Artificial intelligence in underground development: a study of TBM performance

Abstract

Modelling tunnel boring machine (TBM) performance is an important aspect in tunnel operations. The use of artificial intelligence techniques such as artificial neural networks has been recently introduced to this subject and the results from such applications prove their potential in making accurate prognosis. This paper presents a review of feed-forward artificial neural network (ANN) development and furthermore it illustrates their application by the use of two cases studies from Italian and Greek underground projects, where the TBM performance is modelled. The results obtained show that the developed ANNs can efficiently generalise the TBM behaviour in their respective geotechnical environment, having a reliable, effective and consistent performance.

Estimation of TBM Penetration rate using Gradient Boosting-based Algorithms

“…The response to the demand has been increasing worldwide since computer-based and programming-based techniques, like artificial intelligence (AI), came to the attention of researchers and industries; that started in the early 21 st century when Grima Alvarez employed artificial neural networks (ANN) and adaptive neuro-fuzzy inference system (ANFIS) to predict TBM performance (Grima et al, 2000). These techniques have drawn various attention where other researchers also used to estimate TBM performance by employing different input parameters including rock type, uniaxial compressive rock strength (UCS), Brazilian tensile strength (BTS), cutter life index (CLI), Young's modulus, BI, DPW, α, RQD, percentage of Quartz, RMR, TBM thrust and torque, joints spacing and conditions (Js, Jc), punch slope index (PSI), cohesion, internal friction angle, Poisson's ratio, density, RPM, cutter torque (CT), thrust force (TF), and AR (Afradi et al, 2019;Benardos, 2008;Eftekhari et al, 2010;Gholami et al, 2012;Gholamnejad & Tayarani, 2010;Oraee et al, 2012;Salimi & Esmaeili, 2013;Torabi et al, 2013;Yagiz et al, 2009;Zhu et al, 2021); the prediction performance of such models is varied from 0.69 to 0.939 from the R-squared point of view. Or using other derived algorithms based on neural networks (NN) like deep NN (DNN) (Koopialipoor et al, 2019), probabilistic NN (PNN) (Harandizadeh et al, 2021), back-propagation NN (BPNN) (Yan et al, 2023), convolutional NN (CNN) (L. .…”