THE USE OF STATISTICAL APPROACHES IN APPROXIMATING TUNNEL BORING MACHINE PERFORMANCE PARAMETERS

One of the challenging issues in deep excavation area is predicting the performance of tunnel boring machine (TBM) in difficult rock mass conditions. A suitable estimation level of penetration rate (PR) and advance rate (AR) of TBM may reduce the risks related to high capital costs and scheduling for tunneling. This paper presents both linear and non-linear multiple regression models/equations to predict PR and AR of TBM. To obtain this aim, the Pahang-Selangor Raw Water Transfer (PSRWT) tunnel in Malaysia was investigated and the data collected along the tunnel and generated in the laboratory via rock tests to be used for the proposed models. A database comprising of 560 datasets in different tunnel distance were prepared in which uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), rock quality designation (RQD), rock mass rating (RMR), thrust force (TF) and revolution per minute (RPM) were set as model inputs to approximate PR and AR. Then, several linear and non-linear equations were constructed and the best models were selected according to a simple ranking method. Based on the obtained results, although both linear and non-linear models are suitable for estimating PR and AR of TBM, non-linear model shows higher performance capacity in predicting PR and AR compared to the linear one.

Tunnel boring machine, Penetration rate, Advance rate, Linear and non-linear multiple regression.