1 - 16
• Main Authors: Torre, Emiliano, Marelli, Stefano, Embrechts, Paul, Sudret, Bruno
• Format: Journal Article
• Language: English
• Published: Barking Elsevier Ltd 01.01.2019; Elsevier Science Ltd; Elsevier
• Subjects: Applications; Computer simulation; Dependence; Failure analysis; Finite element analysis; Finite element method; Importance sampling; Input dependencies; Machine Learning; Methodology; Multivariate analysis; Polynomial chaos expansions; Polynomials; Probabilistic inference; Probabilistic models; Reliability analysis; Reliability engineering; Representations; Statistical analysis; Statistics; Structural reliability; Uncertainty; Uncertainty quantification; Vine copulas; Uncertainty quantification; Input dependencies; Reliability analysis; Polynomial chaos expansions; Vine copulas

Systems subject to uncertain inputs produce uncertain responses. Uncertainty quantification (UQ) deals with the estimation of statistics of the system response, given a computational model of the system and a probabilistic model of its inputs. In engineering applications it is common to assume that...