Pushover-ML: A Machine Learning approach to predict a trilinear approximation of pushover curves for low-rise reinforced concrete frame buildings
The seismic design of low-rise RC building frames often relies on elastic procedures, limiting the evaluation of nonlinear behavior due to practical constraints such as computational cost. While the research community has applied Machine Learning (ML) to predict the seismic response, existing tools...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-05-01
|
| Series: | SoftwareX |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352711025000895 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The seismic design of low-rise RC building frames often relies on elastic procedures, limiting the evaluation of nonlinear behavior due to practical constraints such as computational cost. While the research community has applied Machine Learning (ML) to predict the seismic response, existing tools often require prior knowledge and expertise to manage dependencies, configure programming environments, and execute code in languages such as Python. This paper introduces Pushover-ML, a graphical user interface (GUI) designed to efficiently predict a trilinear approximation of pushover curves for low-rise RC frames using an ML-based approach. The user-friendly executable provides insights into the structure's seismic capacity through the yielding, maximum capacity, and collapse points of the pushover curve. Pushover-ML bridges the gap between advanced ML techniques and practical engineering applications, enabling accurate and efficient seismic response predictions. |
|---|---|
| ISSN: | 2352-7110 |