• Geomechanics and Engineering
• /
• v.39 no.1
• /
• pp.27-41
• /
• 2024

Accurately estimating the performance of tunnel boring machines (TBMs) is crucial for mitigating the substantial financial risks and complexities associated with tunnel construction. Machine learning (ML) techniques have emerged as powerful tools for predicting non-linear time series data. In this research, six advanced meta-heuristic optimization algorithms based on long short-term memory (LSTM) networks were developed to predict TBM penetration rate (TBM-PR). The study utilized 1125 datasets, partitioned into 20% for testing, 70% for training, and 10% for validation, incorporating six key input parameters influencing TBM-PR. The performances of these LSTM-based models were rigorously compared using a suite of statistical evaluation metrics. The results underscored the profound impact of optimization algorithms on prediction accuracy. Among the models tested, the LSTM optimized by the particle swarm optimization (PSO) algorithm emerged as the most robust predictor of TBM-PR. Sensitivity analysis further revealed that the orientation of discontinuities, specifically the alpha angle (α), exerted the greatest influence on the model's predictions. This research is significant in that it addresses critical concerns of TBM manufacturers and operators, offering a reliable predictive tool adaptable to varying geological conditions.

• Tunnel and Underground Space
• /
• v.13 no.6
• /
• pp.413-420
• /
• 2003

Modelling for performance prediction of mechanical excavation is discussed in this paper. Two of the most successful performance prediction models, namely theoretical based CSM model and empirical based NTH model, are discussed and compared. The basic principles of rock cutting with disc cutters, especially Constant Cross Section cutters, are discussed and a theoretical model developed is introduced to provide an estimate of disc cutting forces as a function of rock properties and the cutting geometry. General modelling logic for the performance prediction of mechanical excavation is introduced. CSM computer model developed and currently used at the Earth Mechanics Institute(EMI) of the Colorado School of Mines is discussed. Example of input and output of this model is illustrated for the typical operation by Tunnel Boring Machine(TBM).

• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.6
• /
• pp.35-42
• /
• 2019

In this paper, we analyze the practical problems of the regenerative start method proposed to improve the starting characteristics of the TBM(Tunnel boring machine)system and propose a solution. In order to solve the speed ripple problem in the previous research results, we first analyze the problems occurring in the system and propose a method to compensate for them. By applying the improved method to the actual system, we compared the results with the conventional system and verified the effect of the proposed method.

• Journal of the Korean GEO-environmental Society
• /
• v.11 no.2
• /
• pp.43-52
• /
• 2010

Shield method is the tunnel boring method that propels a steel cylinder in the ground and excavates tunnels at once. After Marc Isambard Brunel started using the method for the Thames Riverbed Tunnel excavation in London, many kinds of TBM (Tunnel Boring Machine) developed and applied for the construction of road, railway, electricity channel, pipeline, etc. In comparison with NATM concept that allows to observe ground condition and copes with difficulty. The machine selected before starting construction is not able to be changed during construction in shield TBM. Therefore the machine should be designed based on the ground survey result and experiment, so that the tunnel might be excavated effectively by controlling penetration speed, excavation depth and cutter head speed according to the ground condition change. This research was conducted to estimate penetration depth, excavate speed, wear of disc cutter on Boondang Railway of the Han Riverbed Tunnel ground condition by TBM performance prediction models such as NTNU, $Q_{TBM}$, Total Hardness, KICT-SNU and compare the estimated value with the field data. The estimation method is also used to analyze the reason of poor excavation efficiency at south bound tunnel.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.5
• /
• pp.641-656
• /
• 2019

In this study, a machine learning method was proposed to use in predicting optimal replacement period of shield TBM (Tunnel Boring Machine) disc cutter. To do this, a large dataset of ground condition, disc cutter replacement records and TBM excavation-related data, collected from a shield TBM tunnel site in Korea, was built and they were used to construct a disc cutter replacement period prediction model using a machine learning algorithm, SVM (Support Vector Machine) and to assess the performance of the model. The results showed that the performance of RBF (Radial Basis Function) SVM is the best among a total of three SVM classification functions (80% accuracy and 10% error rate on average). When compared between ground types, the more disc cutter replacement data existed, the better prediction results were obtained. From this results, it is expected that machine learning methods become very popularly used in practice in near future as more data is accumulated and the machine learning models continue to be fine-tuned.

