The purpose of this study is to use a novel approach to estimate the tunnel boring machine (TBM) penetration rate in diverse ground conditions. Methods. The methods used in this study include ant colony optimization (ACO), bee colony optimization (BCO) and the particle swarm optimization (PSO). Moreover, a comprehensive database was created based on machine performance using penetration rate (m/h) as an output parameteras well as intact rock and rock mass parameters including uniaxial compressive strength (UCS) (MPa), Brazilian tensile strength (BTS) (MPa), rock quality designation (RQD) (%), cohesion (MPa), elasticity modulus (GPa), Poisson's ratio, density(g/cm 3 ), joint angle (deg.) and joint spacing (m) as input parameters. Findings. Results showed that the analyses yielded several realistic and reliable models for predicting penetration rate of TBMs. ACO model has R 2 = 0.8830 and RMSE = 0.6955, BCO model has R 2 = 0.9367 and RMSE = 0.5113 and PSO model has R 2 = 0.9717 and RMSE = 0.3418. Originality. Prediction of TBM penetration rate using these methods has been carried out in the Sabzkooh water conveyance tunnel for the first time. Practical implications. According to the results, all three approaches are very effective but PSO yields more precise and realistic findings than other methods.
Tunnel boring machines (TBMs) are designed to excavate underground spaces and widely used in tunneling, civil and mining projects. TBM performance prediction substantially deals with the evaluation of machine's penetration rate and the number of consumed disc cutters. There are various methods and equations to predict the TBMs performance in the literature. In this paper, we predicted the penetration rate and number of consumed disc cutters in Beheshtabad water conveyance tunneling project, one of the major water conveyance tunneling projects in Iran, using Artificial Neural Network (ANN) and Support Vector Machine (SVM) methods. Results showed that both approaches are very effective but SVM yields more precise and realistic findings than ANN.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.