ArticlePublisher preview available

Influence of Subjectivity in Geological Mapping on the Net Penetration Rate Prediction for a Hard Rock TBM

DOI:10.1007/s00603-018-1408-2
Authors:
Yongbeom Seo at Norwegian University of Science and Technology
Yongbeom Seo
Javier Macias at JMConsulting-Rock Engineering AS
Javier Macias
  • JMConsulting-Rock Engineering AS
Pål Drevland Jakobsen at Norwegian Geotechnical Institute
Pål Drevland Jakobsen
Amund Bruland at Norwegian University of Science and Technology
Amund Bruland
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

The net penetration rate of hard rock tunnel boring machines (TBM) is influenced by rock mass degree of fracturing. This influence is taken into account in the NTNU prediction model by the rock mass fracturing factor (ks). ks is evaluated by geological mapping, the measurement of the orientation of fractures and the spacing of fractures and fracture type. Geological mapping is a subjective procedure. Mapping results can therefore contain considerable uncertainty. The mapping data of a tunnel mapped by three researchers were compared, and the influence of the variation in geological mapping was estimated to assess the influence of subjectivity in geological mapping. This study compares predicted net penetration rates and actual net penetration rates for TBM tunneling (from field data) and suggests mapping methods that can reduce the error related to subjectivity. The main findings of this paper are as follows: (1) variation of mapping data between individuals; (2) effect of observed variation on uncertainty in predicted net penetration rates; (3) influence of mapping methods on the difference between predicted and actual net penetration rate.
Figures - available from: Rock Mechanics and Rock Engineering
This content is subject to copyright. Terms and conditions apply.
  • A preview of this full-text is provided by Springer Nature.
  • Learn more
Preview content only
Content available from Rock Mechanics and Rock Engineering
This content is subject to copyright. Terms and conditions apply.
Vol.:(0123456789)
1 3
Rock Mechanics and Rock Engineering (2018) 51:1599–1613
https://doi.org/10.1007/s00603-018-1408-2
ORIGINAL PAPER
Inuence ofSubjectivity inGeological Mapping ontheNet
Penetration Rate Prediction foraHard Rock TBM
YongbeomSeo1· FranciscoJavierMacias1,2· PålDrevlandJakobsen1,2· AmundBruland1
Received: 9 November 2015 / Accepted: 12 January 2018 / Published online: 20 January 2018
© Springer-Verlag GmbH Austria, part of Springer Nature 2018
Abstract
The net penetration rate of hard rock tunnel boring machines (TBM) is influenced by rock mass degree of fracturing. This
influence is taken into account in the NTNU prediction model by the rock mass fracturing factor (ks). ks is evaluated by geo-
logical mapping, the measurement of the orientation of fractures and the spacing of fractures and fracture type. Geological
mapping is a subjective procedure. Mapping results can therefore contain considerable uncertainty. The mapping data of a
tunnel mapped by three researchers were compared, and the influence of the variation in geological mapping was estimated
to assess the influence of subjectivity in geological mapping. This study compares predicted net penetration rates and actual
net penetration rates for TBM tunneling (from field data) and suggests mapping methods that can reduce the error related
to subjectivity. The main findings of this paper are as follows: (1) variation of mapping data between individuals; (2) effect
of observed variation on uncertainty in predicted net penetration rates; (3) influence of mapping methods on the difference
between predicted and actual net penetration rate.