• Computers and Concrete
• /
• v.22 no.2
• /
• pp.239-248
• /
• 2018

TBM penetration rate is a function of intact rock properties, rock mass conditions and TBM operational parameters. Machine rate of penetrationcan be predicted by knowledge of the ground conditions and its effects on machine performance. The variation of TBM operational parameters such as penetration rate and thrust plays an important role in its performance. This study presents the results of the analysis on the TBM penetration rates in schistose rock types present along the alignment of Golab tunnel based on the analysis of a TBM performance database established for every stroke through different schistose rock types. The results of the analysis are compared to the results of some empirical and theoretical predictive models such as NTH and QTBM. Additional analysis was performed to find the optimum thrust and revolution per minute values for different schistose rock types.

• Geomechanics and Engineering
• /
• v.30 no.1
• /
• pp.75-91
• /
• 2022

This paper aims to show how to use several Machine Learning (ML) methods to estimate the TBM penetration rate systematically (TBM-PR). To this end, 1125 datasets including uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), punch slope index (PSI), distance between the planes of weakness (DPW), orientation of discontinuities (alpha angle-α), rock fracture class (RFC), and actual/measured TBM-PRs were established. To evaluate the ML methods' ability to perform, the 5-fold cross-validation was taken into consideration. Eventually, comparing the ML outcomes and the TBM monitoring data indicated that the ML methods have a very good potential ability in the prediction of TBM-PR. However, the long short-term memory model with a correlation coefficient of 0.9932 and a route mean square error of 2.68E-6 outperformed the remaining six ML algorithms. The backward selection method showed that PSI and RFC were more and less significant parameters on the TBM-PR compared to the others.

• Tunnel and Underground Space
• /
• v.32 no.6
• /
• pp.345-352
• /
• 2022

The existing NATM (New Austrian Tunneling Method) has induced civil compliants due to blasting vibration and noise. Machanized excavation methods such as TBM (Tunnel Boring Machine) are being adopted in the planning and construction of tunneling projects. Shield TBM method is composed of repetition processes of TBM excavation and segment installation, the machine has to be stopped during the later process. Consecutive excavation technology using helical segment is under developing to minimize the stoppage time. The modification of thrust jacks and module are planned to ensure the advance force acting on the inclined surface of helical segment. Also, the integrated system design of hydraulic circuit will be remodeled. This means that the system deactivate the jacks on the installing segment while the others automatically act the thrusting forces on the existing segments. This report briefly introduces the mechanical research part of the current consecutive excavation technological development project of TBM.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.6D
• /
• pp.623-629
• /
• 2010

Recently, the Shield TBM (Tunnel Boring Machine) construction method is used gradually to increase at the Tunnel Constructin site. However the design and application of the Shield TBM were carried out without sufficient investigation of the ground conditions in the construction site. Due to insufficient understanding to the corresponding equipment is frequently occurring unexpected construction cost and extension of a construction period. The most suitable alternative construction method was determined by analyzing tunneling rate, duration, construction cost of shield machine and tunneling data of alternative method. The result of the case study is suggested as follows. First, the accurate soil exploration on the construction site should be preceded to prevent from tunneling stoppage and schedule delay. Second, the most suitable selection of the shield machine to the ground conditions of the construction site should be executed based on the investigation. Third, the best alternative method for boring of hard rock section is 'hard rock blasting after open cut and cover method'.

• Tunnel and Underground Space
• /
• v.13 no.3
• /
• pp.196-206
• /
• 2003

In order to investigate the workability of Raise Boring Machine(RBM) such as utilization, penetration rate and advance rate, a vertical shaft of 98 m in length and 3.05 m in diameter was constructed in the layer of conglomerate by using the RBM in this study. In addition, field data from tow different construction sites including water-pump power plant tunnel, roadway tunnel and mining tunnel by RBM were collected and analyzed. The results show that the average weekly bored length is 19.3 m and its average utilization is between 54.3 % and 75.1 % very higher than that of the TBM(Tunnel Boring Machine). It also turns out that the bit force increases linearly with respect to the increase of the RPM(revolution per minute) of RBM. However, the net penetration rate decreases with the increase of bit force, RPM of RBM and depth of shaft. The findings of this study can be used to provide the useful information for the design of shaft and the selection of RBM.