Correlation between the rock mass properties and maximum horizontal stress:A case study of overcoring stress measurements

Understanding the mechanical properties of the lithologies is crucial to accurately determine the horizontal stress magnitude.To investigate the correlation between the rock mass properties and maximum horizontal stress,the three-dimensional(3D)stress tensors at 89 measuring points determined using an improved overcoring technique in nine mines in China were adopted,a newly defined characteristic parameter C_(ERP)was proposed as an indicator for evaluating the structural properties of rock masses,and a fuzzy relation matrix was established using the information distribution method.The results indicate that both the vertical stress and horizontal stress exhibit a good linear growth relationship with depth.There is no remarkable correlation between the elastic modulus,Poisson's ratio and depth,and the distribution of data points is scattered and messy.Moreover,there is no obvious relationship between the rock quality designation(RQD)and depth.The maximum horizontal stress σ_(H) is a function of rock properties,showing a certain linear relationship with the C_(ERP)at the same depth.In addition,the overall change trend of σ_(H) determined by the established fuzzy identification method is to increase with the increase of C_(ERP).The fuzzy identification method also demonstrates a relatively detailed local relationship betweenσ_H and C_(ERP),and the predicted curve rises in a fluctuating way,which is in accord well with the measured stress data.

Evaluation of slope stability through rock mass classification and kinematic analysis of some major slopes along NH-1A from Ramban to Banihal, North Western Himalayas

The network of Himalayan roadways and highways connects some remote regions of valleys or hill slopes,which is vital for India’s socio-economic growth.Due to natural and artificial factors,frequency of slope instabilities along the networks has been increasing over last few decades.Assessment of stability of natural and artificial slopes due to construction of these connecting road networks is significant in safely executing these roads throughout the year.Several rock mass classification methods are generally used to assess the strength and deformability of rock mass.This study assesses slope stability along the NH-1A of Ramban district of North Western Himalayas.Various structurally and non-structurally controlled rock mass classification systems have been applied to assess the stability conditions of 14 slopes.For evaluating the stability of these slopes,kinematic analysis was performed along with geological strength index(GSI),rock mass rating(RMR),continuous slope mass rating(CoSMR),slope mass rating(SMR),and Q-slope in the present study.The SMR gives three slopes as completely unstable while CoSMR suggests four slopes as completely unstable.The stability of all slopes was also analyzed using a design chart under dynamic and static conditions by slope stability rating(SSR)for the factor of safety(FoS)of 1.2 and 1 respectively.Q-slope with probability of failure(PoF)1%gives two slopes as stable slopes.Stable slope angle has been determined based on the Q-slope safe angle equation and SSR design chart based on the FoS.The value ranges given by different empirical classifications were RMR(37-74),GSI(27.3-58.5),SMR(11-59),and CoSMR(3.39-74.56).Good relationship was found among RMR&SSR and RMR&GSI with correlation coefficient(R 2)value of 0.815 and 0.6866,respectively.Lastly,a comparative stability of all these slopes based on the above classification has been performed to identify the most critical slope along this road.

Probabilistic analysis of tunnel face seismic stability in layered rock masses using Polynomial Chaos Kriging metamodel

Face stability is an essential issue in tunnel design and construction.Layered rock masses are typical and ubiquitous;uncertainties in rock properties always exist.In view of this,a comprehensive method,which combines the Upper bound Limit analysis of Tunnel face stability,the Polynomial Chaos Kriging,the Monte-Carlo Simulation and Analysis of Covariance method(ULT-PCK-MA),is proposed to investigate the seismic stability of tunnel faces.A two-dimensional analytical model of ULT is developed to evaluate the virtual support force based on the upper bound limit analysis.An efficient probabilistic analysis method PCK-MA based on the adaptive Polynomial Chaos Kriging metamodel is then implemented to investigate the parameter uncertainty effects.Ten input parameters,including geological strength indices,uniaxial compressive strengths and constants for three rock formations,and the horizontal seismic coefficients,are treated as random variables.The effects of these parameter uncertainties on the failure probability and sensitivity indices are discussed.In addition,the effects of weak layer position,the middle layer thickness and quality,the tunnel diameter,the parameters correlation,and the seismic loadings are investigated,respectively.The results show that the layer distributions significantly influence the tunnel face probabilistic stability,particularly when the weak rock is present in the bottom layer.The efficiency of the proposed ULT-PCK-MA is validated,which is expected to facilitate the engineering design and construction.

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.

Effect of eccentric and inclined loading on the bearing capacity of strip footing placed on rock mass

This paper deals with the bearing capacity determination of strip footing on a rock mass in hilly area by considering the influence of inclined and eccentric loading. Applying the generalized HoekBrown failure criterion, the failure behavior of the rock mass is modeled with the help of the power cone programming in the lower bound finite element limit analysis framework. Using bearing capacity factor(Ns), the change in bearing capacity of the strip footing due to the occurrence of eccentrically inclined loading is presented. The variations of the magnitude of Ns are obtained by examining the effects of the Hoek-Brown rock mass strength parameters(uniaxial compressive strength(sci), disturbance factor(D), rock parameter(mi), and Geological Strength Index(GSI)) in the presence of different magnitudes of eccentricity(e) and inclination angle(λ) with respect to the vertical plane, and presented as design charts. Both the inclined loading modes, i.e., inclination towards the center of strip footing(+λ) and inclination away from the center of strip footing(-λ), are adopted to perform the investigation. In addition, the correlation between the input parameters and the corresponding output is developed by utilizing the artificial neural network(ANN). Additionally, from sensitivity analysis, it is observed that inclination angle(λ) is the most sensitive parameter. For practicing engineers, the obtained design equation and design charts can be beneficial to understand the bearing capacity variation in the existence of eccentrically inclined loading in mountain areas.

Numerical manifold method for thermo-mechanical coupling simulation of fractured rock mass

As a calculation method based on the Galerkin variation,the numerical manifold method(NMM)adopts a double covering system,which can easily deal with discontinuous deformation problems and has a high calculation accuracy.Aiming at the thermo-mechanical(TM)coupling problem of fractured rock masses,this study uses the NMM to simulate the processes of crack initiation and propagation in a rock mass under the influence of temperature field,deduces related system equations,and proposes a penalty function method to deal with boundary conditions.Numerical examples are employed to confirm the effectiveness and high accuracy of this method.By the thermal stress analysis of a thick-walled cylinder(TWC),the simulation of cracking in the TWC under heating and cooling conditions,and the simulation of thermal cracking of the SwedishÄspöPillar Stability Experiment(APSE)rock column,the thermal stress,and TM coupling are obtained.The numerical simulation results are in good agreement with the test data and other numerical results,thus verifying the effectiveness of the NMM in dealing with thermal stress and crack propagation problems of fractured rock masses.

Investigation of anisotropic strength criteria for layered rock mass

Layered rock mass is a type of engineering rock mass with sound mechanical anisotropy,which is generally unfavorable to the stability of underground works.To investigate the strength anisotropy of layered rock,the Mohr-Coulomb and Hoek-Brown criteria are introduced to establish the two transverse isotropic strength criteria based on Jaeger's single weak plane theory and maximum axial strain theory,and parameter determination methods.Furthermore,the sensitivity of strength parameters(K 1,K 2,and K 3)that are used to characterize the anisotropy strength of non-sliding failure involved in the strength criteria and confining pressure are investigated.The results demonstrate that strength parameters K 1 and K 2 affect the strength of layered rock samples at all bedding angles except for the bedding angle of 90°and the angle range that can cause the shear sliding failure along the bedding plane.The strength of samples at any bedding angle decreases with increasing K 1,whereas the opposite is for K 2.Except for bedding angles of 0°and 90°and the bedding angle range that can cause the shear sliding along the bedding plane,K 3 has an impact on the strength of rock samples with other bedding angles that the specimens'strength increases with increase of K 3.In addition,the strength of the rock sample increases as confining pressure rises.Furthermore,the uniaxial and triaxial tests of chlorite schist samples were carried out to verify and evaluate the strength criteria proposed in the paper.It shows that the predicted strength is in good agreement with the experimental results.To test the applicability of the strength criterion,the strength data of several types of rock in the literature are compared.Finally,a comparison is made between the fitting effects of the two strength criteria and other available criteria for layered rocks.