Keywords Hard rock TBM· Geological mapping· NTNU prediction model· Net penetration rate· Rock mass fracturing
factor (ks)
Abbreviations
TBM Tunnel boring machine
NTNU Norwegian University of Science and
Technology
DRI™ Drilling Rate Index™
CSM Colorado School of Mines
UCS Uniaxial compressive strength
RMR Rock mass rating
RQD Rock quality designation
RMi Rock Mass Index
GSI Geological Strength Index
CLI™ Cutter Life Index™
MSJ Marked single joints
CV Coefficient of variation
SD Standard deviation
List of symbols
ks Rock mass fracturing factor
α Orientation of the weakness plane
αf Dip angle of the weakness plane
αt Bearing of the tunnel axis
αs Strike angle of the weakness plane
ks-avg Average rock mass fracturing factor
li Tunnel length of fracturing class “i” in the
mapped section
ks-i Rock mass fracturing factor for class “i” in the
mapped section
ks-tot Total rock mass fracturing factor
ksi Rock mass fracturing factor for set number “i”
n Number of fracturing sets
Im Average net penetration rate
Inj Net penetration rate for subsection j
Ij Length of subsection j
Tbj Net time (machine hours) used to bore subsection
j
̄
X
Mean of the results
* Yongbeom Seo
yongbeom.seo@ntnu.no
1 Department ofCivil andTransport Engineering, NTNU,
7491Trondheim, Norway
2 SINTEF Building andInfrastructure, Rock Engineering,
7465Trondheim, Norway
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
... Ates et al. [6] recommended empirical models for predicting torque, thrust, and other design parameters. With the continuous improvement of test equipment, multifactor prediction models represented by the Norwegian University of Science and Technology (NTNU) model [16,17] have been widely used. In addition to the above empirical methods, many scholars have also made some research progress in theoretical analysis methods. ...
Prediction Method of TBM Tunneling Parameters Based on Bi-GRU-ATT Model
Article
Full-text available
  • Feb 2022
With tunnel boring machines (TBMs) widely used in tunnel construction, the adaptable adjustment of TBM operating status has become a research focus. Since the prediction of tunnel geological conditions is still challenging before excavating, the prediction of important TBM operating parameters plays an important role in the research on TBM adaptable adjustment. This paper proposes an intelligent prediction method of TBM tunneling parameters based on bidirectional gate recurrent unit incorporating attention mechanism (Bi-GRU-ATT) and selects a complete tunneling cycle to predict the tunneling parameters of the TBM complete tunneling cycle. Relying on the TBM3 bid section of Jilin Water Supply Project, 21 key parameters of the complete tunneling cycle are selected as the input features of the model to realize the prediction of four tunneling parameters in the complete driving cycle section of TBM. Compared with the Bi-GRU, GRU, and Long Short-Term Memory (LSTM) models, it can be seen that the Bi-GRU-ATT model has a goodness of fit for predicting TBM tunneling parameters above 0.92, and the average absolute percentage error is less than 1.8%. The results show that the prediction method of TBM tunneling parameters based on Bi-GRU-ATT model proposed in this paper has stronger learning and prediction capabilities. This prediction method provides a more feasible auxiliary intelligent decision-making method for TBM aided intelligent construction.
View
... Predicting TBM performance is important to safe and efficient tunneling. In the preceding decade, numerous methods for predicting TBM performance have been proposed (Salimi et al., 2018(Salimi et al., , 2016Koopialipoor et al., 2019Koopialipoor et al., , 2018Fatemi et al., 2018;Armaghani et al., 2018;Seo et al., 2018;Jamshidi, 2018;Vergara and Saroglou, 2017;Minh et al., 2017;Jakubowski et al., 2017;Fattahi and Babanouri, 2017;Adoko et al., 2017). Generally, these methods can be divided into two categories, namely, theoretical and empirical (Hamidi et al., 2010;Salimi and Rostami, 2019). ...
