TBM performance prediction in rock tunneling

Introduction

A key factor in the successful application of a Tunnel Boring Machine (TBM) in tunneling is the ability to develop accurate penetration rate estimates for determining project schedule and costs. While use of hard rock TBM has become standard method of excavation in tunnels of over 1.5-2 km in length, the estimation of machine performance in many of the challenging ground types has not reached sufficient degree of accuracy. Rate of penetration (ROP), defined as the distance the machine advances in a given time while boring in rock, is a complex parameter that not only depends upon intact and rock mass properties (strength, fractures, and texture of rock) but also machine specifications including thrust and torque requirement. During the past three decades, numerous TBM performance prediction models for evaluation of TBM have been proposed. In brief, all the TBM performance prediction models can be divided into two distinguished approaches, namely theoretical and empirical ones. The theoretical models which are primarily developed by using indentation tests or full-scale laboratory cutting tests provide an estimate of cutting forces based on cutter and cutting geometry and spacing and penetration of the cut The disadvantage of these tests is that it does not completely represent the real rock mass conditions as the TBM disc cutters encounter in the field. On the other hand, an empirical method does have some strength (taking into account rock mass conditions) as well as some shortcomings. The main deficiency of the empirical models is the absence of cutting force, cutter geometry, cutting geometry and ability to match machine thrust and torque/power in various ground conditions.

Growth of TBM manufacturing technology and existence of some shortcomings in the prediction models have made it necessary to perform more research on the development of the new models. In this regards, collecting the data from the previous and current tunneling projects is under progress.

The goal of investigation is, developing a new equation for estimation of TBM performance in different geology conditions. Moreover, due to the complexity of TBM performance prediction, in addition to develop a new empirical equation, different artificial intelligence (AI) such as Artificial Neural Network, Fuzzy logic, ANFIS Support Vector Machine for analysis of data will be conducted and developed. Finally, in terms of AI and Regression analysis, set of graphs and new classification will be developed for estimation of Tunnel Boring Machine performance in hard rock condition. Also, most common rock mass classification including RMR, Q, GSI system will be utilized then the relationship between these classification systems with performance of Tunnel Boring Machine is investigated.

 

Current status

Base on the collected data, preliminary results have been presented and additional work is underway to expand the database to include other rock types and operational conditions, for improving the available models for general applications.

 

Preliminary output

In terms of analyzing the available data and the results of Principle Component Analysis (PCA) to find the most effective parameters on TBM performance (Fig. 1), it can be concluded that, among the rock strength, the uniaxial compressive strength is the most effective parameters and between rock mass characteristics the joint spacing has major effect on TBM penetration rate. In addition to rock material and rock mass characteristics, TBM parameters including thrust and power are main parameters used for TBM performance estimation. The machine specifications and in particular operational parameters including thrust and power represent the amount of forces and torque delivered to rock via cutterhead and disc cutters to initiate fracture propagation in rock.

Furthermore, different AI models have been developed containing artificial neural networks (ANN), adaptive neuro-fuzzy inference system (ANFIS) and support vector regression (SVR) for prediction of TBM performance. Based on the results, the SVR model has better accuracy for estimation of TBM performance.

 

Figure 1. Principal components analysis for some features.

 

Future work

Rock mass classification systems are often applied in many empirical design practices in rock engineering, some contrasting with the original intent and applications of these classification systems, for example, estimation of TBM performance in various ground conditions. While accurate estimation of machine performance significantly impacts the costs and schedule of TBM tunneling projects, parameters used in these classifications are more related to support design and not rock mass boreablity. The results of many investigations on this issue has shown a weak correlation between TBM rate of penetration and rock mass classification. This limitation can be overcome by fine tuning the rock mass classifications input parameters to represent influence of rock mass properties on TBM performance. In this regards, developing new equations for TBM performance prediction using rock mass classification will be investigated.

↑ Seitenanfang

contact: Alireza Salimi