Rock mass quality classification is the foundation of initial underground engineering design and construction.In order to evaluate rock mass quality more accurately，this study used beluga whale optimization（BWO）to optimize random forest model（RF），a BWO-RF model which can be used for rock mass quality evaluation was proposed.At the same time，the rock mass quality evaluation models of sparrow search al-gorithm optimized random forest（SSA-RF），particle swarm optimization optimized random forest（PSO-RF）and non-optimized random forest（RF） were constructed for comparison.Before the models construction，a data-base containing 131 engineering cases data was established through literature review and field test data collec-tion.After writing the code of models construction，the training and testing of the four models were completed by using the database.Based on the model test results，five model evaluation indexes，accuracy，precision，recall，F1 score and AUC，were used to compare and select the best model of the four kinds of rock mass quality eva-luation models.The results show that the BWO-RF model has the best performance among the four kinds of rock mass quality evaluation models，and each evaluation indexes of model are better than the other three mo-dels，indicating that the BWO-RF model has better practicability in the evaluation of rock mass quality.Through the test set，the prediction accuracy of BWO-RF model proposed in this study is 90%，which can provide a reliable reference for practical engineering construction and has practical engineering application value.

In view of the low accuracy and uneven water distribution of water content in rock samples with different water saturations，a preparation method of rock samples with different water saturations and uniform distribution of water content was proposed based on the principle of osmotic technology and chemical thermodynamics.The basic principle consists of two parts：One is to control the concentration and molecular weight size of polyethylene glycol（PEG） in organic solution，so as to precisely control the matrix suction values of the solution water and water in the sample to determine the final water saturation of the sample.The other is to control the type of supersaturated solution placed in the constant humidity environment of the rock sample，so as to precisely control the chemical potentials of the component water when the total weight of the sample is stable，and the final water saturation of the sample is determined. The water saturation of rock samples prepared by this mothed are qualified samples with uniform saturation distribution.The water rationality of the method for accurately preparing rock samples with uniform water saturation distribution was also verified by nuclear magnetic resonance imaging（NMRI） experiments.Meanwhile，a device is designed independently to prepare rock samples with different saturations under constant humidity environment，and the constant strain rate compression test of sandstone under different saturation conditions was carried out by using the device.The test results show that the uniaxial compressive strength and elastic modulus of sandstone decreases with the increase of water saturation.After the peak，the post-peak stress decline slows and the brittleness of the sample decreases with the increase of water saturation. With the increase of confining pressure，the compaction stage of sandstone sample is weakened，and the deformation of the sample undergoes the process of elastic-brittle→elastic-plastic→strain hardening.The results of the study provide new methods and ideas for the accurate preparation of rock samples with different water saturations.

To delve into the size-related effects and directional properties of the roughness on rock fracture surfaces，this paper focus on analyzing the roughness correction coefficient C within the refined formula of the cubic law，as it effectively characterizes the roughness of natural rock fractures. Since Louis first introduced the cubic law，numerous scholars have since proposed numerous modified formulas，including those by Zhang Youtian，Zimmerman，and Xiong et al. Through conversion，we acquire the calculation formula for the roughness correction coefficient and utilize collected natural fractures for related computations.High-precision 3D scanning technology was used to scan natural rock samples and acquire roughness data of fracture surfaces. By combining this data with publicly available high-precision CT scan data of rough rock fractures，we generate spatial coordinates.Both formulaic and numerical methods were used to calculate and analyze the roughness correction coefficient C. Using the formulaic approach，the roughness correction coefficient for sample sizes ranging from 10% to 100% of the fracture surface were calculated and varying results were obtained，which indicates that roughness exhibits a scale effect. According to the numerical method，a well-fitting ellipse was obtained，indicating that the roughness correction coefficient possesses directionality and can be expressed using tensor notation. This conclusion is further supported by calculating the JRC value of rough fractures and their surface and crack width fractal dimensions，revealing that the roughness of fracture surfaces exhibits scale effects and anisotropy. Upon further investigation，it is discovered that the roughness correction coefficient tensor or roughness tensor，when combined with average crack width，can be utilized to form a single crack permeability tensor that quantifies the rough surface in complex rough crack network models

A suitable charging structure in blasting engineering can effectively improve the utilization rate of explosives and thus improve the blasting effect.Based on the RHT（Riedel-Hiermaier-Thomamodel）dynamic response mechanics relationship，ANSYS/LS-DYNA software was used to study the effect of the uncoupling coefficient K on the blasting load of rock body under the condition of eccentric uncoupled charge and the damage of rock body during blasting.The effective stress，vibration velocity，seismic wave energy and damage conditions under uncoupling coefficients K=1.0，1.5，2.0，2.5，3.0 and 3.5 were analyzed by establishing a single-hole eccentric uncoupled blasting model.In addition，the relationship between burst center distance l and damage degree D，mass vibration velocity PPV and burst center distance l，and vibration velocity v_pp and damage degree D under different conditions of uncoupling coefficients K were investigated.The results show that with the increase of the uncoupling coefficient K，the effective stress and peak vibration velocity of the coupled and uncoupled sides and the peak energy of the seismic wave decreases slightly.The range of the crushed zone and the fracture zone decreases gradually，and the radius of the damage in the crushed zone is larger than the rate of decrease in the fracture zone.The damage degree of rock body and the mass vibration velocity v_pp are gradually reduced with the increase in the burst center distance. The greater the particle vibration velocity v_pp is，the larger the damage degree is.When v_pp reached 50.4 cm/s， the damage variable D in the rock body reached the damage failure threshold of 0.19.When v_pp>140.6 cm/s，the damage degree D>0.80.

The bunch holes，which consist of multiple dense and parallel holes and detonated simultaneously，are widely used in mines because of its high rock-breaking efficiency.With the worldwide depletion of resources in the shallow part of the earth，many mines are transitioning to the deep mining stage，and the influence of the complex high stress environment on the effect of blasting on rock breaking in the deep mining process needs to be further studied.In order to explore the law of blasting damage and fracture of bunch holes cutting blasting in deep mining under high stress stope environment，LS-DYNA finite element numerical analysis software was used to study.Combined with the mine blasting engineering practice，the planar numerical model with the same size as the actual working conditions was established，which contained four charge holes and one auxiliary empty hole.The rupture damage process of the bunch holes under no static stress was analyzed.On this basis，in the case of one-way static stress and two-way equivalent static stress，when the static stress is 20 MPa and 40 MPa respectively，the blasting fracture damage of the bunch holes cutting blasting was analyzed，and the number of failures units of the numerical model under different static stress states was compared. In addition，the hole spacing and the arrangement of peripheral auxiliary holes were optimized under different static stress conditions，and engineering experiments were carried out to verify the reliability of the optimization results by counting the chunk rate of different schemes.The results show that compared with the no static stress condition，the expansion of the bunch holes blasting damage cracks will be suppressed under the unidirectional static stress and bidirectional equivalent static stress conditions.With the increase of static stress，the number of model failure units decreases gradually，and the inhibition of crack extension is more significant.When the static stress reaches 40 MPa，no penetration crack can be formed between the four bunch holes and the center empty hole.In particular，in the bidirectional equivalent static stress state，the inter-hole crack penetration mode of the bunch holes changes from a square penetration to a circular penetration.Practice shows that according to the size of the static stress to adjust the bunch holes spacing and peripheral auxiliary holes arranged way can reduce the rate of large pieces of slot blasting，so as to obtain a good slot blasting rock-breaking effect.

As the demand for transportation increases in China，more and more highway tunnels adopt the form of super-large section.However，there is still a lack of clear guidance for the design and construction of super-large section tunnels with four lanes or more in the current highway tunnel specifications. There is still a need for in-depth research on the section shape，construction methods，and construction mechanics of super-large section tunnels. Existing research indicates that the use of the benching method for excavation construction of super-large section tunnels is feasible. To obtain the optimized schemes for bench length and anchor rod spacing，numerical simulation was performed in this study to investigate seven different scenarios based on the upper and lower bench construction sections of a certain super-large section flat structure tunnel.The scenarios included bench lengths of 30 m，40 m，50 m，and 60 m，as well as anchor rod longitudinal spacings of 1.0 m，1.5 m，and 2.0 m. The distribution patterns of the plastic zone in the surrounding rock mass，the stress of the surrounding rock，and the deformation of the surrounding rock were analyzed.The study results indicate that the overall maximum principal stress around the tunnel shows a trend of “arch foot>arch waist>arch crown”.An increase in bench length or anchor rod spacing leads to a significant increase in the maximum principal stress at the arch crown. Overall，the settlement and horizontal convergence values of the tunnel arch crowns increase with the increase in bench length. An increase in anchor rod spacing requires other support structures in the initial support to exert stronger control over surrounding rock deformation，potentially leading to the destruction of the support structure. Based on the numerical results and on-site monitoring data，considering factors such as construction efficiency and rock stability，the optimized scheme with a bench length of 50 m and an anchor rod spacing of 1 m was applied in the construction site.The results of this research has high reference value for the optimal selection of excavation and support parameters for super-large section flat structure tunnels.

The open-pit mining area is a complex scene，and the traveling obstacle detection is seriously interfered by dust noise such as dust and particles，which makes it difficult to accurately identify obstacles，especially at night when the light is poor，which is not conducive to correct decision-making，thus affecting the safety and overall efficiency of unmanned operation.In view of the above problems，a YOLOv8n-based YOLOv8n-Enhanced algorithm for detecting traveling obstacles in open-pit mining areas was proposed.The algorithm is mainly improved in three aspects：Firstly，for the problems of serious interference by dust noise and poor light at night，a C2fCA module structure was proposed instead of the original C2f module，which utilizes the shared weights and context-aware weights to enhance the dependency relationship between different locations of the image，mitigate the noise interference，and improve the feature extraction ability of the model.Secondly，to trade-off the accuracy and real-time performance of the open-pit obstacle detection model，the Neck end of the model was reconstructed，and the lightweight convolutional techniques GSConv and VoV-GSCSP modules were used to reduce the complexity of the computation and network structure，and realize a higher computational cost-effectiveness of the detector.Finally，for the situation that there is a large gap between the quality of data in the open-pit mining area，especially at night when there is insufficient light，and low-quality data will affect the ability of the model to learn features in training，the loss function was optimized to solve the problem of the bounding box regression equilibrium between the samples of different qualities，to improve the ability of the model to generalize and accelerate the convergence.The experimental results show that the improved algorithm in this paper reduces the computational GFLOPs of the model by about 8.5% and the number of parametric params by about 3% while maintaining the real-time performance，and improves the mean Average Precision（mAP） of the YOLOv8n by 1.8% and 2.6% in daytime and nighttime scenarios，respectively，and realizes obstacle recognition at different scales in daytime and nighttime scenes.

Intelligent mining is an important direction for the future development of the mining industry.The construction of mine intelligent ventilation system is a key part of promoting the digital and intelligent development of mines.Under the current background of energy conservation and emission reduction，the ways of digital and information technology to achieve low-carbon mining，efficient utilization and intelligent control of non-ferrous metal resources，is the important technical guarantee to achieve the strategy of “carbon peaking and carbon neutrality”.The key to realize the high-efficiency utilization of non-ferrous metal resources is to adopt the intelligent ventilation technology to reduce the energy consumption.In the current process of pro-moting the construction of intelligent mine ventilation system，a common hot and difficult issue is to realize the optimal regulation of airflow of mine ventilation system under the condition of ventilation on-demand.The ventilation optimization regulation of intelligent ventilation system requires that on the premise of satisfying the dynamic ventilation on-demand in different periods，the ventilation optimization theory based on fluid network was adopted to obtain a ventilation optimization regulation scheme that meets the requirements of safety，technology and economic.Then the airflow distribution and air pressure distribution of the ventilation network were adjusted to ensure the safe，reliable，stable and economic operation of the mine ventilation system.In order to solve the problem of difficulty in solving the optimization and regulation model of nonlinear mine ventilation network，based on the basic mathematical model of ventilation network airflow regulation，the method of airflow control and optimization of ventilation network was analyzed，and the objectives and constraints of the airflow regulation optimization model based on multi-objective mixed integer programming were analyzed.A two-step airflow control optimization mathematical model based on the mixed integer programming method by improving the two-step ventilation optimization method was proposed.The mathematical model is a multi-objective programming linear model with the goals of minimum ventilation energy consumption，minimum number of regulation equipment and optimal position of regulation equipment，so that the solution result of the mathematical model is more in line with the actual regulation needs.The mathematical model can constraint the number and regulating ways of regulation position by introducing the mixed integer programming method，and constraint the position of regulation schemes according to the actual situation of the mine by introducing the regulation level of branches，which improve the flexibility of ventilation regulation optimization schemes.In addition，by solving the corresponding regulation schemes of airflow distribution multiple times，the mathematical model can obtain a solution scheme that approximately satisfies the solution result of regulation mathematical model with unknown airflow，while avoiding the problem of non-convergence in solving nonlinear models.We construct a calculation example for the mine ventilation optimization regulation problem，and verify the reliability of ventilation regulation mathematical model based on the calculation of airflow distribution.

The high and steep slopes formed by open-pit mining have potential safety hazards due to landslides，posing a significant threat to the safety of downstream residents’ lives and property.Using modern monitoring technology to monitor the stability of high and steep slopes is a necessary means to prevent disasters.The existing research theory basically relies on the corresponding monitoring technology to realize the slope monitoring and management，and lacks the system information integration platform，which is not conducive to the management of the mine in the overall safety production operation process.Therefore，a comprehensive framework for online monitoring data integration on the internet has been designed in this paper by using the principles of radar and GNSS monitoring technology.This framework allows for full time and comprehensive monitoring of displacement，blasting vibration，and other aspects of high and steep slopes in the Nannihu mining site and waste dump.Based on the data information obtained by the monitoring system，the landslide deformation characteristics and landslide inducing factors are analyzed.The appropriate monitoring indicators are selected to establish the threshold model，and the early warning threshold is calculated according to the model，which can divide the corresponding early warning level.The on-line monitoring system realizes the modern slope monitoring target of high precision，large area，non-contact，all-day，all-weather and early detection.The on-line monitoring system is safer，more efficient and more accurate than traditional manual monitoring，and plays a significant role in disaster warning.

Copper is an important key metal raw material.It is of great significance to comprehensively recover and utilize copper ores containing precious metals such as gold and silver.The process mineralogy and beneficiation test of a copper sulfide ore containing gold and silver in Henan Province have been carried out.The results show that the main useful element copper content is 0.82%，the associated beneficial groups are sulfur，gold and silver，the main useful metal minerals are chalcopyrite，chalcocite，pyrite，gangue minerals are mainly quartz.The combined collector of butyl xanthoxanthate and butyl ammonium black is used in mixed flotation，the total amount of agent is 120 g/t，and the process flow of one coarse，two fine and three sweeps is adopted.The new collector TB1021 is used in copper sulfur separation flotation，and the process flow of one coarse，three fine and three sweeps is adopted.Finally，a copper concentrate with a copper grade of 15.21% and a copper recovery of 80.13%，a gold grade of 3.02 g/t，a gold recovery of 66.51%，a silver grade of 160.43 g/t and a silver recovery of 41.82%，and a sulfur concentrate with a sulfur grade of 49.13% and a recovery of 54.34% were obtained.

With the increasing mining depth of metal mines，the problem of high temperature damage in mines is becoming more and more serious，which has become an urgent problem to be solved in deep mining of metal mines.At present，high-temperature mine cooling technology can be divided into two categories：Non-artificial refrigeration cooling technology and artificial refrigeration cooling technology.The latter can effectively solve the cooling problem of high-temperature mines，and is widely used in high-temperature mines at home and abroad.However，a large amount of heat will be generated during the operation of the artificial cooling equipment，which is easy to cause heat accumulation，thus affect the cooling effect.Therefore，it is a reasonable and feasible method to use the local ventilation system of the mine to discharge the heat generated by the operation of the cooling equipment.The high-temperature single-headed excavation roadway on the west side of the -20 m level of the north along the vein in the west section of the Dahongshan copper mine was taken as the research object，and numerical simulations were conducted by Fluent software to investigate the effects of the vent duct height，the distance of the exhaust vent duct lagging the forced vent duct and the exhaust forced ratio on the cooling effect of artificial cooling.The results show that the height of forced vent duct has an obvious influence on the cooling effect，and the optimal height is 1.0 m.The cooling effect is the best when the height of exhaust vent duct is equal to that of forced vent duct，and the optimal height is 1.0 m.The distance between exhaust vent duct and forced vent duct is too large，which is not conducive to cooling，and the optimal distance is 5.0 m.The exhaust forced ratio is too small or too large to form a good wind circulation in the driving drift，and the cooling effect is poor，the optimal exhaust forced ratio is 2.0.The research results can provide guidance for the selection of artificial cooling parameters for high temperature mine ventilation assisted cooling.

In order to realize the prediction of CO concentration in the single-head roadway of the mine，based on the monitoring data of CO concentration in the heading face of the single-head roadway in the 1800 transport lane of Laochang tin mine in Yunnan Province.Firstly，the MATLAB curve fitting toolbox was used to fit the curve of the change of CO concentration with time in the single-head roadway，and the mathematical model of the change of CO concentration with time in the single-head roadway of the mine was established.Through the model，the time required for the CO concentration value in the single-head roadway of the mine to reach the CO concentration value required by the safety regulations was obtained.Then，the convolutional neural network time series prediction model（CNN model） and the BP neural network time series prediction model（BP model） were used to predict the CO concentration，and the two evaluation indexes of R² and RMSE were compared.The results show that the BP neural network time series prediction model has the better prediction effect on the CO concentration of the single-head roadway，which provides an accurate and reliable theoretical basis for the monitoring and control of the CO concentration value of the single-head roadway in the mine.

In order to remove the noise components mixed in the blasting vibration signals of open-pit mine，a noise reduction method based on the PEMD-MPE algorithm was proposed.This algorithm obtains a completely orthogonal Intrinsic Mode Function （IMF） components through Adaptive Orthogonal Empirical Mode Decomposition （PEMD）.Subsequently，it performs a randomness test on the IMF components and calculates its Mean Power Entropy （MPE）.Finally，based on a preset entropy threshold of 0.6，it determines whether a component is noise.If the obtained MPE is greater than 0.6，the component is identified as a noise component and needs to be removed，thus achieving the purpose of noise recluction.Applying this algorithm to denoise measured open-pit mining explosion vibration signals，the results indicate that compared to the EMD-MPE and EEMD-MPE algorithms，the proposed algorithm improves the signal-to-noise ratio by 3.520 dB and 1.107 dB，respectively.It exhibits the best denoising effect，with the smallest reconstruction standard deviation and root mean square error，providing better fidelity to the original signal.Using Adaptive Optimal Kernel （AOK） time-frequency analysis technology to analyze the signal waveforms before and after denoising，a comparison reveals consistent main frequencies.Throughout the denoising process，peak energy and energy in the main frequency band （0~300 Hz） do not show a significant decrease.This indicates that the PEMD-MPE algorithm，while preserving the authenticity of the real signal，more effectively removes noise components.

The application of cantilever type roadheader in coal mines has shown great advantages，and gradually promoted in metal mines.However，the cutting teeth are subjected to huge loads and strong impacts in the process of cutting the rock body，which leads to serious wear and tear of the cutting teeth，thus restricting the further popularization and application of the cantilever type roadheader.In order to optimize the cutting angle and improve the force on the teeth，Rhino software and PFC^3D software were used to establish a rock model of the cutting angle of the teeth，which was simulated and verified by the existing test data，and on the basis of which the linear cutting simulation of the teeth was carried out on the granite by considering the cutting angles of different modifications，such as 45°，50°，55°，60°，65°.The results show that the cutting simulation test based on PFC^3D software has reliable accuracy，and the relative errors between the simulation test and the indoor test results are 13.59% and 16.92%，respectively.The force on the teeth is different under different cutting angles，and the cutting angle of the teeth can be improved at 55°.It provides a reference for the design of cutter teeth structure of cantilever type roadheader.

The metal mine is an important industry type for China’s social-economic development in China.In recent years，mines have been constructing safety digitalization，and improving the level of mine safety management and control capacity is the key aspect of achieving enterprise transformation and high-quality development.The Internet of Things，cloud computing，roboticized equipment，and modern mining deve-lopment，safety production and other technologies are deeply integrated，gradually forming an intelligent security control system that integrates situational awareness，dynamic prediction，and intelligent warning.During this process，metal mining enterprises still face problems such as lack of safety management，unclear decision-making targets，and frequent safety accidents.Therefore，during the construction of metal mine safety digi-talization，a new evaluation index system for metal mine safety control capacity was proposed.The CWM method was used to comprehensively analyze the influencing factors of the safety control capacity of metal mining enterprises，and the CWM-TOPSIS method was used to construct an evaluation model for the safety control capacity of metal mines.The safety management and control capacity of five gold enterprises in Shandong Province was evaluated.AHP-TOSIS and EWM-TOPSIS model were used to verify the CWM-TOSIS model.The results show that the order of the most significant indicators affecting the safety management and control capacity of metal mining enterprises is emergency response ability （x₁₁） > safety technology level （x₁₀） > risk intelligent early warning ability （x₁₃），and Enterprise 1 has the highest safety control capacity，which is consistent with the actual situation.The model and method can be adapted to the safety control capacity evaluation of metal mines.

During deep tunnelling using drill-and-blast method，excavation damaged zone （EDZ） is inevitably induced in surrounding rocks due to the coupled impacts of blast loading and dynamic initial stress unloading and thus affect the structure stability.Therefore，it is very important to predict EDZ depth before roadways excavation.Relying on the field measurements of EDZ in several underground mines as the research object，300 data samples were collected.Four mainstream hyperparametric optimization algorithms，i.e.，genetic algorithm （GA），gray wolf optimization algorithm（GWO），particle swarm optimization algorithm（PSO），and salp swarm algorithm （SSA），were used to optimize the XGBoost algorithm and to construct four hybrid models for EDZ prediction.Comparative analysis of predictive model performance was conducted in terms of R²，RMSE，MAE and MAPE，along with a sensitivity analysis of the influencing parameters.Finally，the optimal PSO-XGBoost model was applied to a transportation roadway in an underground mine for engineering validation.The results show that the GA-XGBoost，GWO-XGBoost，PSO-XGBoost，and SSA-XGBoost models achieve the best predictive performance with swarm sizes of 90，70，60 and 100，respectively.Among them，the PSO-XGBoost model demonstrates the best predictive performance with correlation coefficients of 0.9244 and 0.8787 in the training and testing sets，respectively.Moreover，compared to bench models（XGBoost，RF，SVM and LightGBM），both the prediction accuracy and stability of the optimized models are improved.The tunnel diameter（TD） and rock mass geological strength index（GSI） have the most significant influence on the loosened zone thickness，along with a noticeable impact from the vertical principal stress.The application results of the optimized XGBoost model in practical engineering show that the error between the measured value and the predicted value is within 10%，indicating that the PO-XGBoost is of significance for engineering application.

In order to study the load transfer characteristics and the slippage process of anchorage segment of the pressure-type anchor cable in soil strata，the formula of shear stress and axial force distribution of the anchor cable under external load was derived based on theoretical analysis.In addition，the load transfer characteristics was analyzed.The influence degree of soil compressive modulus，anchor diameter，load and Poisson’s ratio on axial force and shear stress were analyzed in detail.The shear stress and axial force in the anchorage segment of the cable has the same distribution function and is all power functions.The shear stress and axial force are the largest near the bearing plate，but far from the bearing plate，the shear stress and axial force gradually decrease to zero.The magnitude and distribution range of the shear stress and axial force in the anchorage segment mainly depend on the anchorage force，and are related to the diameter，elastic modulus and Poisson ratio of anchor body，the compressibility modulus and Poisson ratio of soils，and friction angle of the interface between anchors and soils.In the form of the formula，if the anchorage segment is infinite，the anchoring force P can be infinite in theory.The ultimate anchoring force depends on the strength of the anchors around the bearing plate，the soil and the interface.The position of the maximum shear stress is only related to the soil properties，material parameters and radius of anchorage in anchorage segment of prestressed cable，not related to the magnitude of load.The diameter of anchors also has significant influence on the distribution of shear stress and axial force.With different loads，the distribution form，range and peak position of shear stress is the same，which is independent of load.Under the same conditions，the larger the compressive modulus of soil，the larger of shear stress，and the faster the shear stress decline trendation.The Poisson’s ratio has little effect on the shear stress.When the value of Poisson’s ratio is different，the mechanical property of the whole anchorage system under external load is basically unchanged，and also the value of shear stress.The range of action remains and the axial force is unchanged.In order to verify the correctness of the theoretical analysis results，the load transfer test results of the anchor cable were obtained by in-situ tests of clay and sandy soil.The in-situ test results show that the theoretical analysis results are in good agreement with the engineering results.The research results can provide a basis for analyzing the stress state and optimizing design of soil anchorage engineering.

In order to solve the problem of fuzzy uncertainty of index parameters between adjacent grades of rock quality evaluation， the theory of topology was introduced and the topological rock mass quality grading model based on optimal combination weight was proposed to evaluate the quality of rocks in underground mining of the phosphate mine in Kunyang No.2 mine.According to the geological characteristics of the mine，six influence indexes，namely，uniaxial saturated compressive strength of rock（R_c），rock quality index（RQD），joint spacing（J_d），structural surface condition（J_f），groundwater status（W） and geostress influence coefficient（Z） were selected，and each index was divided into five grades.Then，the subjective and objective weights of the indicators were determined by the improved hierarchical analysis and entropy weight methods，respectively，and the moment estimation method was introduced to optimize the combination of subjective and objective weights.Finally，the modified RMR method and Q-system method were applied to evaluate the quality of the rock mass，and the grading results were compared with the quality classification results of the rock mass based on the optimal combination weights.The results show that，among the five rock masses to be evaluated，the evaluation result for the upper ore body rock is Class Ⅱ，with a characteristic value of 2.57，indicating good rock mass quality and certain self-stabilizing ability.It is determined that the support method of anchor spraying and hanging net should be adopted.For the lower ore body，the characteristic value is 3.68，indicating that the rock mass quality is approaching Class Ⅳ，and the stability of the surrounding rock in the roadway is poor.Immediate initial spraying should be carried out after the excavation of the roadway to control the deformation of the surrounding rock，followed by the use of the anchor spraying and hanging net support method.There are differences in the grading results of the upper ore body N₃，and the grading results of the rest of the rock bodies by the proposed method are the same as those of the modified RMR method.The overall low evaluation results using the Q-system method are caused by the failure to take into account the role of geostress and the discontinuity in the grading between the grades.In addition，the model is able to determine the degree to which the rock mass quality grade leans towards the adjacent grade.The classification results are consistent with the geological exploration report of the Kunyang No.2 mine，and are more accurate than the modified RMR method and Q-system method，validating the reliability of the method proposed in this paper and providing technical support for underground mining.

Rare earth is a key raw material for the development of strategic emerging industries in major countries.Clarifying the evolution pattern and competitive characteristics of global rare earth trade is of great strategic significance for seizing the commanding heights of future industrial development.Based on the perspective of the whole rare earth industry chain，the trade data of seven rare earth products from 1999 to 2022 were selected，and the clustering，centrality and community structure characteristics of the global rare earth trade network were analyzed by complex network analysis method.Competition intensity index and interdependence index were introduced to explore the competition and cooperation relationship in rare earth trade.The research shows that the scale of global rare earth trade continues to expand，the trade network space coverage is wider，connectivity and accessibility are gradually enhanced，with typical “small world” network characteristics.In the rare earth trade，the trend of a few countries occupying most of the trade links is becoming more and more obvious，and the difference in trade resources between nodes is very large.In addition，the rare earth trade is significantly divided into communities，but the composition of members within each community is unstable.From the perspective of competition and cooperation，the average competition intensity of rare earth trade has a small cyclical change，and the competition relationship is mainly reflected in the countries （regions） with a large volume of rare earth trade，and the interdependence tends to be low，and the trend of diversification of trade partners is obvious.On the whole，the competition and cooperation relationship between China and other rare earth trade participating countries（regions） is characterized by “high competitiveness and low dependency”.

Mine filling technology is an important technical means for the construction of green mines，and full pipe transportation is a very important technology in the filling operation of underground metal mines.Full pipe transportation can minimize the contact area between filling slurry and air，reduce the impact on the filling pipeline，extend the service life of the filling pipeline，and improve the efficiency of mining filling operations.Aiming at the problem of the long distance between the newly discovered edge ore body and the filling station in Fankou lead-zinc mine and the high difficulty of transportation，the surface pipeline SL1 and underground pipeline L2-2 in the mine design plan were selected as the research objects to study the optimization plan of the filling pipeline in Fankou lead-zinc mine.Firstly，using theoretical formulas and based on the filling data of Fankou lead-zinc mine，the filling line and full pipe rate of SL1 pipeline and L2-2 pipeline were calculated when transporting graded tailings and fine tailings，respectively.The comparison was made using the optimal full pipe rate of 0.8 as the standard.The results show that both pipelines are in a state of under pipe when transporting graded tailings，and are in a state of over pipe when transporting fine tailings，which do not meet the optimal full pipe rate and need optimization.Secondly，through formula derivation and calculation，the ideal horizontal pipe diameter and the hydraulic slope after diameter change when transporting different slurry were obtained.Finally，numerical simulation was used to verify the calculation results of pipe diameter optimization.A pipeline model was constructed using CFD.The vertical pipeline was taken as 5 m，the horizontal pipeline was 23 m，the total length of the pipeline was 28 m，and the curvature radius at the bend of the pipeline was 0.55 m.The horizontal pipe diameter was changed.Fluent software was used to simulate the full pipe transportation before and after the diameter change，and key data such as flow velocity and full process resistance were obtained when transporting graded tailings and fine tailings.By comparing and analyzing the pressure of the pipeline and the maximum outlet flow rate，it is concluded that SL1 and L2-2 can transport graded tailings by gravity after optimizing the pipe diameter，while fine tailings can’t be transported by gravity.However，the pumping pressure is significantly reduced，so the calculation results are reasonable.Therefore，this optimization plan is relatively reasonable and has strong guiding significance for mining filling operations.