Rock Mass Quality Rating Based on the Multi-Criteria Grey Metric Space

Assessment of rock mass quality significantly impacts the design and construction of underground and open-pit mines from the point of stability and economy.This study develops the novel Gromov-Hausdorff distance for rock quality(GHDQR)methodology for rock mass quality rating based on multi-criteria grey metric space.It usually presents the quality of surrounding rock by classes(metric spaces)with specified properties and adequate interval-grey numbers.Measuring the distance between surrounding rock sample characteristics and existing classes represents the core of this study.The Gromov-Hausdorff distance is an especially useful discriminant function,i.e.,a classifier to calculate these distances,and assess the quality of the surrounding rock.The efficiency of the developed methodology is analyzed using the Mean Absolute Percentage Error(MAPE)technique.Seven existing methods,such as the Gaussian cloud method,Discriminant method,Mutation series method,Artificial neural network(ANN),Support vector machine(SVM),Grey wolf optimizer and Support vector classification method(GWO-SVC)and Rock mass rating method(RMR)are used for comparison with the proposed GHDQR method.The share of the highly accurate category of 85.71%clearly indicates compliance with actual values obtained by the compared methods.The results of comparisons showed that the model enables objective,efficient,and reliable assessment of rock mass quality.

Inverting the rock mass P-wave velocity field ahead of deep buried tunnel face while borehole drilling

Imaging the wave velocity field surrounding a borehole while drilling is a promising and urgently needed approach for extending the exploration range of the borehole point.This paper develops a drilling process detection(DPD)system consisting of a multifunctional sensor and a pilot geophone installed at the top of the drilling rod,geophones at the tunnel face,a laser rangefinder,and an onsite computer.A weighted adjoint-state first arrival travel time tomography method is used to invert the P-wave velocity field of rock mass while borehole drilling.A field experiment in the ongoing construction of a deep buried tunnel in southwestern China demonstrated the DPD system and the tomography method.Time-frequency analysis of typical borehole drilling detection data shows that the impact drilling source is a pulse-like seismic exploration wavelet.A velocity field of the rock mass in a triangular area defined by the borehole trajectory and geophone receiving line can be obtained.Both the borehole core and optical image validate the inverted P-wave velocity field.A numerical simulation of a checkerboard benchmark model is used to test the tomography method.The rapid convergence of the misfits and consistent agreement between the inverted and observed travel times validate the P-wave velocity imaging.

A Novel ISSA–DELM Model for Predicting Rock Mass Permeability

In pumped storage projects,the permeability of rock masses is a crucial parameter in engineering design and construction.The rock mass permeability coefficient(K)is influenced by various geological parameters,and previous studies aimed to establish an accurate relationship between K and geological parameters.This study uses the improved sparrow search algorithm(ISSA)to optimize the parameter settings of the deep extreme learning machine(DELM),constructing a prediction model with flexible parameter selection and high accuracy.First,the Spearman method is applied to analyze the correlation between geological parameters.A sample database is built by comprehensively selecting four geological indexes:burial depth,rock quality designation(RQD),fracture density characteristic index(FD),and rock mass integrity designation(RID).Hence,the defects of the sparrow search algorithm(SSA)are enhanced using the improved strategy,and the initial input weights of the DELM are optimized.Finally,the proposed ISSA–DELM model is employed to predict the permeability coefficient of rock mass in the entire study area.The results showed that the predictive performance of the model is superior to that of the DELM and SSA–DELM.Therefore,this model successfully provides insights into the distribution characteristics of rock mass permeability at engineering sites.

Load-bearing characteristics and energy evolution of fractured rock masses after granite and sandstone grouting

