Risk Assessment and Estimation of TBM Penetration Rate Using RES-Based Model

Abstract

The tunnel boring machine (TBM) penetration rate (PR) estimation is one of the crucial and complex tasks encountered frequently to excavate the mechanical tunnels. Predicting the PR is a nonlinear and multivariable complex problem and establishing relationship between rock properties and PR is not a simple task. To overcome this problem, in this paper, based on the basic concepts of a rock engineering systems (RES) approach, a model for the estimation of the PR and the risk associated is presented. The suggested model involves 6 effective parameters on PR (uniaxial compressive strength (UCS), ratio of the maximum load (PSI), average distance between planes of weakness (DPW), angle between tunnel axis and the planes of weakness (Alpha), Brazilian tensile strength (BTS) and rock fracture class (RFC)), while retaining simplicity as well. The performance of the RES model is compared with multiple regression models. The estimation abilities offered using RES and multiple regression models were presented by using field data given in open source literatures. The results indicate that the RES based model predictor with less mean square error (MSE) and higher coefficient of determination (R2) performs better than the other models (used in this study).