Prediction of TBM performance in fresh through weathered granite using empirical and statistical approaches

This study aims to develop several equations for predicting penetration rate (PR) and advance rate (AR) of tunnel boring machine (TBM) in fresh, slightly weathered and moderately weathered zones in granite rock mass. To reach study objectives, 12,649 m of the Pahang-Selangor Raw Water Transfer (PSRWT) tunnel in Malaysia was studied in both laboratory and field. In order to demonstrate the need for developing new equations for prediction of TBM performance, two well-known empirical models namely QTBM and Rock Mass Excavatability (RME) were applied and evaluated. It was found that the obtained results from these two empirical models are not accurate enough while, more accurate models are needed to propose. To get better performance results, linear multiple regression (LMR) and non-linear multiple regression (NLMR) models were built and proposed to estimate TBM PR and TBM AR. These equations were proposed for each weathering zone including fresh, slightly weathered and moderately weathered. Statistical indices including coefficient of determination (R-2), root mean square error (RMSE), variance account for (VAF), rank value and total rank values were implemented and achieved to evaluate the accuracy of each model. It was found that both LMR and NLMR models are able to provide an acceptable accuracy level to estimate TBM performance with R-2 ranges from 0.5 to 0.7. However, the performance capacity of the NLMR equations was slightly better than the proposed LMR equations. The proposed equations in this study are considered as suitable, simple and practical models that can be used in field of TBM, however, they should be used when the same predictors with their ranges and conditions would be available.

Automated summary

This study successfully predicted the penetration rate and advance rate of tunnel boring machines in different weathering zones through the development of new equations, demonstrating the accuracy of these new models.