Experiments on grouting-reinforced rock mass specimens with different particle sizes and features were carried out in this study to examine the effects of grouting reinforcement on the load-bearing characteristics of fractured rock mass.The strength and deformation features of grouting-reinforced rock mass were analyzed under different loading manners;the energy evolution mechanism of grouting-reinforced rock mass specimens with different particle sizes and features was investigated;the energy dissipation ratio and post-peak stress decreasing rate were employed to evaluate the bearing stability of grouting-reinforced rock mass.The results show that the strength and ductility of granite-reinforced rock mass(GRM)under biaxial loading are higher than that of sandstone-reinforced rock mass(SRM)under uniaxial loading.Besides,the energy evolution characteristics of grouting-reinforced rock mass under uniaxial and biaxial loading mainly could be divided into early,middle,and late stages.In the early stage,total,elastic,and dissipation energies were quite small with flatter curves;in the middle stage,elastic energy increased rapidly,whereas dissipation energy increased slowly;in the late stage,dissipation energy increased sharply.The energy dissipation ratio was used to represent the pre-peak plastic deformation.Under uniaxial loading,this ratio increased as the particle size increased and the pre-peak plastic deformation of grouting-reinforced rock mass became larger;under biaxial loading,it dropped as the particle size increased,and the pre-peak plastic deformation of grouting-reinforced rock mass became smaller.The post-peak stress decline rate A_(v) was used to assess the post-peak bearing performance of grouting-reinforced rock mass.Under uniaxial loading,parameter A_(v) exhibited reduction as the particle size kept increasing,and the ability of post-peak of grouting-reinforced rock mass to allow deformation development was greater,and the bearing capacity was greater;under biaxial loading,A_(v) increased with the particle size,and the ability of post-peak of grouting-reinforced rock mass to allow deformation development was low and the bearing capacity was reduced.The findings are considered instrumental in improving the stability of the roadway-surrounding rock by granite and sandstone grouting.

Failure behavior and strength model of blocky rock mass with and without rockbolts

To better understand the failure behaviours and strength of bolt-reinforced blocky rocks,large scale extensive laboratory experiments are carried out on blocky rock-like specimens with and without rockbolt reinforcement.The results show that both shear failure and tensile failure along joint surfaces are observed but the shear failure is a main controlling factor for the peak strength of the rock mass with and without rockbolts.The rockbolts are necked and shear deformation simultaneously happens in bolt reinforced rock specimens.As the joint dip angle increases,the joint shear failure becomes more dominant.The number of rockbolts has a significant impact on the peak strain and uniaxial compressive strength(UCS),but little influence on the deformation modulus of the rock mass.Using the Winkler beam model to represent the rockbolt behaviours,an analytical model for the prediction of the strength of boltreinforced blocky rocks is proposed.Good agreement between the UCS values predicted by proposed model and obtained from experiments suggest an encouraging performance of the proposed model.In addition,the performance of the proposed model is further assessed using published results in the literature,indicating the proposed model can be used effectively in the prediction of UCS of bolt-reinforced blocky rocks.

Tree-Based Solution Frameworks for Predicting Tunnel Boring Machine Performance Using Rock Mass and Material Properties

Tunnel Boring Machines(TBMs)are vital for tunnel and underground construction due to their high safety and efficiency.Accurately predicting TBM operational parameters based on the surrounding environment is crucial for planning schedules and managing costs.This study investigates the effectiveness of tree-based machine learning models,including Random Forest,Extremely Randomized Trees,Adaptive Boosting Machine,Gradient Boosting Machine,Extreme Gradient Boosting Machine(XGBoost),Light Gradient Boosting Machine,and CatBoost,in predicting the Penetration Rate(PR)of TBMs by considering rock mass and material characteristics.These techniques are able to provide a good relationship between input(s)and output parameters;hence,obtaining a high level of accuracy.To do that,a comprehensive database comprising various rock mass and material parameters,including Rock Mass Rating,Brazilian Tensile Strength,and Weathering Zone,was utilized for model development.The practical application of these models was assessed with a new dataset representing diverse rock mass and material properties.To evaluate model performance,ranking systems and Taylor diagrams were employed.CatBoost emerged as the most accurate model during training and testing,with R2 scores of 0.927 and 0.861,respectively.However,during validation,XGBoost demonstrated superior performance with an R2 of 0.713.Despite these variations,all tree-based models showed promising accuracy in predicting TBM performance,providing valuable insights for similar projects in the future.

Deformation characteristics and damage ontologies of soft and hard composite rock masses under impact loading

