Predicting TBM penetration rate in hard rock condition: A comparative study among six XGB-based metaheuristic techniques

A reliable and accurate prediction of the tunnel boring machine(TBM) performance can assist in minimizing the relevant risks of high capital costs and in scheduling tunneling projects.This research aims to develop six hybrid models of extreme gradient boosting(XGB) which are optimized by gray wolf optimization(GWO), particle swarm optimization(PSO), social spider optimization(SSO), sine cosine algorithm(SCA), multi verse optimization(MVO) and moth flame optimization(MFO), for estimation of the TBM penetration rate(PR).To do this, a comprehensive database with 1286 data samples was established where seven parameters including the rock quality designation, the rock mass rating, Brazilian tensile strength(BTS), rock mass weathering, the uniaxial compressive strength(UCS), revolution per minute and trust force per cutter(TFC), were set as inputs and TBM PR was selected as model output.Together with the mentioned six hybrid models, four single models i.e., artificial neural network, random forest regression, XGB and support vector regression were also built to estimate TBM PR for comparison purposes.These models were designed conducting several parametric studies on their most important parameters and then, their performance capacities were assessed through the use of root mean square error, coefficient of determination, mean absolute percentage error, and a10-index.Results of this study confirmed that the best predictive model of PR goes to the PSO-XGB technique with system error of(0.1453, and 0.1325), R~2 of(0.951, and 0.951), mean absolute percentage error(4.0689, and 3.8115), and a10-index of(0.9348, and 0.9496) in training and testing phases, respectively.The developed hybrid PSO-XGB can be introduced as an accurate, powerful and applicable technique in the field of TBM performance prediction.By conducting sensitivity analysis, it was found that UCS, BTS and TFC have the deepest impacts on the TBM PR.     

Estimation of the performance of the tunnel boring machine (TBM) using uniaxial compressive strength and rock mass rating classification (RMR)–A case study from the Deccan traps,India

The competency of any TBM in any geological condition is determined by a rock or rock mass breakage process. A 12.24 km long tunnel between Maroshi and Ruparel College was excavated by Brihanmumbai municipal corporation (BMC) to improve water supply system of greater Mumbai, India, using open-type hard rock tunnel boring machines (TBMs). In this paper an attempt has been made to establish the relationship between rock mass characteristics i.e. RMR and UCS of the Deccan trap rocks and TBMs performance characteristics for 5.83 km long Maroshi–Vakola tunnel section of the Maroshi–Ruparel college tunnel project. To analyze the effect of variable rock mass conditions on the TBM performance, the operating parameters i.e. thrust force, torque and RPM of the machine, were recorded and intact rock strength was determined. The effect of rock mass properties on machine penetration rate (PR) and the relation with other operational parameters were analyzed. The rock strength affects the rock behaviour under compression. When the rolling cutters indent the rock, the stress exerted must be higher than the rock strength i.e.; the rock strength is directly relevant to the performance of TBM. Studies show that the penetration rate decreases with increase in uniaxial compressive strength (UCS). The comparison of measured penetration rate with empirical model developed by Graham, in which, the penetration rate is computed using UCS and average thrust per cutter, showed good agreement with coefficient of determination (R²), i.e. 0.97. The study shows that the TBM performance was maximum in rock mass rating (RMR) range from 40 to 75, while slower penetration was recorded both in very poor and very good rock masses.     

TBM Performance Analysis in Pyroclastic Rocks: A Case History of Karaj Water Conveyance Tunnel

Karaj Water Conveyance Tunnel (KWCT) is 30-km long and has been designed for transferring 16 m³/s of water from Amir-Kabir dam to northwest of Tehran. Lot No. 1 of this long tunnel, with a length of 16 km, is under construction with a double shield TBM and currently about 8.7 km of the tunnel has been excavated/lined. This paper will offer an overview of the project, concentrating on the TBM operation and will review the results of field performance of the machine. In addition to analysis of the available data including geological and geotechnical information and machine operational parameters, actual penetration and advance rates will be compared to the estimated machine performance using prediction models, such as CSM, NTNU and Q_TBM. Also, results of analysis to correlate TBM performance parameters to rock mass characteristics will be discussed. This involves statistical analysis of the available data to develop new empirical methods. The preliminary results of this study revealed that the available prediction models need some corrections or modifications to produce a more accurate prediction in geological conditions of this particular project.     

