Strength Stress Ratio Research Articles

Stress-strain state zoning model and novel large deformation classification method for squeezing tunnels

Accurately assessing the level of large deformation in squeezing tunnels is of great importance for the design of support structures. The current large deformations classification methods fail to account for the effect of support pressure on the relative deformation of surrounding rock. In this study, a theoretical model for the stress–strain state zoning of the surrounding rock in squeezing tunnels is established. A sensitivity analysis of the model parameters is conducted. The distribution law of the strength-stress ratio, relative deformation, and relative support pressure indices in 43 field monitoring squeezing tunnels are statistically analyzed. A novel large-deformation classification method based on the relative support pressure index is proposed and applied in case studies. The parameter sensitivity analysis shows that mechanical parameters of the surrounding rock significantly affect the deformation and the stress–strain state zoning of the surrounding rock. The stress–strain state zoning and deformation of the surrounding rock are also primarily dependent on the matching relationship between strength-stress ratio and support pressure. Statistical analyses indicate that there is a nonlinear negative correlation between the relative deformation of the surrounding rock and the strength-stress ratio, as well as the relative support pressure, which includes the steel arch relative support pressure index (SSPI) and the anchor bolt relative support pressure Index (ASPI). The key to controlling surrounding rock deformation lies in the rational matching of support pressure to the strength-stress ratio. In the novel large deformation classification method, three key indexes of strong stress ratio, relative deformation and relative support pressure are quantitatively included, and squeezing large deformation is divided into three levels. Moreover, the case study demonstrated that the novel large deformation classification method has significant guiding importance in verifying feasibility and optimizing support design schemes during both the design and construction phases.

A strength-stress coupling criterion for rockburst: Inspirations from 1114 rockburst cases in 197 underground rock projects

Rockburst is a kind of engineering geological disaster induced by artificial engineering excavation. Qualitative and quantitative statistical analysis of rockburst cases is the primary work and research basis for rockburst prediction and discrimination. In this study, 1114 rockburst cases are collected and the qualitative and quantitative characteristics are analyzed. It is found that rockburst is prone to occur in rocks that are hard, complete, dry, and unweathered. Strong rockburst occurs more frequently in rocks with high uniaxial compressive strength (σc), elastic modulus, and elastic energy index, and in rock masses with high initial maximum principal stress (σ1)and tangential stress. Notably, rockburst may occur only when σc≥ 60 MPa and σ1≥ 10 MPa. On this basis, the “Human–Rock–Environment” three-element mechanism of rockburst is presented. Guided by this conception, the rationality and defects of the traditional strength-stress ratio (SSR) criterion are investigated. The SSR criterion does not adequately consider the absolute difference between the rock strength and geostress, as well as their combined effects. Subsequently, a strength-stress coupling (SSC) criterion is proposed from a two-dimensional scale.This criterion is characterized by considering both the absolute boundary conditions of σcand σ1, as well as their coupling interval. Five intervals of weak, weak–moderate, moderate, moderate–strong, and strong rockburst are divided in the two-dimensional chart and their distinguishing conditions are given. The existence of two transition zones for weak–moderate and moderate–strong reflects the coupling effect of rock strength and geostress when rockburst occurs, confirming the rationality and scientificity of the SSC criterion. The verification results show that the SSC criterion features higher accuracy as compared with the SSR criterion.

Failure mechanism and deformation prediction of soft rock tunnels based on a combined finite–discrete element numerical method

