Assessment of Seismic Liquefaction of Soils Using Swarm-Assisted Optimization Algorithm

Abstract

Assessment of liquefaction potential of soils due to the earthquake has been carried out in this research using the nature-inspired Metaheuristic swarm-assisted algorithm (PSO). An assessment has been made on the basis of actual field data from the previous research. The field data consists of 59 sets having variables of total stress of soil (⌠<inf>o</inf>), effective stress of the soil (⌠′<inf>o</inf>), percentage fines, mean size of soil particles (D<inf>50</inf>), standard penetration value (SPT), the equivalent dynamic shear stress (T<inf>av</inf>/⌠′<inf>o</inf>), maximum horizontal acceleration at ground surface (a/g) and the earthquake magnitude (M). PSO-based models were developed for both single variable and multivariable linear approaches. The results revealed that for the assessment of liquefaction of soils, the developed PSO models perform good estimations in terms of the errors and convergent solution. And also, with a damping coefficient and varying input variables, there is a significant improvement in the best solution. These developed models can be useful for practicing engineers in the field. © 2021, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.