Spatial prediction of landslide susceptibility in western Serbia using hybrid support vector regression (SVR) with GWO,BAT and COA algorithms

In this study, we developed multiple hybrid machine-learning models to address parameter optimization limitations and enhance the spatial prediction of landslide susceptibility models. We created a geographic information system database, and our analysis results were used to prepare a landslide inventory map containing 359 landslide events identified from Google Earth, aerial photographs, and other validated sources. A support vector regression (SVR) machine-learning model was used to divide the landslide inventory into training (70%) and testing (30%) datasets. The landslide susceptibility map was produced using 14 causative factors. We applied the established gray wolf optimization (GWO) algorithm, bat algorithm (BA), and cuckoo optimization algorithm (COA) to fine-tune the parameters of the SVR model to improve its predictive accuracy. The resultant hybrid models, SVR-GWO, SVR-BA, and SVR-COA, were validated in terms of the area under curve (AUC) and root mean square error (RMSE). The AUC values for the SVR-GWO (0.733), SVR-BA (0.724), and SVR-COA (0.738) models indicate their good prediction rates for landslide susceptibility modeling. SVR-COA had the greatest accuracy, with an RMSE of 0.21687, and SVR-BA had the least accuracy, with an RMSE of 0.23046. The three optimized hybrid models outperformed the SVR model (AUC = 0.704, RMSE = 0.26689), confirming the ability of metaheuristic algorithms to improve model performance.     

Prediction model of tunnel boring machine performance by ensemble neural networks

The penetration rate of a tunnel boring machine (TBM) depends on many factors ranging from the machine design to the geological properties. Therefore it may not be possible to capture this complex relationship in an explicit mathematical expression. In this paper, we propose an ensemble neural network (ENN) to predict TBM performance. Based on site data, a four-parameter ENN model for the prediction of the specific rock mass boreability index is constructed. Such a neural-network-based model has the advantages of taking into account the uncertainties embedded in the site data and making appropriate inferences using very limited data via the re-sampling technique. The ENN-based prediction model is compared with a non-linear regression model derived from the same four parameters. The ENN model outperforms the non-linear regression model.     

Intelligent assistant driving method for tunnel boring machine based on big data

Acta Geotechnica - During tunnel construction with tunnel boring machine (TBM), the TBM drivers determine the driving parameters depending only on their own experiences. Inappropriate TBM driving...     

Developing new equations for TBM performance prediction in carbonate-argillaceous rocks: a case history of Nowsood water conveyance tunnel

Nowsood water conveyance tunnel is 49 km long and has been designed for transferring 70 m³/s water from Sirvan river southward to Dashte Zahab plain in the west of Iran. This long tunnel has been divided into three sections, namely 1A, 1B and 2. By April 2008, about 5.3 km of the lot 2 of this project, with a total length of 26 km, were excavated by a double-shield TBM. The bored section of tunnel passed through different geological units of three main formations of the Zagross mountain ranges which mainly consist of weak to moderately strong argillaceous-carbonate rocks. This paper will offer an overview of the project, concentrating on the TBM operation, and review the results of the field performance of the machine. Also results of statistical analyses to evaluate correlation of TBM performance parameters with rock mass characteristics will be discussed. The results of machine performance analysis indicated that there are strong relationships between geomechanical parameters and TBM performance parameters in this particular project. In this research some empirical equations and a chart have been developed to estimate TBM performance parameters in similar cases based on common rock mass properties.     

