A fuzzy logic model to predict the performance of hard rock tunnel boring machine

Prediction of tunnel boring machine (TBM) is one of the most crucial and decisive issues in underground excavation projects. Precise estimation of machine performance can significantly mitigate the capital costs of mechanical excavation project. The main objective of this study is to estimate the TBM penetration rate by constructing a fuzzy inference system analysis. For this purpose, rule-based (Mamdani model) fuzzy logic were employed to build a fuzzy model and 34 TBM field datasets including Q rock mass classification system, rock material properties and machine characteristics along the route of the tunnel were compiled. Hence, the F-Q (fabric index of Q rock mass classification system), F-f (the ratio of uniaxial compressive strength and load per cutter) and F-alpha were determined as input parameters. In order to verify the validity of the two models, the predicted penetration rate and the measured penetration rate gained from the field records were compared. Results picked out form this predictor model revealed that this model has a strong capability for estimation of TBM performance with a correlation coefficient of 81.5%.