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

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.

Automated summary

This paper presents a novel framework for quantifying the uncertainty in the inverse problems of suspended nonstructural systems. The framework combines machine learning and model-driven stochastic Gaussian process model calibration to account for geometric complexity through Bayesian inference. The proposed framework is validated using a large-scale shaking table test and simulated data, showing computational soundness, scalability, and optimal generalizability.