Estimation of uniaxial compressive strength at tunnel face using TBM operation data

One of the most crucial factors influencing the excavation speed of TBM tunnels is rock strength. Laboratory tests in geotechnical investigations can determine rock strength, but determining the UCS for the entire TBM excavation section is impossible. To apply the appropriate operation parameters during TBM excavation, determining the rock strength that affects the excavation speed is essential. Therefore, the objective of this study is to estimate rock strength using machine data obtained during TBM excavation. The slurry shield TBM excavation of the rock strata provided the TBM machine data and the UCS required for the analysis. The data were split in a ratio of 7:3 for training and testing, pre-processed with scaling, and outlier removal. According to the findings, the Adaboost model is inferred to be the most accurate at predicting UCS from TBM excavation data, with root-mean-square error and determination coefficient values of 5.14 and 0.96, respectively.