What question did this study set out to answer?

To develop and evaluate machine learning models capable of rapidly predicting the seismic response and limit-state performance of reinforced concrete moment-resisting frames.

March 14, 2023Archives of Civil and Mechanical EngineeringOpen Access

Machine learning-based seismic response and performance assessment of reinforced concrete buildings

View Full Paper

Ask AI

Bookmark

Authors

FKFarzin KazemiGdańsk University of Technology NANeda AsgarkhaniGdańsk University of Technology RJRobert JankowskiGdańsk University of Technology

Discussion

Member takes

Overview

Computational study demonstrates machine learning models predict seismic drift in reinforced concrete frames, suggesting efficient performance assessment without heavy computational costs.

Key Points

To develop and evaluate machine learning models capable of rapidly predicting the seismic response and limit-state performance of reinforced concrete moment-resisting frames.
Generated 92,400 data points by performing incremental dynamic analyses across 165 reinforced concrete moment-resisting frame configurations (2 to 12 stories; bay lengths of 5.0 m, 6.1 m, and 7.6 m) subjected to near-fault ground motions.
Trained and refined machine learning models in Python, including artificial neural networks and extreme gradient boosting, to predict structural response parameters.
Validated the generalized framework using an independent five-story reinforced concrete building and integrated the models into a graphical user interface.
Optimized machine learning algorithms achieved superior R² values for estimating the maximum interstory drift ratio across various frame geometries.
Trained artificial neural networks and extreme gradient boosting algorithms accurately predicted median incremental dynamic analysis curves, facilitating rapid seismic limit-state assessment.
Validation against a five-story reinforced concrete test structure demonstrated strong predictive capability while substantially cutting analytical time and computational cost.

Cite This Study

Kazemi et al. (2023) studied this question.

synapsesocial.com/papers/69d8a2741dfc3877cabed972 https://doi.org/10.1007/s43452-023-00631-9

View Full Paper

Ask AI

Bookmark

March 14, 2023Archives of Civil and Mechanical EngineeringOpen Access

Machine learning-based seismic response and performance assessment of reinforced concrete buildings

View Full Paper

Ask AI

Bookmark

Authors

FKFarzin KazemiGdańsk University of Technology NANeda AsgarkhaniGdańsk University of Technology RJRobert JankowskiGdańsk University of Technology

Discussion

Member takes

Overview

Computational study demonstrates machine learning models predict seismic drift in reinforced concrete frames, suggesting efficient performance assessment without heavy computational costs.

Key Points

To develop and evaluate machine learning models capable of rapidly predicting the seismic response and limit-state performance of reinforced concrete moment-resisting frames.
Generated 92,400 data points by performing incremental dynamic analyses across 165 reinforced concrete moment-resisting frame configurations (2 to 12 stories; bay lengths of 5.0 m, 6.1 m, and 7.6 m) subjected to near-fault ground motions.
Trained and refined machine learning models in Python, including artificial neural networks and extreme gradient boosting, to predict structural response parameters.
Validated the generalized framework using an independent five-story reinforced concrete building and integrated the models into a graphical user interface.
Optimized machine learning algorithms achieved superior R² values for estimating the maximum interstory drift ratio across various frame geometries.
Trained artificial neural networks and extreme gradient boosting algorithms accurately predicted median incremental dynamic analysis curves, facilitating rapid seismic limit-state assessment.
Validation against a five-story reinforced concrete test structure demonstrated strong predictive capability while substantially cutting analytical time and computational cost.

Cite This Study

Kazemi et al. (2023) studied this question.

synapsesocial.com/papers/69d8a2741dfc3877cabed972 https://doi.org/10.1007/s43452-023-00631-9

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark