Parameter Identification of Consolidation Settlement Based on Multi-Objective Optimization

Due to the complexity of natural ground condition, consolidation settlement is usually difficult to predict. In this study, a Pareto multi-objective optimization based back analysis method for consolidation settlement is presented in this study. The model is a coupled flow and deformation model for unsaturated soil foundation which is implemented in the interactive multiphysics software environment COMSOL. A multi-objective optimization algorithm AMALGAM is adopted to identify soil parameters based on multiple types of measurements. A case history of a highway trial embankment is used to demonstrate the proposed back analysis method. The observed displacement and pore-water pressures are utilized simultaneously to estimate the mechanical and hydraulic parameters of the soil. The results show that the bi-objective Pareto front exhibits a sharp rectangular pattern. When only displacement is used in back analysis, the numerical model with optimized soil parameters cannot simulate pore-water pressure very well, and vice versa. However, the back analyzed soil parameters of the compromise solution from the bi-objective back analysis can reasonably simulate both the displacement and pore-water pressure and predict the settlement well.