A significant part of the energy consumed during the tunnelling process of Earth Pressure Balanced (EPB) Tunnel Boring Machines (TBMs) is related to the main drive, consisting of a set of motors driving the rotation of the cutting wheel. An energy efficient EPB design requires the optimization of the main drive to avoid over- or under powering of the machine. Key aspect is therefore a precise and reliable estimation of the expected cutting wheel torque. In this paper state-of-the-art torque estimation models are compared to supervised machine learning (ML) approaches, including classification and regression trees (CART), support vector machines (SVM), Gaussian process regression (GPR) and decision tree ensembles (DTE). Feature selection algorithms are compared to models using manually selected input features. ML models are evaluated using accuracy metrics, residual analyses, and model validation. Torque prediction for a real-world validation project shows that utilization rates can be increased distinctively due to the application of ML techniques. @en

In recent years, numerous models have been developed to predict the torque requirements of Tunnel Boring Machines operating in soft ground. Friction between the TBM cutterhead and surrounding ground is a main contribution to the required torque. As a result, friction coefficients are a major input factor in recent torque estimation models. Conditioned soil samples have been tested in a direct shear box, to assess whether a common lab setup can be used to improve the determination of TBM-soil friction coefficients. Three different soils were each mixed with water, bentonite or foam and sheared against a TBM steel plate. Test results show unconditioned friction coefficients matching with those reported from literature. Compared to the unconditioned friction coefficients, all bentonite-conditioned specimens show reduced friction coefficients, while foam-conditioned sand specimens do not show any reduction in friction. Data from literature and the laboratory tests resulted in two regressions, for sand and clay respectively, relating steel-soil friction coefficients to relative density, grain size, liquid limit and steel roughness.@en