Rock mass quality prediction on tunnel faces with incomplete multi-source dataset via tree-augmented naive Bayesian network

Rock mass quality serves as a vital index for predicting the stability and safety status of rock tunnel faces.In tunneling practice,the rock mass quality is often assessed via a combination of qualitative and quantitative parameters.However,due to the harsh on-site construction conditions,it is rather difficult to obtain some of the evaluation parameters which are essential for the rock mass quality prediction.In this study,a novel improved Swin Transformer is proposed to detect,segment,and quantify rock mass characteristic parameters such as water leakage,fractures,weak interlayers.The site experiment results demonstrate that the improved Swin Transformer achieves optimal segmentation results and achieving accuracies of 92%,81%,and 86%for water leakage,fractures,and weak interlayers,respectively.A multisource rock tunnel face characteristic(RTFC)dataset includes 11 parameters for predicting rock mass quality is established.Considering the limitations in predictive performance of incomplete evaluation parameters exist in this dataset,a novel tree-augmented naive Bayesian network(BN)is proposed to address the challenge of the incomplete dataset and achieved a prediction accuracy of 88%.In comparison with other commonly used Machine Learning models the proposed BN-based approach proved an improved performance on predicting the rock mass quality with the incomplete dataset.By utilizing the established BN,a further sensitivity analysis is conducted to quantitatively evaluate the importance of the various parameters,results indicate that the rock strength and fractures parameter exert the most significant influence on rock mass quality.

Rock mass quality classification based on deep learning:A feasibility study for stacked autoencoders

Objective and accurate evaluation of rock mass quality classification is the prerequisite for reliable sta-bility assessment.To develop a tool that can deliver quick and accurate evaluation of rock mass quality,a deep learning approach is developed,which uses stacked autoencoders(SAEs)with several autoencoders and a softmax net layer.Ten rock parameters of rock mass rating(RMR)system are calibrated in this model.The model is trained using 75%of the total database for training sample data.The SAEs trained model achieves a nearly 100%prediction accuracy.For comparison,other different models are also trained with the same dataset,using artificial neural network(ANN)and radial basis function(RBF).The results show that the SAEs classify all test samples correctly while the rating accuracies of ANN and RBF are 97.5%and 98.7%,repectively,which are calculated from the confusion matrix.Moreover,this model is further employed to predict the slope risk level of an abandoned quarry.The proposed approach using SAEs,or deep learning in general,is more objective and more accurate and requires less human inter-vention.The findings presented here shall shed light for engineers/researchers interested in analyzing rock mass classification criteria or performing field investigation.

Determination and applications of rock quality designation(RQD)

Characterization of rock masses and evaluation of their mechanical properties are important and challenging tasks in rock mechanics and rock engineering. Since in many cases rock quality designation （RQD） is the only rock mass classification index available, this paper outlines the key aspects on determination of RQD and evaluates the empirical methods based on RQD for determining the deformation modulus and unconfined compressive strength of rock masses. First, various methods for determining RQD are presented and the effects of different factors on determination of RQD are highlighted. Then, the empirical methods based on RQD for determining the deformation modulus and unconfined compressive strength of rock masses are briefly reviewed. Finally, the empirical methods based on RQD are used to determine the deformation modulus and unconfined compressive strength of rock masses at five different sites including 13 cases, and the results are compared with those obtained by other empirical methods based on rock mass classification indices such as rock mass rating （RMR）, Q-system （Q） and geological strength index （GSI）. It is shown that the empirical methods based on RQD tend to give deformation modulus values close to the lower bound （conservative） and unconfined compressive strength values in the middle of the corresponding values from different empirical methods based on RMR, Q and GSI. The empirical methods based on RQD provide a convenient way for estimating the mechanical properties of rock masses but, whenever possible, they should be used together with other empirical methods based on RMR, Qand GSI.

Rapid profiling rock mass quality underneath tunnel face for Sichuan-Xizang Railway

The Sichuan-Xizang Railway is a global challenge,surpassing other known railway projects in terms of geological and topographical complexity.This paper presents an approach for rapidly profiling rock mass quality underneath tunnel face for the ongoing construction of the Sichuan-Xizang Railway.It adopts the time-series method and carries out the quantitative analysis of the rock mass quality using the depth-series measurement-while-drilling(MWD)data associated with drilling of blastholes.A tunnel face with 15 blastholes is examined for illustration.The results include identification of the boundary of homogeneous geomaterial by plotting the blasthole depth against the net drilling time,as well as quantification of rock mass quality through the recalculation of the new specific energy.The new specific energy profile is compared and highly consistent with laboratory test,manual logging and tunnel seismic prediction results.This consistency can enhance the blasthole pattern design and facilitate the dynamic determination of charge placement and amount.This paper highlights the importance of digital monitoring during blasthole drilling for rapidly profiling rock mass quality underneath and ahead of tunnel face.It upgrades the MWD technique for rapid profiling rock mass quality in drilling and blasting tunnels.

Classification and rating of disintegrated dolomite strata for slope stability analysis

Although disintegrated dolomite,widely distributed across the globe,has conventionally been a focus of research in underground engineering,the issue of slope stability issues in disintegrated dolomite strata is gaining increasing prominence.This is primarily due to their unique properties,including low strength and loose structure.Current methods for evaluating slope stability,such as basic quality(BQ)and slope stability probability classification(SSPC),do not adequately account for the poor integrity and structural fragmentation characteristic of disintegrated dolomite.To address this challenge,an analysis of the applicability of the limit equilibrium method(LEM),BQ,and SSPC methods was conducted on eight disintegrated dolomite slopes located in Baoshan,Southwest China.However,conflicting results were obtained.Therefore,this paper introduces a novel method,SMRDDS,to provide rapid and accurate assessment of disintegrated dolomite slope stability.This method incorporates parameters such as disintegrated grade,joint state,groundwater conditions,and excavation methods.The findings reveal that six slopes exhibit stability,while two are considered partially unstable.Notably,the proposed method demonstrates a closer match with the actual conditions and is more time-efficient compared with the BQ and SSPC methods.However,due to the limited research on disintegrated dolomite slopes,the results of the SMRDDS method tend to be conservative as a safety precaution.In conclusion,the SMRDDS method can quickly evaluate the current situation of disintegrated dolomite slopes in the field.This contributes significantly to disaster risk reduction for disintegrated dolomite slopes.