Full-scale linear cutting test in Chongqing Sandstone and the comparison with field TBM excavation performance

In this study, determination of some machine parameters and performance prediction for tunnel boring machine (TBM) are conducted based on laboratory rock cutting test. Firstly, laboratory full-scale linear cutting test is carried out using 432-mm CCS (constant cross section) disc cutter in Chongqing Sandstone. Then, the input parameters for TBM cutterhead design are extracted; some TBM specifications are determined and then compared to the manufactured values. Finally, laboratory full-scale linear cutting test results are compared with the field TBM excavation performance data collected in Chongqing Yangtze River Tunnel. Results show that laboratory full-scale linear cutting test results, combined with some engineering considerations, can be used for the preliminary and rough design of TBM machine capacity. Meanwhile, combined with some modification factors, it can also well predict the field TBM excavation performance.

     

New equations for predicting the field penetration index of tunnel boring machines in fractured rock mass

The Queens Water Tunnel No. 3, stage 2 having 7.5-km length and 7-m diameter, is excavated by a high-power tunnel boring machine (TBM) underneath Brooklyn and Queens area for distributing freshwater throughout the New York City, USA. This paper offers a review of the project by considering the TBM performance and rock mass interaction. Using the individual cutter force, intact, and mass rock properties, TBM performance by means of field penetration index (FPI) was predicted and compared with actual results obtained in the field. Further, the study involves statistical analysis of the laboratory and field data including machine, intact, and mass rock properties to develop new empirical equations to estimate FPI. It is stated that the FPI, also converted to the rate of penetration, could be estimated utilizing intact and mass rock properties together with cutter force for similar type of rocks with correlation coefficient of 0.88.     

Study of the influence of geotechnical parameters on the TBM performance in Tehran–Shomal highway project using ANN and SPSS

Alborz twin tunnel along with an exploratory or service tunnel between the two main tunnels, are the longest tunnels section in Tehran–Shomal highway with 6.3 km length. The service tunnel is designed to be used for geological investigations, ventilation, transportation during the construction of main tunnels, water drainage, ground improvement by grouting, and emergency exit. An open tunnel boring machine (TBM) of Wirth Company was used to drive this service tunnel. With regard to the fact that in such mechanized tunneling projects, performance of the TBMs is of the most importance, which affects the economy and timing of the projects; on the other hand, geotechnical conditions of the region play a significant role in this respect, this effect was investigated during this study. In this study, two main elements of the TBM performance including the rate of penetration and utilization factor were investigated using artificial neural network and Statistical Package for Social Sciences. It is shown that geotechnical conditions have considerable effect on the rate of penetration. Whereas, utilization is largely affected by management and non-rock mass-related parameters including delays, wasted times, maintenance, labor, etc. With regard to the available data, four parameters including uniaxial compressive strength (UCS), friction angle, Poisson’s ratio, and cohesion were selected to be studied. Based on assessments conducted using these approaches, the rate of effectiveness of four selected parameters on penetration rate, in a descending order, was as follows: UCS, friction angle, Poisson’s ratio, and cohesion. For increasing utilization, it was concluded that minimizing time delays by good management is the most effective way. Furthermore, with regard to the relative error percentages and the coefficient of correlation of the input and output data, it was concluded that the method artificial neural network yields more reliable results than the statistical approach.     

Experimental and analytical studies of the parameters influencing the action of TBM disc tools in tunnelling