TBM performance prediction with Bayesian optimization and automated machine learning
Article
Accurately predicting the performance of a tunnel boring machine (TBM) is important to safe and efficient tunneling. The application of machine learning algorithms to TBM performance prediction creates several challenges. Such prediction is a nontrivial task involving procedures such as data preprocessing, selection of a machine learning algorithm and optimization of the related hyperparameters. The demand for expert knowledge has restricted the application of machine learning methods to TBM performance prediction, and it is meaningful to study predicting TBM performance automatically. In this paper, we explore three approaches to TBM performance prediction using Bayesian optimization and automated machine learning (AutoML). In the first study, Bayesian optimization is used to determine the optimal hyperparameters of various machine learning algorithms, including support vector regression (SVR), decision tree, bagging tree, random forest and AdaBoost. We attain the minimum mean squared error (MSE) values of 3.135x10^-2 and 3.177x10^-2 for a decision tree and SVR, respectively. In the second approach called the neural architecture search (NAS), the optimal combination of architecture, hyperparameters and the training procedure of an artificial neural network is found in a single operation. We obtain the optimal results of 3.514x10^-2 and 3.237x10^-2 if complete and simplified NAS are used, respectively. In the third method, the best combination of a data preprocessing method, a machine learning model and the related hyperparameters is found, and an optimal MSE value of 3.148x10^-2 is obtained using AutoML. In all three studies, we obtain state-of-the-art prediction results that are superior to a previous best prediction result of 3.500x10^-2. The prediction results prove that Bayesian optimization and AutoML are powerful tools that can not only effectively predict TBM performance but also reduce the demand for expert knowledge of machine learning.
View
... Yagiz (2017) proposed new equations for predicting the field penetration index of tunnel boring machines in fractured rock mass. Seo et al. (2018) studied the influence of geological mapping on the net penetration rate prediction. Mikaeil et al. (2018) developed a fuzzy multi criteria decisionmaking approach for hard rock TBMs. ...
Prediction of the Penetration Rate and Number of Consumed Disc Cutters of Tunnel Boring Machines (TBMs) Using Artificial Neural Network (ANN) and Support Vector Machine (SVM)—Case Study: Beheshtabad Water Conveyance Tunnel in Iran
Article
Full-text available
  • Jan 2019
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.
View
Total loads modeling and geological adaptability analysis for mixed soil-rock tunnel boring machines
Article
  • Oct 2021
Rock-soil interface mixed ground (RSI) is often encountered in tunnel construction. The excavation loads of tunnel boring machines (TBMs) are controlled by the interaction characteristics between TBM and rock/soil layers. The different properties of rock and soil cause the varying interaction range and stress distribution. Currently, there have been several studies available to estimate excavation loads under RSI, and the conclusion is that the total loads increase with increasing the rock layer proportion in the excavation face. However, the previous studies cannot take the difference of rock / soil properties into account, except for the calculation of cutters loads. Therefore, the interaction characteristics between RSI and TBM is unclear. This paper analyzes the interaction characteristics between TBM’s main components and complex geological conditions (e.g., layered soil, layered rock, and RSI condition). A model is proposed to calculate the total thrust and total torque assuming quasi-static equilibrium of the tunneling equipment. The rationality and applicability of the model are discussed and verified by two typical projects. Furthermore, the geological adaptability is discussed in terms of the excavation difficulty and the matching relationship between total torque and total thrust. The results indicate that when the rock layer proportion in the excavation face increases, the reduction of overall extrusion and friction loads is 1.5 times higher than the increase of disc cutters breaking load. The total loads and the ratio of the total torque to total thrust decrease approximately linearly. There is a power function relationship between the excavation difficulty index and the penetration depth. The results of this study provide an important reference for the total loads design of equipment propulsion systems and the parameter adjustment during tunnel construction.
