Fuzzy clustering theory based rock mass cuttability classification prediction model for TBM tunnelling

Abstract

In this paper, systematic and quantitative studies are carried out regarding rock cuttability by a TBM under complex geological conditions. Based on the fuzzy clustering theory and construction sample analysis, a rock cuttability classification prediction model is established by using the rate of penetration (ROP) as a basic index. The ROP is related to four property indexes: rock uniaxial compressive strength, rock integrity coefficient, the angle between the rock structural plane and the tunnel axis, and water seepage. In the prediction model, the rock mass cuttability is classified into three levels: good, moderate, and poor. Furthermore the prediction model is refined to improve model precision and geological applicability. According to the prediction results for the West Qinling tunnel and the Dahuofang water conveyance tunnel projects, the ROP of the TBM basically agrees with that of the prediction model, proving the feasibility, scientific soundness, and availability of the rock cuttability classification prediction model and providing a basis for the selection, design, and construction of the TBM.