Seismic Reliability Analysis of Structures by an Adaptive Support Vector Regression-Based Metamodel

The dual metamodeling approach is usually adopted to tackle the stochastic nature of earthquakes in seismic reliability analysis relying on the lognormal response assumption. Alternatively, a direct response approximation approach where separate metamodels are constructed for each earthquake is attempted here avoiding prior distribution assumption. Further, an adaptive support vector regression-based metamodeling is proposed that selects new training samples near the failure boundary with due consideration to accuracy and efficiency. The effectiveness of the approach is elucidated by comparing it with the results obtained by the direct Monte Carlo simulation technique and a state-of-the-art active learning-based Kriging approach.

Automated summary

The dual metamodeling approach is used to address the stochastic nature of earthquakes, while avoiding prior distribution assumption, a direct response approximation approach is attempted here. Furthermore, an adaptive support vector regression-based metamodeling is proposed for selecting new training samples near the failure boundary with consideration to accuracy and efficiency. The effectiveness of the approach is demonstrated by comparing it with direct Monte Carlo simulation technique and an active learning-based Kriging approach.