Multiscale hierarchical analysis of rock mass and prediction of its mechanical and hydraulic properties

Engineering geological and hydro-geological characteristics of foundation rock and surrounding rock mass are the main factors that affect the stability of underground engineering. This paper presents the concept of multiscale hierarchical digital rock mass models to describe the rock mass, including its structures in different scales and corresponding scale dependence. Four scales including regional scale,engineering scale, laboratory scale and microscale are determined, and the corresponding scaledependent geological structures and their characterization methods are provided. Image analysis and processing method, geostatistics and Monte Carlo simulation technique are used to establish the multiscale hierarchical digital rock mass models, in which the main micro-and macro-structures of rock mass in different geological units and scales are reflected and connected. A computer code is developed for numerically analyzing the strength, fracture behavior and hydraulic conductivity of rock mass using the multiscale hierarchical digital models. Using the models and methods provided in this paper, the geological information of rock mass in different geological units and scales can be considered sufficiently,and the influence of downscale characteristics(such as meso-scale) on the upscale characteristics(such as engineering scale) can be calculated by considering the discrete geological structures in the downscale model as equivalent continuous media in the upscale model. Thus the mechanical and hydraulic properties of rock mass may be evaluated rationally and precisely. The multiscale hierarchical digital rock mass models and the corresponding methods proposed in this paper provide a unified and simple solution for determining the mechanical and hydraulic properties of rock mass in different scales.

Geotechnical investigations and remediation design for failure of tunnel portal section： a case study in northern Turkey

Mass movements are very common problems in the eastern Black Sea region of Turkey due to its climate conditions, geological, and geomorphological characteristics. High slope angle, weathering, dense rainfalls, and anthropogenic impacts are generally reported as the most important triggering factors in the region. Following the portal slope excavations in the entrance section of Cankurtaran tunnel, located in the region, where the highly weathered andesitic tuff crops out, a circular toe failure occurred. The main target of the present study is to investigate the causes and occurrence mechanism of this failure and to determine the feasible remedial measures against it using finite element method(FEM) in four stages. These stages are slope stability analyses for pre-and postexcavation cases, and remediation design assessments for slope and tunnel. The results of the FEM-SSR analyses indicated that the insufficient initial support design and weathering of the andesitic tuffs are the main factors that caused the portal failure. After installing a rock retaining wall with jet grout columns and reinforced slope benching applications, the factor of safety increased from 0.83 to 2.80. In addition toslope stability evaluation, the Rock Mass Rating(RMR), Rock Mass Quality(Q) and New Austrian Tunneling Method(NATM) systems were also utilized as empirical methods to characterize the tunnel ground and to determine the tunnel support design. The performance of the suggested empirical support design, induced stress distributions and deformations were analyzed by means of numerical modelling. Finally, it was concluded that the recommended stabilization technique was essential for the dynamic long-term stability and prevents the effects of failure. Additionally, the FEM method gives useful and reasonably reliable results in evaluating the stability of cut slopes and tunnels excavated both in continuous and discontinuous rock masses.

Cross-project prediction for rock mass using shuffled TBM big dataset and knowledge-based machine learning methods

Extensive research has confirmed the successful prediction of rock mass quality in tunnel boring machine(TBM)construction using machine learning methods based on big data collected during the boring process.However,the developed model cannot be applied to a new project owing to the different mechanical and environmental features involved in different projects.This study tries to combine the datasets of three TBM projects whose cutterhead diameters are 5.2,7.9 and 9.8 m,respectively.In this study,machine learning focused on predictions of a binary rock mass quality system was implemented using this unified dataset by adding the diameter and disc cutter number as new attributes into the input.The process consists of:(1)individual learning for the three respective datasets,(2)shuffled learning for the unified dataset containing randomly distributed information from the three projects,and(3)crossed learning aimed at validating that the algorithm developed on the unified dataset can produce predictions with equally acceptable accuracies as those obtained in the individual learning.It is anticipated that with more datasets joining this cross-project learning,we will be able to develop a machine learning algorithm that is suitable for new projects with a wide range of cutterhead diameters and disc cutter numbers at the beginning of the tunnel excavation.

Feedback on a shared big dataset for intelligent TBM Part Ⅱ:Application and forward look

This review discusses the application scenarios of the machine learning-supported performance prediction and the optimization effi-ciency of tunnel boring machines(TBMs).The rock mass quality ratings,which are based on the Chinese code for geological survey,were used to provide"labels"suitable for supervised learning.As a result,the generation of machine prediction for rock mass grades reason-ably agreed with the ground truth documented in geological maps.In contrast,the main operational parameters,i.e.,thrust and torque,can be reasonably predicted based on historical data.Consequently,18 collapse sections of the Yinsong project have been successfully predicted by several researchers.Preliminary studies on the selection of the optimal penetration rate and cost were conducted.This review also presents a summary of the main achievements in response to the initiatives of the Lotus Pool Contest in China.For the first time,large and well-documented TBM performance data has been shared for joint scientific research.Moreover,the review discusses the technical problems that require further study and the perspectives in the future development of intelligent TBM construction based on big data and machine learning.