The large deformation mechanisms and predictions of soft rock tunnels have always been important but difficult to solve problems in the field of geotechnical engineering. A combined finite–discrete element numerical simulation method (FDEM) was used to study large deformation mechanism, classification and prediction. The deformations or failure mechanisms of tunnel surrounding rock were revealed, and the failure modes and displacement value prediction of surrounding rock with different strength-stress ratios were also investigated. The following conclusions were obtained: (1) Critical hysteresis damping can be adopted to simulate the progressive large deformation process of soft rock tunnels, which can obtain the final deformation of unsupported tunnels and avoid a dynamic response. (2) Under concentrated tangential stress, the surrounding rock undergoes X-shaped conjugate shear fracture, and this type of fracture network continues to propagate toward the depth of the surrounding rock until the model reaches a stable state; the reduction in tunnel cross-section is mainly caused by the macroscopic movement and volume expansion of rock fragments, and the latter is due to the generation of a large number of macroscopic voids. (3) As the strength-stress ratio decreases, the deformation or failure modes of the surrounding rock can be divided into four categories: elastic–plastic deformation, closed fracturing, shear dilation and broken expansion. (4) Finally, a prediction equation for isotropic and homogeneous soft rock unsupported tunnel deformation with general size driven by hydrostatic in situ stress is obtained, which indicates that with an increasing strength-stress ratio, the surrounding rock displacement decreases as an exponential function, with a correlation coefficient of R2 = 0.997. In addition, the robustness and reliability of the prediction equation is verified.

Robust model for tunnel squeezing using Bayesian optimized classifiers with partially missing database

Accurately predicting and estimating the squeezing and ground response to tunneling remains challenging. Moreover, tunnel-squeezing hazards are much more likely to occur in deeply buried long tunnels with complex engineering-geological environments. Therefore, a high-performance predictive model for tunnel squeezing is necessary. A superior ensemble classifier is put forward in this study, which is composed of four individual classifiers (gradient boosting classifier, extra-trees classifier, AdaBoost classifier, and Logistic regression classifier) and two optimization algorithms (Bayesian optimization (BO) and sparrow search algorithm (SSA)). The training database covers five parameters: tunnel depth (H), rock tunneling quality index (Q), tunnel diameter (D), support stiffness (K), and strength stress ratio (SSR), about which the basic information is accessible at the early design phases. However, the dataset compiled from the literature is insufficient. Thus, the ten proposed methods are used to replace the missing values. During the model training process, BO shows its strong ability to optimize seventeen hyperparameters. When applied to tune the classifiers' weights, SSA achieves a fast and efficient performance. The novel Shapley Additive Explanations–LightGBM method indicates that the K is the most important input feature, followed by SSR, Q, H, and D, respectively. The ensemble classifier is then validated using the test set and additional historical case projects. The validation shows that the model can achieve an accuracy of 98% (i.e., the error rate is 2%) on the test set, higher than those achieved by previous prediction models. Moreover, the predicted probability could provide warning information for timely support measures. Finally, the application results are illustrated through tests on the tunnel sections that have not yet been excavated in the line of the Sichuan–Tibet railway project. The applied predictive tendencies and laws are in line with the practical experience. In summary, the proposed model's prediction results are reasonable, and its prediction will be more accurate as more data is collected and trained for prewarning the tunnel squeezing hazard.

Numerical study on the induced seismicity through hydraulic fracturing in the deep heat mining project in Basel, Switzerland using a pseudo dynamic model and considering fully hydro-mechanical coupling

The induced seismicity associated with fluid injection (hydraulic fracturing) or forced circulation has become a key issue in the safe exploitation of hot dry rock geothermal energy. In this paper, the induced seismicity in the Deep Heat Mining Project in Basel, Switzerland was numerically studied by using a developed 3D numerical code with considering the fully hydro-mechanical coupling and the dynamic stress transfer effects. The numerical study was based on history matching of the injection pressure and the statistic characteristics of the induced seismicity. Further, sensitivity analysis of some parameters relating to injection operation, slip behavior and strength-stress ratio was conducted. According to this study, the following conclusions were obtained: 1) the temporal and spatial distribution of the seismic events can be divided into an indirectly induced zone, a directly induced zone, and a quiet zone; 2) large seismic events can be found within the indirectly induced zone; 3) the released energy is not exactly proportional to the injection volume; 4) when the geological condition is well-known, the maximum induced seismicity by fluid injection can be approximately determined; 5) the slip behavior and the strength-stress state are dominate factors for inducing large seismic events.

