Development of a hard rock TBM performance prediction model using RMR input parameters

Abstract

Despite the widespread use of TBMs in tunneling projects, accurate estimation of TBM performance, especially in various geological conditions, can still be challenging. Comparing the most common rock mass classification systems, the RMR classification system shows a better correlation with the TBM penetration rate, due to using uniaxial compressive strength (UCS) as an input parameter in this classification system. RMR parameters are available in most tunneling projects, but it should be noted that the RMR system, was developed to classify the rock mass conditions in terms of tunnel stability and support design, and the effective parameters were selected accordingly. So, when the target is to use the RMR input parameters to predict the TBM performance, modifications should be made in the parameters. This study is offered a new TBM performance prediction model based on the input parameters of the Basic RMR in different rock types, through machine learning-based regression analysis. The proposed model has been developed based on the analysis of a comprehensive database of TBM performance in various rock types. Because different rock types have different fabrics and respond differently to shear forces, incorporating the effects of rock type into performance prediction models can improve the accuracy of estimates, for this reason, the lithology parameter is also included in the model. The model can be practical, especially in the design phase of a tunneling project. © 2024 The Author(s).