Soil Liquefaction Assessment Using Soft Computing Approaches Based on Capacity Energy Concept

Abstract

Soil liquefaction is one of the most complicated phenomena to assess in geotechnical earthquake engineering. The conventional procedures developed to determine the liquefaction potential of sandy soil deposits can be categorized into three main groups: Stress-based, strain-based, and energy-based procedures. The main advantage of the energy-based approach over the remaining two methods is the fact that it considers the effects of strain and stress concurrently unlike the stress or strain-based methods. Several liquefaction evaluation procedures and approaches have been developed relating the capacity energy to the initial soil parameters, such as the relative density, initial effective confining pressure, fine contents, and soil textural properties. In this study, based on the capacity energy database by Baziar et al. (2011), analyses have been carried out on a total of 405 previously published tests using soft computing approaches, including Ridge, Lasso & LassoCV, Random Forest, eXtreme Gradient Boost (XGBoost), and Multivariate Adaptive Regression Splines (MARS) approaches, to assess the capacity energy required to trigger liquefaction in sand and silty sands. The results clearly prove the capability of the proposed models and the capacity energy concept to assess liquefaction resistance of soils. It is also proposed that these approaches should be used as cross-validation against each other. The result shows that the capacity energy is most sensitive to the relative density.

Keywords

• soil liquefaction
• capacity energy
• Ridge
• Lasso &amp
• LassoCV
• Random Forest