The use of tunnel boring machines (TBMs) is increasingly popular in tunnelling. One of the most important aspects in the use of these machines is to assess with certain accuracy the effectiveness of the action of the discs on the cutter-head in the different rock types to be excavated. A specific machine, called an intermediate linear cutting machine (ILCM), has been developed at the Politecnico di Torino in order to study, on a reduced scale in detail in the laboratory, the interaction between the discs of the TBM and the rock: this machine allows a series of grooves to be cut on a rock sample of 0.5 × 0.3 × 0.2 m, through the rolling of a 6.5-in. disc, and evaluation, during testing, of the parameters associated with the action of the cutting tool. The parameters measured during the tests were compared with the results obtained employing two analytical methods widely used for predicting the performance of TBMs: the Colorado School of Mines (CSM) model and the Norwegian University of Science and Technology (NTNU) model. The latter showed a greater ability to reproduce tests conducted using the ILCM. However, as with the CSM model, it does not allow the optimal excavation condition (the ratio, which minimizes the specific energy of excavation, between the groove spacing and the penetration of the disc), necessary for the correct design of the TBM cutter-head, to be identified. An example, based on a real case of a tunnel in Northern Italy, allowed a demonstration of how the NTNU model provides results in line with the measurements taken during the excavation and represents, therefore, a model that is able to reliably simulate both laboratory tests and the action of a TBM on site. The NTNU model, together with the results of the tests with ILCM targeted on the identification of the optimal conditions of excavation, may allow the correct dimensioning of the TBM cutter-head to be attained in order to effectively implement the excavation.     

TBM岩体可掘性预测及其分级研究

全断面硬岩隧道掘进机（tunnel boring machine, TBM）对岩体条件极其敏感,且其前期投入较大,准确地评估岩体可掘性、预测TBM掘进性能对TBM隧道施工至关重要。基于来自中国、伊朗两国涵盖3种不同岩性的5条TBM施工引水隧洞约300组现场数据,以现场贯入度指数FPI为岩体可掘性评价指标,分析了岩石单轴抗压强度UCS、岩体完整性指数 、岩体主要结构面与洞轴线的夹角?、隧洞直径D等与岩体可掘性之间的关系;探讨了适用于岩体可掘性研究的岩体参数统一方法,进一步建立了精度较高的（相关系数为0.768）岩体可掘性经验预测方法。基于该预测方法,运用K中心聚类分析方法,将岩体可掘性分为6类,探讨了不同岩体可掘性条件下TBM平均单刀推力、刀盘转速分布规律,相应成果可为实际工程中TBM施工隧洞岩体可掘性评估、掘进参数的选择、施工进度的安排提供一定的指导。     

A Predictive Model for Estimating the TBM Penetration Rate Based on Hybrid ICA-ANN and DEA-AHP Algorithms

Geotechnical and Geological Engineering - Predicting the penetration rate of tunnel boring machine (TBM) is a complex and challenging task that plays a crucial role in the schedule planning and...     

Parameter Study on Prediction Methods for TBM Penetration Rate

Penetration rate prediction of Tunnel Boring Machine (TBM) is the first step to advance prediction process of mechanized tunnelling. In this research, influence of effective parameters on TBM penetration rate is investigated by sensitivity analysis of three main TBM performance prediction methods; Norwegian University of Science and Technology (NTNU), rock mass index (RMi) and Q_TBM. Based on these analyses, it is shown that applied thrust per disc and joint spacing in NTNU and RMi models have more influence on penetration rate. In Q_TBM model, Q value, applied thrust per disc and induced biaxial stress are more effective.     

基于颗粒流模型的TBM滚刀破岩过程数值模拟研究

为了研究全断面岩石掘进机（TBM）盘型滚刀的破岩机制及其影响因素,采用颗粒流方法建立了岩石与滚刀的二维数值模型,实现了对TBM滚刀破岩过程的模拟。分析表明,滚刀的破岩过程可分为冲击挤压破碎、大量微裂纹生成、张拉性主裂纹扩展3个阶段,证实了滚刀破岩的挤压-张拉破坏理论。在滚刀侵入深度相同的前提下,随着刀圈刃角以及刃宽的增加,滚刀下的压碎区也相应增大,张拉性主裂纹数目增多,滚刀的破岩能力提高;与平刃刀圈相比,楔刃刀圈的“楔块劈裂”作用更加显著,使径向裂纹扩展得更快且更深入岩石内部。TBM滚刀对强度较高或较低岩石的破坏损伤较小,而对中等强度的岩石破坏损伤最为显著。     

