Predicting the Maximum Seismic Response of the Soil-Pile-Superstructure System Using Random Forests

ABSTRACT

Seismic fragility analysis has been considered an efficient method for seismic performance assessment of soil-pile-superstructure systems (SPSSs). However, seismic fragility analysis based on widely used numerical methods often requires a significant investment of time. In this study, a new approach based on the random forest technique, which can be applied to fragility analysis, is proposed to accurately and efficiently predict the maximum seismic response of SPSSs. Moreover, based on the random forest technique, the velocity RMS is found to exhibit a stronger correlation with the maximum seismic response of SPSSs in liquefiable soils.

KEYWORDS:

• Seismic response
• soil-pile-superstructure system
• finite element analysis
• ground motion parameters
• random forests
• machine learning

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work is supported by the National Natural Science Foundation of China [Grant Nos. 41902287, 51908153, and 52020105002], The Science and Technology Planning Project of Guangdong Province [2020A1414010284], and the Science and Technology Planning Project of Guangzhou [202102010436].