As one of the most common occurring geological landforms in deep rock formations, the dynamic mechanical properties of layered composite rock bodies under impact loading have been widely studied by scholars. To study the dynamic properties of soft and hard composite rocks with different thickness ratios, this paper utilizes cement, quartz sand and gypsum powder to construct soft and hard composite rock specimens and utilizes a combination of indoor tests, numerical calculations, and theoretical analyses to investigate the mechanical properties of soft and hard composite rock bodies. The test results reveal that:(1) When the proportion of hard rock increases from 20% to 50%, the strength of the combined rock body increases by 69.14 MPa and 87 MPa when the hard rock face and soft rock face are loaded, respectively;however, when the proportion of hard rock is the same, the compressive strength of the hard rock face impact is 9%-17% greater than that of the soft rock face impact;(2) When a specimen of soft and hard combined rock body is subjected to impact loading, the damage mode involves mixed tension and shear damage, and the cracks generally first appear at the ends of the specimen, then develop on the laminar surface from the impact surface, and finally end in the overall damage of the soft rock part. The development rate and the total number of cracks in the same specimen when the hard rock face is impacted are significantly greater than those when the soft rock face is impacted;(3) By introducing Weibull’s statistical strength theory to establish the damage variables of soft-hard combined rock bodies, combined with the DP strength criterion, the damage model and the Kelvin body are concatenated to obtain a statistical damage constitutive model, which can better fit the full stress-strain curve of soft-hard combined rock body specimens under a single impact load.

A modified smoothed particle hydrodynamics method considering residual stress for simulating failure and its application in layered rock mass

Residual strength is an indispensable factor in evaluating rock fracture,yet the current Smoothed Particle Hydrodynamics(SPH)framework rarely considers its influence when simulating fracture.An improved cracking strategy considering residual stress in the base bond SPH method was proposed to simulate failures in layered rocks and slopes and verified by experimental results and other simulation methods(i.e.,the discrete element method).Modified Mohr–Coulomb failure criterion was applied to distinguish the mixed failure of tensile and shear.Bond fracture markψwas introduced to improve the kernel function after tensile damage,and the calculation of residual stress after the damage was derived after shear damage.Numerical simulations were carried out to evaluate its performance under different stress and scale conditions and to verify its effectiveness in realistically reproducing crack initiation and propagation and coalescence,even fracture and separation.The results indicate that the improved cracking strategy precisely captures the fracture and failure pattern in layered rocks and rock slopes.The residual stress of brittle tock is correctly captured by the improved SPH method.The improved SPH method that considers residual strength shows an approximately 13%improvement in accuracy for the safety factor of anti-dip layered slopes compared to the method that does not consider residual strength,as validated against analytical solutions.We infer that the improved SPH method is effective and shows promise for applications to continuous and discontinuous rock masses.

Restraint effect of partition wall on the tunnel floor heave in layered rock mass

The presence of horizontal layered rocks in tunnel engineering significantly impacts the stability and strength of the surrounding rock mass,leading to floor heave in the tunnel.This study focused on preparing layered specimens of rock-like material with varying thickness to investigate the failure behaviors of tunnel floors.The results indicate that thin-layered rock mass exhibits weak interlayer bonding,causing rock layers near the surface to buckle and break upwards when subjected to horizontal squeezing.With an increase in the layer thickness,a transition in failure mode occurs from upward buckling to shear failure along the plane,leading to a noticeable reduction in floor heave deformation.The primary cause of significant deformation in floor heave is upward buckling failure.To address this issue,the study proposes the installation of a partition wall in the middle of the floor to mitigate heave deformation of the rock layers.The results demonstrate that the partition wall has a considerable stabilizing effect on the floor,reducing the zone of buckling failure and minimizing floor heave deformation.It is crucial for the partition wall to be sufficiently high to prevent buckling failure and ensure stability.Through simulation calculations on an engineering example,it is confirmed that implementing a partition wall can effectively reduce floor heave and enhance the stability of tunnel floor.

Frost heaving and crack initiation characteristics of tunnel rock mass in cold regions under low-temperature degradation

Water freezing in rock fractures causes volumetric expansion and fracture development through frost heaving.This study introduces a novel analytical model to investigate how uneven freezing force and surrounding rock pressure influence fracture initiation,based on mass conservation,elasticity,and water-ice phase transition principles.A model for rock fracture initiation considering freezing temperature,uneven freezing expansion,in-situ stress,and lateral pressure was proposed based on fracture mechanics.Equations for stress intensity factors were developed and validated using the phase field method.The effects of rock elastic modulus anisotropy and critical fracture energy density on fracture initiation were also discussed.The results show that the values of KI and KII exhibit an upward trend as the freezing temperature,uneven expansion,in-situ stress,and lateral pressure increase.The uneven freezing expansion has the most significant influence on KI and KII values among these parameters.As the uneven freezing expansion coefficient increases to 0.5,the fracture initiation mode shifts from tensile fracture to shear fracture.As the lateral pressure coefficient increases to 1,the fracture initiation mode shifts from tensile fracture to shear fracture.Rock elastic modulus anisotropy causes fractures to propagate in a clockwise direction,forming a'butterfly'pattern.Critical fracture energy density an isotropy causes counterclockwise deviation in propagation direction,resulting in branching paths and an'H'-shaped pattern.

