Recorded Versus Synthetic Ground Motions: A Comparative Analysis of Structural Seismic Responses

2025-02-26

AI Summary

This study systematically evaluates the applicability discrepancy between synthetically generated ground motions (produced by stochastic spectral-compatible models) and recorded ground motions in structural response prediction. Method: Employing uniform response spectrum matching to control spectral characteristics, the study conducts Monte Carlo simulation, Sobol’ global sensitivity analysis, and extreme-value statistics—including tail-index estimation—to quantitatively compare probabilistic distributions, higher-order moments, tail behavior, and long-period structural displacements. Contribution/Results: For the first time under spectral consistency constraints, it decouples aleatory from epistemic uncertainty and concurrently quantifies extreme-response discrepancies. Results indicate that overall response deviations are generally ≤15%, supporting synthetic motions for conventional design; however, deviations exceed 50% for long-period drifts, higher-order moments, and tail behavior—revealing the inability of current stochastic models to reproduce the non-Gaussian extreme dynamics inherent in real earthquake records.

Technology Category

• Reasoning under Uncertainty: Stochastic Optimization
• Machine Learning: Calibration & Uncertainty Quantification
• Planning, Routing, and Scheduling: Scheduling under Uncertainty

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Web data generation and simulation
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

This paper presents a comparative analysis of structural seismic responses under two types of ground motion inputs: (i) synthetic motions generated by stochastic ground motion models and (ii) recorded motions from an earthquake database. Five key seismic response metrics - probability distributions, statistical moments, correlations, tail indices, and variance-based global sensitivity indices - are systematically evaluated for two archetypal structures: a 12-story medium-period building and a high-rise long-period tower. Both ground motion datasets are calibrated to a shared response spectrum, ensuring consistency in spectral characteristics, including spectral median, variance, and correlation structure. The analysis incorporates both aleatory uncertainties from ground motion variability and epistemic uncertainties associated with structural parameters, providing a comprehensive comparison of seismic responses. The results demonstrate close agreement in global response characteristics, including distributions, correlations, and sensitivity indices, between synthetic and recorded motions, with differences typically within 15%. However, significant discrepancies are observed under extreme conditions, particularly in tail behavior, higher-order moments, and drift responses of long-period structures, with differences exceeding 50%. These discrepancies are attributed to the non-Gaussian features and complex characteristics inherent in recorded motions, which are less pronounced in synthetic datasets. The findings support the use of synthetic ground motions for evaluating global seismic response characteristics, while highlighting their limitations in capturing rare-event behavior and long-period structural dynamics.

Problem

• Compares structural seismic responses from synthetic and recorded ground motions
• Evaluates five stochastic response metrics for medium and long-period structures
• Assesses suitability of synthetic motions for design versus rare-event limitations

Innovation

• Synthetic motions replicate recorded global seismic responses
• Synthetic motions show limitations in long-period extreme events
• Stochastic models calibrated to shared target response spectrum