Study on the Prediction of the Uniaxial Compressive Strength of Rock Based on the SSA-XGBoost Model

The uniaxial compressive strength of rock is one the important parameters characterizing properties masses in geotechnical engineering. To quickly and accurately predict rock, a new SSA-XGBoost optimizer prediction model was produced to 290 samples. With four parameters, namely, porosity (n,%), Schmidt rebound number (Rn), longitudinal wave velocity (Vp, m/s), point load (Is(50), MPa) as input variables (UCS, output variables, built based on model. verify effectiveness model, empirical formulas, XGBoost, SVM, RF, BPNN, KNN, PLSR, other models were also established compared with All evaluated using root mean square error (RMSE), correlation coefficient (R2), absolute (MAE), variance interpretation (VAF). results calculated by (R2 = 0.84, RMSE 19.85, MAE 14.79, VAF 81.36), are best among all models. Therefore, for dataset tested.