Characteristics of rockburst and early warning of microseismic monitoring at qinling water tunnel

In-depth records of the geological and rockburst data during the construction of the Qinling Water Tunnel of the Han to the Wei River project, analysis of the initial stress field distribution rules of the rock mass within a buried depth of 700 m. According to the situation of the site, strength stress ratio criterion, Russense criterion and Hoek criterion, were analysed and modified, and the rockburst tendency criterion suitable for Qinling Water Tunnel was proposed. According to the rockburst data, the rockbursts in the TBM tunnelling section of Qinling Water Tunnel were classified and counted, and the characteristics and risk reasons of rockbursts were summarized. The microseismic monitoring technology was applied to carry out the monitoring and early warning of rockburst in Qinling Water Tunnel, so as to achieve the effect of real-time monitoring. By comparing the microseismic monitoring results with the statistical data of rockburst, it was found that the microseismic monitoring has high accuracy of rockburst prediction. Finally, combined with two typical cases of rockburst prediction, it can be found that microseismic monitoring can reflect the preparation process of rockburst and use artificial intelligence for rockburst early warning, which has great practical significance for the prevention of rockburst.

A Gray Wolf Optimization-Based Improved Probabilistic Neural Network Algorithm for Surrounding Rock Squeezing Classification in Tunnel Engineering

Surrounding rock squeezing deformation is a common and prominent hazard in tunnel engineering projects, which often induces the shield jamming disaster during the TBM tunneling process. Based on the 139 groups of historical squeezing deformation cases, this study developed a hybrid PCA-IWGO-PNN model for squeezing classification. According to the influencing factors and characteristics of squeezing deformation, the strength-stress ratio, tunnel burial depth, tunnel equivalent diameter, rock mass quality index, and support stiffness were selected to establish the prediction index system of squeezing level. Because the probabilistic neural network (PNN) requires that the input variables are independent, principal component analysis (PCA) was used to preprocess the original data to eliminate the correlation between prediction indexes and achieve dimensionality reduction. The spread coefficient was the critical hyper-parameter in the PNN, and the improved gray wolf optimization (IGWO) algorithm was used to realize its efficient automatic optimization. Then, the PNN model was applied to engineering practice. Only 1 of 20 test samples was misjudged, achieving the 95% prediction accuracy. Finally, the comparison analysis with the artificial neural network (ANN) model, support vector machine (SVM) model, and random forest (RF) model was conducted. Among them, the PNN model achieved the highest prediction accuracy, followed by the artificial neural network (85%), RF (85%), and SVM (80%). In addition, the PNN model had the fastest running speed, which only consumed 5.6350 s, while the running time of ANN, SVM, and RF was 8.8340, 6.2290, and 6.9260 s, respectively. The hybrid PCA-IWGO-PNN model developed in this research provides an effective method for surrounding rock squeezing classification, and it has superiorities in both prediction accuracy and running speed.

Numerical Simulation of the Stress Field in Repeated Mining of Coal Seams Based on In Situ Stress Measurement

We established an evaluation index of the rock mass stress state for underground coal mines using the strength-stress ratio based on the measured in situ stress and the generalized Hoek–Brown strength criterion. Three in situ stress states, σcm/σ1m < 1.4 (high), 1.4 < σcm/σ1m < 3.6 (medium), and σcm/σ1m > 3.6 (low), were established based on the value of the unconfined compressive strength (σcm) and the maximum principal stress of the rock mass (σ1m). This index classifies the Burtai mine as a medium-high in situ stress field, which is in agreement with the on-site situation, establishing the reliability of the index. The working face was a three-dimensional geological model based on the log sheets. The initial conditions for the model were determined using the combined measurements of the in situ stress regression model. We performed numerical simulations of the roof stress field distribution under repeated mining. Mining the overlying coal seam leads to significant variation in the value and direction of the main roof, σ1, within the range of the front abutment pressure under the pillar and gob. Along the main roof strike direction, σ1 under the pillar is 1.5 times that under the gob, and the σ1 direction under the pillar is deflected by 5°, which is 30° smaller than that under the gob. This provides a reference for optimized underground coal mining.