Mechanical response of surrounding rock of tunnels constructed with the TBM and drill-blasting method

There are two kinds of excavation methods in underground engineering: the tunnel boring machine (TBM) and the drill-blasting method. A large number of studies have shown that the deformation and failure, the degree of disturbance, the stability and the reinforcement measures of surrounding rock using the TBM and drill-blasting method vary from each other. To accurately master these macroscopic damages, it is necessary to focus on the investigation of the micro-mechanical responses of the surrounding rock. Scanning electron microscopy tests, acoustic emission tests and tunnel acoustic detection tests were carried out to analyze the mechanical response of surrounding rock of tunnels, which were excavated in marble by, respectively, the TBM and the drill-blasting method. The tests results showed that most of the rock fractures cut by TBM is wipe along the crystal, and the failure mechanism is mainly cutting, while most of the rock fractures induced by the TBM coincide with crystal planes, its mechanism is mainly tensile. The stress–strain curves of rocks cut by the TBM method are rather flat around the peak strength, which means a strong resistance to deformation around the peak load. The response of AE for the rock cut by the TBM method appears after larger strains than the response of the rock constructed by the drill-blasting method. This suggests that the resistance to damage is higher under TBM excavation conditions. The relaxation depths of the tunnel excavated by the drill-blasting method are larger than the tunnel excavated by the TBM method. The research can provide more insight into tunnel failure mechanisms and provide a framework for reinforcement measures.     

基于可拓理论的围岩稳定分类方法的研究

在双护盾TBM（tunnel boring machine）的隧洞施工中,将可拓理论与洞室围岩稳定评价相结合。基于碴料和掘进参数的地质编录所提供的地质信息,选取了能够反映围岩稳定综合特性的评价指标,确定围岩稳定类型和预测前方岩体情 况。在物元理论、可拓集合论和关联函数运算的基础上,建立了隧洞围岩稳定分类的可拓评价方法,其中引进了隶属度的概念和一种定量的指标权重的确定方法,并在山西引黄工程的双护盾TBM隧洞施工中用此分类方法对某两段围岩进行了稳定分 类,得到的稳定分类结果与实际情况吻合。     

软土地层TBM开挖面支护压力计算模型及可视化

隧道掘进机（TBM）近年来在世界范围内得到了广泛应用,通常通过完全充满压力仓的泥土或泥浆来支护开挖面。但在较差的地层和水力条件下,开挖面失稳时有发生。事实上,TBM开挖面的支护压力的大小直接决定了施工安全及地表变形。基于所建立的开挖面支护压力计算模型,并考虑复合地层下土体分层带来的影响,通过计算机编程方法,建立了界面友好、使用便捷的开挖面支护压力可视化计算平台(TBM Studio);并结合阿拉斯加隧道、钱江隧道工程实例进行了不同模型结果的验证分析,给出了各模型计算结果的差异性;讨论了软土复合地层条件下,土体自稳性对开挖面稳定的影响,认为软土地层中定量确定有效支护压力和水头高度至关重要,研究为正确评价TBM开挖面稳定性提供了相应的计算模型。     

基于引力搜索法的隧道围岩微震定位研究

针对目前深埋隧道围岩微震源定位难且精度不高等问题,采用启发式算法——引力搜索法（GSA）对隧道围岩微震源位置进行搜索,并将该算法与粒子群算法和单纯形法的搜索结果进行对比。发现在双速度模型和三速度模型下,引力搜索法相较于粒子群算法和单纯形法,都具有快速收敛、精度较高的优点,且与震源位置的距离能够控制在10 m以内。对双速度模型,引力搜索法的精度相对于单纯形法提高了83.71%,相对于粒子群算法提高了7.77%。对三速度模型,引力搜索法的精度相对于单纯形法提高了70.67%,相对于粒子群算法提高了39.36%。可见,该方法为深埋隧道微围岩震源定位提供了一种新思路。