View
DEVELOPMENT OF A FORCE ESTIMATION MODEL FOR ROCK FRAGMENTATION WITH DISC CUTTERS THROUGH THEORETICAL MODELING AND PHYSICAL MEASUREMENT OF CRUSHED ZONE PRESSURE
Thesis
Full-text available
  • May 1997
Accurate and reliable estimation of cutting forces acting on a disc cutter is a prerequisite in developing models for design optimization and performance prediction of Tunnel Boring Machines (TBM). The ability to accurately predict TBM performance means more reliable estimates of project completion time and costs, hence improved confidence in the application of this technology to a wider sector of the underground construction market. The disc cutting forces are generated and controlled by the load distribution within the contact area between the rock and the cutter. To determine the magnitude and location of load distribution, a series of laboratory cutting tests were designed and performed to measure the strain and stresses within the cutter ring. The measurements were obtained by installing a set of strain gages on each side of the disc ring very close to the tip. The experimental program started by first installing the ring on the hub and measuring the strains caused by the hoop stresses to estimate the pressure between the ring and the hub. A special data reduction program was developed to analyze the gage outputs to determine the load distribution in the contact area. In addition, extensive theoretical and Finite Element Modeling of the disc ring were performed for comparison with experimental observations. The ability of the gages and the data reduction program to accurately measure the loads was verified by running the disc on a steel plate. The actual test program included cutting three rock types, Indiana Limestone, Colorado Red Granite, and Umettela basalt, to develop cutting force and strain measurement data under conditions of soft to hard, brittle rocks. The results of the test program concluded that the commonly accepted notion of cutter ring being loaded over the full length of the contact area is not correct. The analysis of data obtained showed that the loading area is a portion (30-80%) of the contact area, close the center of the arc of contact. The width of the loading area was found to increase with the depth of cutter penetration in softer rocks while in the harder rocks, the extent of loading area was relatively insensitive to changes in penetration. Further, test observations and analysis of the loading area led to the conclusion that the rock failure primarily occurs in front of the cutter. The force estimation equations were developed by regression analysis of an extensive data base of cutting forces using disc cutters in different rock types. These equations were then incorporated into a model designed to predict and optimize TBM performance as a function of rock properties and machine capabilities. The model allows input of any cutter geometry and layout with the capability to produce overall project completion time and schedules. To check the accuracy of the developed model, field performance data from four different tunneling projects were analyzed and compared with model outputs. Overall, a very good correlation was obtained between the model predictions and the field performance of the TBMs.
View
TBM tunnelling in jointed and faulted rock
Article
Full-text available
  • Jan 2000
About this book This book describes the stage by stage development of a new method for predicting the penetration rate (PR) and the advance rate (AR) for tunnel boring machines based on an expanded version of the Q-value, termed QTBM. Some 145 TBM tunnels totalling about 1000 km in length were an-alysed and some simple formulae are developed from the results to esti-mate PR and AR from the QTBM value, or to back-calculate QTBM from performance data. Logging methods, empirical TBM tunnel support de-sign, and numerical verification of support are also treated in this slim but practical book on TBM tunnelling. Penetration rates as high as 10 m/hr, but occasionally as low as 0.1 m/hr, are a function both of the machine and rock mass interaction, and of the cutter force and rock strength interaction. Actual advance rates that could be as high as 5 m/hr for one day, or as low as 0.005 m/hr (while stuck for several months in a major fault zone) are quantitatively ex-plained. This book is a useful source of reference for consultants, contractors and owners of TBM tunnels, and for those involved with feasibility stud-ies, machine and support design and follow-up of tunnel progress. Among the geotechnical community, the book will be useful for geolo-gists, engineering geologists and rock mechanics engineers, and for all civil engineers who have a professional interest in TBM tunnelling.
View
New model for performance production of hard rock TBMs
Article
Full-text available
  • Jan 1993
A new theoretical/empirical model has been developed for performance prediction of hard rock TBMs. The model uses information on the rock properties and cutting geometry to calculate TBM rate of penetration. The model can also be used for TBM cutterhead design optimization to achieve maximum performance in a given geology.
View
TBM performance estimation in rock using Q(TBM)
Article
Full-text available
  • Jan 1999
Nick Barton, Technical Adviser, NGI, Norway, Visiting Professor at the University of Sao Paulo, Brazil, has developed a new method for predicting penetration rate (PR) and advance rate (AR) for TBM tunnelling. This method is based on an expanded Q-system of rock mass classification and average cutter force in relation to the appropriate rock mass strength. Orientation of fabric or joint structure is accounted for, together with the compressive or point load (tensile) strength of the rock. The abrasive or non-abrasive nature of the rock is incorporated via the University of Trondheim cutter life index (CLI). Rock stress level is also considered. The new parameter Q(TBM) can be estimated during feasibility studies, and can also be back calculated from TBM performance during tunnelling.