Optimization Analysis of Excavation Procedure Design of Underground Powerhouses under High In Situ Stress in China

To recommend the excavation procedures and design parameters for underground powerhouses, excavation procedures of fifty-one underground powerhouses in China were summarized and analyzed based on in situ stress conditions. Firstly, the complex stress environment in China was introduced and fifty-one underground powerhouses with their engineering scale, size, lithology, rock classification and in situ stress level were listed in detail. Subsequently, to evaluate the influence of in situ stress levels on excavation procedure design, the correlation between excavation procedures and in situ stress level in three main excavation zones were analyzed accordingly. Moreover, to provide the excavation design recommendations, the strength–stress ratio (SSR) was promoted to analyze and recommend the design parameters, and the blasting excavation design based on the stress transient unloading control was also supplemented. The results show that excavation procedures have different priorities under different in situ stress levels, and the design parameters show an obvious relationship with in situ stress levels. Moreover, the excavation procedure parameters are suggested to adjust accordingly under different SSR. The discussion of influencing factors and specification ensures its rationality and accuracy. It is believed that the summary and recommendations can provide a good reference for excavation procedure optimization of underground powerhouse under high in situ stress.

Study on Rock Mass Classification Methods Used in the Geological Disposal of High-level Radioactive Waste

The conventional methods of rock quality classification mainly focus on the constructability of rock mass, while the long-term safety requirements of the high-level radioactive waste (HLW) disposal are not considered. Therefore, two rock suitability classification methods for geological disposal projects, the Rock Suitability Criterion (RSC) and the QHLW methods were systematically compared by evaluating the rock suitability of the ONKALO site for geological disposal in Finland. The suitability of the ONKALO site was analyzed by evaluating the permeability, geochemical condition, and integrity of rock mass, according to the RSC method. Meanwhile, the QHLW method was used to evaluate the rock mass suitability by analyzing the fault zone influence, hydrogeochemical characteristics, temperature influence, permeability, strength-stress ratio, and integrity of rock mass. Then, the advantages and disadvantages of the two methods were discussed. We found that the results obtained with RSC are consistent with those with QHLW. While RSC gives more detailed chemical properties of groundwater and mechanical characteristics of rock mass, QHLW performs better in evaluating the permeability of rock mass. Besides, RSC is a qualitative or semi-qualitative method with some site-specific criteria, while the QHLW is a quantitative method with higher universality. Based on the above results and conclusions, a new version of QHLW at the tunnel scale was finally proposed by incorporating the advantages of RSC.

Evaluation of the large deformation grade cloud model of surrounding rock based on combination weighting method

On the basis of the analytic hierarchy process (AHP) and entropy weight method, combined with the statistical results of large deformation under the complex geological conditions of Galigongshan tunnel, the grade cloud model evaluation method of the large deformation of surrounding rock is proposed. Its influencing factors are considered to select the strength stress ratio, maximum principal stress, elastic modulus, and uniaxial compressive strength as evaluation indexes. The weight of each index in the evaluation of large deformation is determined through AHP and the entropy weight method. Meanwhile, the combined weights of each influencing factor index are obtained by the optimal combination weighting method of maximizing the sum of squares of total deviation. Aiming at the uncertainty and randomness of the large deformation evaluation of tunnel surrounding rock, the numerical characteristics of a normal cloud model are used to calculate the determination of the evaluation index, and the evaluation results of the large deformation grade of surrounding rock are obtained. In this study, this method is applied to the large deformation evaluation of the surrounding rock of Galigongshan tunnel engineering. Compared with the results of the fuzzy hierarchical comprehensive evaluation method, the model evaluation method established here is more feasible. It combines subjective and objective weights to obtain a reasonable weight in the practical engineering application of significance, as large deformation provides a reliable reference for the prediction and prevention of disasters.

