Prediction of hard rock TBM penetration rate using least square support vector machine

The prediction of tunnel boring machine (TBM) penetration rate is helpful to plan construction time and to control cost. This paper studies on modeling the penetration rate and the rock mass parameters via least square support vector machine (LS-SVM). These rock properties include uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), peak slope index (PSI), distance between planes of weakness (DPW) and the alpha angle. The correlation between the rock parameters and the measured rate of penetration (ROP) is first investigated via LS-SVM. Then the necessity of normalizing the rock parameters is discussed. Finally, comparisons between the LS-SVM and other methods are presented to demonstrate the applicability of LS-SVM for predicting ROP.