View
View
Hard Rock Tunnel Boring: Performance Predictions and Cutter Life Assessments
Thesis
  • Dec 2016
The use of hard rock tunnel boring machines (TBMs) has become widely and generally used with success but in too many cases, due to unanticipated situations and/or inappropriate assessments, with catastrophic consequences. Hard rock tunnel boring involves major investments and high levels of geological risk, which require reliable performance predictions and cutter life assessments. Hard rock tunnel boring leads the interaction between the rock mass and the machine, which is a process of great complexity. The tunnelling system around the excavation process has a great relevance in the final goal of performance predictions for hard rock TBMs, which is the estimation of time and cost. The overall aim of this thesis is to build on the existing knowledge of tunnel boring and wear processes, technology and capacity, thus enhancing performance prediction and cutter life assessments in hard rock tunnel boring projects. The prediction model for hard rock TBMs developed by the Norwegian University of Science and Technology (NTNU) is based on a combination of field performance data, engineering geological back-mapping, and laboratory testing. The development of TBM technology during recent decades, and the possible influence of parameters not considered in previous versions, has made revision of the model to improve prediction accuracy an essential requirement. TBM specifications, the penetration rate and cutter life models have been revised and extended to adapt to the current technology. In addition, it has been incorporated a new definition of the rock mass fracturing, the influence of the cutterhead velocity (rpm) on penetration rate on the in response to the results of in-situ trials 'RPM tests' and cutter thrust on cutter life. The tunnel length effect on time consumption for the tunnelling activities has been introduced for the estimation of the machine utilization and therefore in the advance rate model. A revised and extended version of the current version of the NTNU prediction model for TBM performance and cutter life has been published included in this thesis. Cutter consumption and parameters such as cutter ring wear play a significant role in performance and cost predictions, especially in hard rock conditions. Cutter wear involves a complex tribological system in interaction with the geological properties of the rock mass. Existing laboratory test methods fail to reproduce wear behaviour encountered during tunnel boring. Because of this, and the importance of cutter wear, it was considered of interest to develop a new rock abrasivity test method for tool life assessments in hard rock tunnel boring: The Rolling Indentation Abrasion Test (RIAT). Understanding the processes and failure mechanisms during cutter wear enabled new knowledge to be applied as part of the development of the new rock abrasivity test method for laboratory use. In addition, cutter wear mechanisms affecting cutter rings during tunnel boring might lead to better cutter consumption predictions and future improvements in cutter ring development.
View
View
View
Engineering classification of rock masses
Article
  • Dec 1973
A classification based on properties of rock materials and rock masses, termed the Geomechanics Classification, is proposed. The geological parameters are specified and the functional features of the Geomechanics Classification is demonstrated by applying it to the selection of primary support in tunnel design. The proposed classification system is based on a detailed study of all existing major rock classifications.
View
Rock Mass influence on Hard Rock TBM Performance Prediction
Conference Paper
  • May 2014
Prediction of construction time and cost is important to make a reliable choice of excavation method; to avoid schedule delays, budget overruns and to control the risk management for tunnelling projects. For hard rock Tunnel Boring Machine (TBM) projects, penetration rate is a key factor for estimating advance rates, cutter life and excavation costs. The penetration rate depends on intact rock properties, rock mass properties, and machine parameters. Rock mass fracturing factor (ks) is used to include the rock mass influence on the rock boreability for performance and cost prediction for tunnel boring. The ks considers the simultaneous influence of the degree of fracturing and the orientation of the plane of weakness. In the present paper, different correlations of the ks have been confirmed by analyses of data from a recently finished tunnel project.
View