Seismic site response predictions improve using scarce sensor data
Seismic Site Response Prediction from Sparse Observations Using Finite-Element-Pretrained Latent Dynamics
Machine Learning
Summary
Predicting how the ground shakes during earthquakes is hard because there are few sensors and records to learn from. The authors developed a method called FLARE-T that uses detailed computer simulations combined with limited real measurements to better predict shaking at different depths underground. Their approach reduces errors compared to traditional models, even with limited data, and works for different shaking intensities and directions. This makes earthquake response forecasts more accurate and less dependent on perfectly calibrated starting models.
What this means in practice
- •For geotechnical engineers: Use sparse seismic sensor data combined with dense simulations to improve predictions of ground shaking at multiple soil depths for site design.
- •For earthquake early warning teams: Enhance site-specific shaking forecasts by calibrating response models with limited sensor records, improving warning accuracy across different shaking magnitudes and directions.
Authors
Yi Zhu, Su Chen, Xiaojun Li
Abstract
Numerical site-response predictions often deviate from observations, yet correcting these discrepancies is difficult because records are limited in both sensor coverage and number of events. This study proposes the Transfer-Enabled Forced Latent Autoencoder for Response Equations (FLARE-T) to improve these predictions by learning and calibrating low-dimensional latent dynamics that connect the base acceleration input to acceleration outputs at multiple depths. FLARE-T learns a low-dimensional response manifold and input-driven dynamics from dense finite-element simulations. It then trains a sparse encoder to map simulated sensor responses into the learned coordinates and uses limited records to calibrate the dynamics within them. A short response window initializes each prediction, while the complete base motion drives the response. The framework was evaluated using a layered-soil centrifuge test and the Lotung field vertical array. Test-set results show that FLARE-T improved multi-depth acceleration histories and 5%-damped pseudoacceleration response spectra relative to the original finite-element models, reducing errors at every evaluated sensor for motions of different intensities and, at Lotung, for both horizontal components. Two Lotung source models with different constitutive parameters achieved comparable test-set accuracy, indicating reduced dependence on precise prior calibration. FLARE-T therefore provides a data-efficient means of combining dense numerical response information with limited field records to improve future site-response predictions.