Assessment of small strain modulus in soil using advanced computational models

Abstract

Abstract Small-strain shear modulus ( $$G_0$$ G 0 ) of soils is a crucial dynamic parameter that significantly impacts seismic site response analysis and foundation design. $$G_0$$ G 0 is susceptible to multiple factors, including soil uniformity coefficient ( $$C_u$$ C u ), void ratio (e), mean particle size ( $$d_{50}$$ d 50 ), and confining stress ( $$\sigma '$$ σ ′ ). This study aims to establish a $$G_0$$ G 0 database and suggests three advanced computational models for $$G_0$$ G 0 prediction. Nine performance indicators, including four new indices, are employed to calculate and analyze the model’s performance. The XGBoost model outperforms the other two models, with all three models achieving $$R^2$$ R 2 values exceeding 0.9, RMSE values below 30, MAE values below 25, VAF values surpassing 80%, and ARE values below 50%. Compared to the empirical formula-based traditional prediction models, the model proposed in this study exhibits better performance in IOS, IOA, a20-index, and PI metrics values. The model has higher prediction accuracy and better generalization ability.