Numerical Simulation Research on Rockburst Prediction of Gaoloushan Deep Tunnel Based on Strength-stress Ratio Method

In view of the complex geological environment of deep buried tunnel, the great harm of rock burst, and the serious hidden danger to the underground engineering construction and the safety of the workers, it is necessary to adopt an effective method to reasonably predict the rock burst, so as to carry out the rock burst disaster warning. Based on the strength-stress ratio method, the possibility of rockburst in the deep-buried tunnel of Pingliang-Mianyang National Expressway is predicted in this paper. At the same time, the FLAC3D numerical simulation software was used to build a numerical model to simulate the excavation process of the deep buried tunnel, so as to analyze the stress state of the surrounding rock of the deep buried tunnel, and the ratio of the maximum principal stress σ 1 of the surrounding rock obtained from the simulation calculation to the uniaxial compressive strength Rc of the rock in the excavation process. According to the ratio, the possibility level of rockburst at different length sections along the deep buried tunnel is divided, analyzed and predicted, and compared with the actual situation of rockburst in the field, the results show that the location of rockburst and the rockburst level of the corresponding length are basically the same. This discriminant method can provide reference and relevant reference for rockburst prediction of similar engineering.

Analysis of factors affecting the deformation of soft rock tunnels by data envelopment analysis and a risk assessment model

Deformation of the surrounding rock is an anticipated geological hazard when driving tunnels through soft rock. If deformation is not restricted, it can lead to serious engineering safety hazards. Therefore, studying the factors that affect the deformation of soft rock tunnels is highly significant. This study selected twelve main influencing factors involving subjective and objective factors and four actual indexes to establish a deformation risk assessment system for a soft rock tunnel. Data envelopment analysis (DEA) was innovatively utilized to analyze the efficiency of the twelve factors which affect the deformation of soft rock tunnels. The analytical results showed that the water content of the soft rock was the most efficient factor, followed by the tunnel depth, uniaxial compressive strength, excavation method, strength-stress ratio and groundwater seepage condition. A risk assessment model for the deformation of a soft rock tunnel was established using a normal cloud model and verified by the deformation occurred in typical soft rock tunnel cases. Risk assessment results confirmed that the proposed model was consistent with actual deformation risk degree. This model can guidie significance for the assessment of soft rock tunnel engineering.

Dynamic and Probabilistic Multi-class Prediction of Tunnel Squeezing Intensity

Tunnel squeezing is a time-dependent process that typically occurs in weak or over-stressed rock masses, significantly influencing the budget and time of tunnel construction. This paper presents a new framework to probabilistically predict the potential squeezing intensity and to dynamically update the prediction during construction based on the sequentially revealed ground information. An extensively well-documented database, which contains quantitative data from 154 squeezing sections with 95 unpublished inventories is established. A Decision Tree method is employed to train a probabilistic multi-classification model to predict the tunnel squeezing intensity. The trained classifier is then integrated with a Markovian geologic model, which features embedded Bayesian updating procedures, to achieve a dynamic prediction on the state probabilities of the geologic parameter within the model and the resulting squeezing intensity during excavation. An under-construction tunnel case—Miyaluo #3 tunnel—is used to illustrate the proposed framework. Results show that the Decision Tree classifier, as opposed to other black-box models, is easy to be interpreted. It provides reliable predictive accuracy while leading to insights into the understanding of the squeezing problem. The strength-stress ratio (SSR) is suggested to be the most important factor. Moreover, the implementation of the updating procedures is efficient since only a simple field test (e.g. Point Load index or Schmidt rebound index) is required. Multiple rounds of predictions within the updating process allow different levels of prediction, for example long-range, short-term, or immediate, to be extracted as useful information towards the decision-making of construction operations. Therefore, this framework can serve as a pragmatic tool to assist the selection of optimal primary-support and other construction strategies based on the potential squeezing risk.

