Machine learning and model driven bayesian uncertainty quantification in suspended nonstructural systems

• Published in: Reliability engineering & system safety Vol. 237; p. 109392
• Main Authors: Qin, Zhiyuan, Naser, M.Z.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.09.2023
• Subjects: Blackbox variational inference; Gaussian process; Geometric complexity; Inverse problems; Machine learning; Suspended nonstructural systems; Suspended nonstructural systems; Inverse problems; Gaussian process; Blackbox variational inference; Machine learning; Geometric complexity
• ISSN: 0951-8320; 1879-0836
• DOI: 10.1016/j.ress.2023.109392

•A novel framework for the uncertainty quantification of inverse problems is presented.•This framework adopts ML- and model-driven stochastic GP calibration and inference.•The proposed framework is validated through large shaking table tests. This paper presents a novel framework for the uncertainty quantification of inverse problems often encountered in suspended nonstructural systems. This framework adopts machine learning- and model-driven stochastic Gaussian process model calibration to quantify the uncertainty via a new blackbox variational inference that accounts for geometric complexity through Bayesian inference. The soundness of the proposed framework is validated by examining one of the largest full-scale shaking table tests of suspended nonstructural systems and accompanying simulated (numerical) data. Our findings indicate that the proposed framework is computationally sound and scalable and yields optimal generalizability.