Strength and damage evolution mechanism of rock mass with holes under cyclic loading

The damage and failure law of rock mass with holes is of great significance to the stability control of roadways. This study investigates the mechanical properties and failure modes of porous rock masses under cyclic loading, elucidates the acoustic emission (AE) characteristics and their spatial evolution, and establishes the interrelation among AE, stress, strain, time, and cumulative damage. The results reveal that the rock mass with holes and the intact rock mass show softening and hardening characteristics after cyclic loading. The plastic strain of the rock mass with holes is smaller than that of the intact rock mass, and the stress −strain curve shows hysteresis characteristics. Under uniaxial compression, the pore-bearing rock mass shows the characteristics of higher ringing count, AE energy, b-value peak, and more cumulative ringing count in the failure stage, while it shows lower characteristics under cyclic action. At the initial stage of loading, compared with the intact rock mass, the pore-containing rock mass shows the characteristics of a low b-value. The AE positioning and cumulative damage percentage are larger, and the AE positioning is denser around the hole. The specimen with holes is mainly shear failure, and the complete specimen is mainly tensile shear failure.

Failure mechanism and safety control technology of a composite strata roadway in deep and soft rock masses:a case study

The construction of coal mines often encounters deep composite soft rock roadways,which is characterized by significant deformation and poor stability.To deeply study the failure mechanism and large deformation challenges of a composite strata roadway in deep and soft rock masses,a numerical model of 3DEC tetrahedral blocks was established based on the method of rock quality designation(RQD).The results showed that original support cannot prevent asymmetric failure and large deformation due to the adverse geological environment and unsuitable support design.According to the failure characteristics,a coupling support of“NPR bolt/cable+mesh+shotcrete+steel pipe”was proposed to control the stability of the surrounding rock.The excellent mechanical properties of large deformation(approximately 400 mm)and high constant resistance force(bolt with 180 k N;cable with 350 k N)were evaluated by the tensile tests.The numerical results showed that the maximum deformation was minimized to 243 mm,and the bearing capacity of the surrounding rock of the roadway was enhanced.The field test results showed that the maximum deformation of the surrounding rock was 210 mm,and the forces of the NPR bolt and cable were stable at approximately 180 k N and 350 k N,respectively.This demonstrated the effectiveness of the coupling support with the NPR bolt and cable,which could be a guiding significance for the safety control of large deformation and failure in deep composite soft rock roadways.

Prediction of rock mass classification in tunnel boring machine tunneling using the principal component analysis (PCA)-gated recurrent unit (GRU) neural network

Due to the complexity of underground engineering geology,the tunnel boring machine(TBM)usually shows poor adaptability to the surrounding rock mass,leading to machine jamming and geological hazards.For the TBM project of Lanzhou Water Source Construction,this study proposed a neural network called PCA-GRU,which combines principal component analysis(PCA)with gated recurrent unit(GRU)to improve the accuracy of predicting rock mass classification in TBM tunneling.The input variables from the PCA dimension reduction of nine parameters in the sample data set were utilized for establishing the PCA-GRU model.Subsequently,in order to speed up the response time of surrounding rock mass classification predictions,the PCA-GRU model was optimized.Finally,the prediction results obtained by the PCA-GRU model were compared with those of four other models and further examined using random sampling analysis.As indicated by the results,the PCA-GRU model can predict the rock mass classification in TBM tunneling rapidly,requiring about 20 s to run.It performs better than the previous four models in predicting the rock mass classification,with accuracy A,macro precision MP,and macro recall MR being 0.9667,0.963,and 0.9763,respectively.In Class II,III,and IV rock mass prediction,the PCA-GRU model demonstrates better precision P and recall R owing to the dimension reduction technique.The random sampling analysis indicates that the PCA-GRU model shows stronger generalization,making it more appropriate in situations where the distribution of various rock mass classes and lithologies change in percentage.