Stability evaluation of a double-deck tunnel with diverging section

Due to the various restrictions and problems related to the construction of new roads in urban areas, underground road construction has been receiving a great deal of attention in the field of tunnel engineering. In this study, a double-deck road tunnel with a diverging section was analyzed for the evaluation of its stability. Both numerical analysis and scale model tests were performed, the results were used to develop a stability evaluation method for double-deck tunnels with diverging sections constructed in rocks by NATM. From regression analyses conducted on the results of the numerical analysis, an equation and a chart were derived, these tools allow us to obtain the strength/stress ratio (SSR) for double-deck road tunnels with a diverging tunnel in various diverging conditions quickly and accurately. These tools have great potential to help engineers evaluate the stability of double-deck tunnels in the preliminary design stage.

Investigating the effect of simultaneous extraction of two longwall panels on a maingate gateroad stability using numerical modeling

In longwall mining method, the stability of the maingate and tailgate gateroads is of great importance because these excavations are used for essential services such as ventilation and transportation. The aim of this paper is to develop a numerical model to investigate the stability of a maingate gateroad named P8-maingate during caving and extraction of P7 and P8 coal panels of Tabas Parvadeh underground mine. Stability analysis is carried out by critical shear strain criterion, zone plasticity conditions, support system efficiency and strength-stress ratio conditions using FLAC3D, a software based on finite difference numerical methods. In the critical shear strain criterion method, the stability analysis is conducted by comparing the shear strain increments of the element in the model and Sakurai critical shear strain. In the second stability analysis method based on zone plasticity, the zones indicating shear plastic failures are the potential instability-causing regions. In the support system efficiency method, the applied load and pressure are compared to the yielding strength of the support systems to check their stability. In the Strength-Stress Ratio (SSR) method, a safety factor is determined for the gates, and the strength versus stress plots of the gates are assessed to control the stability condition. The results of numerical modeling indicate the stability of the P8-maingate using these methods. The modeling results are then compared to the instrumentation data to investigate the consistency where a good agreement was observed.

Energy Release Analysis of a Severe Rockburst in a Headrace Tunnel Crossing a Tectonic Stress Zone

When tunneling in a hard and brittle rock mass within a tectonic stress zone, dynamic failure of rock mass‐rockburst may occur. Considering the occurrence of rockburst is generally induced by a sudden release of storage elastic energy, a numerical analysis based on the geotechnical conditions of the headrace tunnels of the Neelum–Jhelum hydroelectric project was carried out to investigate the variations of the storage elastic energy of surrounding rock mass during excavation in the tectonic stress zone. As expected, the numerical results show that the storage elastic energy concentration zones is elliptical around the tunnel due to the influence of the tectonic stress field and that the long axis of the ellipse is perpendicular to the orientation of the maximum principal stress of the tectonic stress. Furthermore, the calculated storage energy concentration zone is consistent with the locations of blasting overbreak in the tunnel. Rockburst predictions were carried out using the strength‐stress ratio and energy criteria to identify the applicability of the criteria in a tectonic stress zone. The comparisons between the predictions and the field observations show that the strength‐stress ratio criteria based on the uniaxial tests do not consider the influence of the tectonic stress on the strength of the rock. These criteria overpredict the extent of the blasting pits in the tectonic stress zone. However, the energy criteria based on the energy conversion of unloading confining pressure tests are able to reflect the influence of the tectonic stress, and the prediction results are more close to the field observations.

