A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Site amplification prediction model of shallow bedrock sites based on machine learning models
https://orcid.org/0000-0002-5453-1582
https://orcid.org/0000-0002-6337-4785
https://orcid.org/0000-0002-3292-9194
Abstract Prediction of the site amplification is of primary importance for a site-specific seismic hazard assessment. A large suite of both empirical and simulation-based site amplification models has been proposed. Because they are conditioned on a few simplified site proxies including time-averaged shear wave velocity up to a depth of 30 m (V S30) and site period (T G), they only provide approximate estimates of the site amplification. In this study, site amplification prediction models are developed using two machine learning algorithms, which are random forest (RF) and deep neural network (DNN). A comprehensive database of site response analysis outputs obtained from simulations performed on shallow bedrock profiles is used. Instead of simplified site proxies and ground motion intensity measures, matrix data which include the response spectrum of the input ground motion and shear wave velocity profile. Both machine learning based models provide exceptional prediction accuracies of both the linear and nonlinear amplifications compared with the regression-based model, producing accurate predictions of both binned mean and standard deviation of the site amplification. Among two machine learning techniques, DNN-based model is revealed to produce better predictions.
Highlights We develop ML-based site amplification models for shallow bedrock sites. We use random forest and deep neural network algorithms to train site response. Main inputs are matrix data, which include response spectrum and Vs profile. ML-based models provide exceptional predictions of surface response spectra. ML-based models are shown to have the potential to replace numerical models.
Site amplification prediction model of shallow bedrock sites based on machine learning models
https://orcid.org/0000-0002-5453-1582
https://orcid.org/0000-0002-6337-4785
https://orcid.org/0000-0002-3292-9194
Abstract Prediction of the site amplification is of primary importance for a site-specific seismic hazard assessment. A large suite of both empirical and simulation-based site amplification models has been proposed. Because they are conditioned on a few simplified site proxies including time-averaged shear wave velocity up to a depth of 30 m (V S30) and site period (T G), they only provide approximate estimates of the site amplification. In this study, site amplification prediction models are developed using two machine learning algorithms, which are random forest (RF) and deep neural network (DNN). A comprehensive database of site response analysis outputs obtained from simulations performed on shallow bedrock profiles is used. Instead of simplified site proxies and ground motion intensity measures, matrix data which include the response spectrum of the input ground motion and shear wave velocity profile. Both machine learning based models provide exceptional prediction accuracies of both the linear and nonlinear amplifications compared with the regression-based model, producing accurate predictions of both binned mean and standard deviation of the site amplification. Among two machine learning techniques, DNN-based model is revealed to produce better predictions.
Highlights We develop ML-based site amplification models for shallow bedrock sites. We use random forest and deep neural network algorithms to train site response. Main inputs are matrix data, which include response spectrum and Vs profile. ML-based models provide exceptional predictions of surface response spectra. ML-based models are shown to have the potential to replace numerical models.
Site amplification prediction model of shallow bedrock sites based on machine learning models
https://orcid.org/0000-0002-5453-1582
https://orcid.org/0000-0002-6337-4785
https://orcid.org/0000-0002-3292-9194
Abstract Prediction of the site amplification is of primary importance for a site-specific seismic hazard assessment. A large suite of both empirical and simulation-based site amplification models has been proposed. Because they are conditioned on a few simplified site proxies including time-averaged shear wave velocity up to a depth of 30 m (V S30) and site period (T G), they only provide approximate estimates of the site amplification. In this study, site amplification prediction models are developed using two machine learning algorithms, which are random forest (RF) and deep neural network (DNN). A comprehensive database of site response analysis outputs obtained from simulations performed on shallow bedrock profiles is used. Instead of simplified site proxies and ground motion intensity measures, matrix data which include the response spectrum of the input ground motion and shear wave velocity profile. Both machine learning based models provide exceptional prediction accuracies of both the linear and nonlinear amplifications compared with the regression-based model, producing accurate predictions of both binned mean and standard deviation of the site amplification. Among two machine learning techniques, DNN-based model is revealed to produce better predictions.
Highlights We develop ML-based site amplification models for shallow bedrock sites. We use random forest and deep neural network algorithms to train site response. Main inputs are matrix data, which include response spectrum and Vs profile. ML-based models provide exceptional predictions of surface response spectra. ML-based models are shown to have the potential to replace numerical models.
Site amplification prediction model of shallow bedrock sites based on machine learning models
Lee, Yong-Gook (author) / Kim, Sang-Jin (author) / Achmet, Zeinep (author) / Kwon, Oh-Sung (author) / Park, Duhee (author) / Di Sarno, Luigi (author)
2023-01-07
Article (Journal)
Electronic Resource
English
Prediction of Site Amplification of Shallow Bedrock Sites Using Deep Neural Network Model
| British Library Conference Proceedings | 2022
Site Response Study and Amplification Factor for Shallow Bedrock Sites
| Springer Verlag | 2020
| Taylor & Francis Verlag | 2022
Site Classification System and Site Coefficients for Shallow Bedrock Sites in Korea
| Taylor & Francis Verlag | 2018