An Empirical Approach for Tunnel Support Design through Q and RMi Systems in Fractured Rock Mass

Empirical systems for the classification of rock mass are used primarily for preliminary support design in tunneling. When applying the existing acceptable international systems for tunnel preliminary supports in high-stress environments, the tunneling quality index (Q) and the rock mass index (RMi) systems that are preferred over geomechanical classification due to the stress characterization parameters that are incorporated into the two systems. However, these two systems are not appropriate when applied in a location where the rock is jointed and experiencing high stresses. This paper empirically extends the application of the two systems to tunnel support design in excavations in such locations. Here, the rock mass characterizations and installed support data of six tunnel projects are used. The back-calculation approach is used to determine the Q value using the Q-system support chart, and these values are then used to develop the equations and charts to characterize the stress reduction factor (SRF), which is also numerically evaluated. These equations and charts reveal that the SRF is a function of relative block size, strength–stress ratio, and intact rock compressive strength. Furthermore, the RMi-suggested supports were heavier than the actual installed ones. If the approximate inverse relation between stress level (SL) and SRF is used, the difference between the actual and the recommended supports increases when using the RMi-recommended rock support chart for blocky ground. An alternate system is made for support recommendation using a Q-system support chart. In this system, the ground condition factor is modified from the available parameters, and a correlation is developed with a modified Q system.

Extension of tunneling quality index and rock mass rating systems for tunnel support design through back calculations in highly stressed jointed rock mass: An empirical approach based on tunneling data from Himalaya

A preliminary support design is the basic output of an empirical rock mass classification system. Tunneling quality index (Q) and rock mass rating (RMR) systems are used specifically for tunnel support in moderately jointed and massive rocks; however, they do not provide any information/guidance for jointed rock masses in a highly stressed environment. Hence, this study focuses on the already supported drill and blast tunnel sections mapping data of four tunneling projects in Pakistan. The study extends the application of these systems to the tunnel support design for highly stressed jointed rock mass through an empirical approach. In each empirical approach, the parameters for the stress condition are suggested. Rock mass quality (Q or Qc) is determined from back calculations using an installed support and tunnel span. Empirical equations and charts are proposed for the stress reduction factor (SRF) characterization. In the proposed equations and charts, the SRF is a function of intact rock strength, relative block size, and strength–stress ratio. In the case of the RMR system, approximately 90% of the sections show that the actual supports are heavier than the suggested supports by RMR89. The RMR was determined from the installed support for each tunnel section through the back calculations. To select the adjustment rating factor for stress, three hypotheses are considered based on the intact rock strength to the major principal stress ratio. The RMR14 suggested support revealed that all sections are heavily supported. A strong correlation exists between RMR89 and RMR14. Three hypotheses are considered for the stress adjustment factor rating selection for RMR14 based on the correlation equation between RMR89 and RMR14. For the evaluation, the application of modified Q and RMR systems are used in the tunnel support in a case study. The exploration reports show that the tunnel will pass through a jointed rock mass under high in-situ stress environments. The comparison shows that heavy support is recommended from the modified systems for tunnel stability.

Predicting tunnel squeezing with incomplete data using Bayesian networks

Tunnel squeezing or time-dependent large deformations due to creep are common in tunnels constructed in weak rock masses at large depth or subjected to high horizontal in situ stresses in tectonically active regions. Squeezing can produce tunnel collapses, budget overruns and construction delays, and being able to predict squeezing is therefore important. This study presents a novel application of Bayesian networks (BNs) to predict squeezing. In particular, we employ a Naïve Bayes classifier based on five parameters – support stiffness (K), Rock Tunneling Quality Index (Q), tunnel depth (H), tunnel diameter (D), and strength–stress ratio (SSR) – about which information is commonly available at early design stages. The Naïve Bayes classifier is “learned”, using the Expectation Maximization algorithm, with a database of 166 tunneling case histories from 7 countries compiled by the authors which is provided as Supplementary material. Then, the Junction Tree algorithm is employed for “belief updating”; i.e., to predict the probability of tunnel squeezing for a given set of (probably incomplete) evidence. The model is validated using 10-fold cross-validation and also using an additional set of case-histories that had not been originally employed to learn the network. Results show that, when compared with other available criteria, the error rate of our BN is among the lowest, but with the advantage that it is able to provide predictions even with incomplete data. Results of a sensitivity analysis to assess the importance of input parameters on the squeezing outcome are also presented. And, finally, a web-based implementation of the proposed BN is provided to improve the ease-of-use of our approach.