In order to grasp the action law of process factors affecting the actual recovery rate of a gold ore and predict the gold recovery rate，the flotation test was carried out by the method of orthogonal experiment.The sensitivity of process factors was analyzed by Pearson coefficient，and the gold recovery rate was predicted by using double hidden layer BP neural network.The results show that the sensitivity of the gold recovery rate to different factors in the production process is in descending order：2^# oil dosage，sodium sulfide dosage，butyl xanthate dosage，copper sulfate dosage and grinding fineness.The reasons for the significant differences in the effects of 2^# oil dosage，sodium sulfide dosage and butyl xanthate dosage on gold recovery rate were also elucidated.On this basis，used three main influencing factors such as 2^# oil dosage，the study selected different input layer to the first implicit layer functions，such as tansig，purelin and logsig，and the rest of the activation functions remained unchanged.The first hidden layer to the second hidden layer function was logsig，and the second hidden layer to the output layer function was purelin.When research used logsig as the activation function，the fitted degree and accuracy are high，the goodness of fit R² is 0.9792，and the relative average error is only 0.666%.The model can be used to predict the recovery rate of gold.The research has certain reference significance for the prediction of metal recovery rate in the production of precious metal mines.

Chute is a common form of deposit development.With the increasing exploitation of alpine and deep resources，high-deep chutes are widely used because of its advantages of simple development methods，convenient transportation and low operating costs.Based on the long-term observation of the production practice of single-segment vertical high-deep chutes，we summarized the flow law of ore in high-deep chute，and put forward that the vertical full-section stage flow is a unique engineering phenomenon of high-deep chute which is different from short chute.Through analysis and research，it is found that there are limitations in the study of ore flow law in high-deep chute by numerical simulation and laboratory model test.Therefore，we adopted the research method of ‘practical understanding+theoretical analysis’，and based on the research carried out by Sun Qicheng et al. of Tsinghua University，the cutting-edge results of the theory of riding chain and arching effect was used to explain the arching mechanism of ore granular in the chute.At the same time，combined with the analysis of gas-solid two-phase flow in the wellbore，the ore flow law of high-deep chute was studied.The causes of the staged flow characteristics of high-deep chute were analyzed，and the influence of the staged flow of high-deep chute on the common stagnation，blockage，wear and spalling phenomena in the production process of chute and the mixing and classification of ore in the chute was revealed.The influence of air flow in the chute on the granular flow was analyzed.On the one hand，the existence of flowing gas in the normal ore drawing process is beneficial to the ore flow and the destruction of the temporary balance arch.On the other hand，the drag force of gas produces the phenomenon of block and powder classification concentration，which promotes the stagnation and blockage of ore in the wellbore.It is proposed that the grain composition，density，water content，cohesiveness of the ore，the diameter of the chute and the roughness of the shaft wall are the basic factors causing the stagnation and blockage of the chute.The flow characteristics of the vertical full-section stage ore are the important reasons for the stagnation and blockage of the high-deep chute.The results open up ideas for the research and the safe and efficient production of high-deep chutes，and can provide reference for the design and use of high-deep chutes.

After mining operations，a large number of broken ore pillars will be left behind，forming a group of ore pillars.The group of ore pillars is of great significance for safe underground operations and the recovery of ore pillars resources.Under the influence of high ground stress，geological weak surface structure，free face generation，blasting vibration and other factors，the ore pillars group is very prone to instability，collapse，resulting in disaster of a large area of instability ore pillars group，thus causing great harm to underground workers and equipment.In order to strengthen the safety control of the mine pillars group，the stability analysis was carried out based on the renormalization group theory and Bayesian network.The renormalization group is a scaling transformation group.The self-similarity transformation of the basic unit can be used to describe the whole system，which is similar to the process in which the instability of a single pillar ultimately leads to the instability of the pillars group.Bayesian network can also overcome the uncertainty in the process of the instability of the pillars group.Therefore，based on the renormalization group theory and Bayesian network，the stability analysis model of pillars group was established.A pillars group in Tongkeng mine was taking as the research object.Firstly，the critical instability probability of the pillars group was derived based on the two-dimensional renormalization group model.Secondly，the multiple logistic regression model was used to deduce the calculation formula of the instability probability of a single pillar，then the stability，instability and instability probability of the pillar was calculated，and the stability probability of a single pillar was input into the Bayesian network to obtain the instability probability of the pillars group.Finally，by comparing with the critical instability probability of the pillars group obtained from the two-dimensional renormalization group model，the stability group of the pillars group was obtained.The research results show that this method can more accurately obtain the instability probability of the ore pillars group，and by defining the stability probability of a single ore pillar to carry out Bayesian network probability inversion and sensitivity analysis on the ore pillars，can design the optimal mining route of the ore pillars，and achieve safe，efficient and low loss recovery of pillars resources in the panel.This method can also provide reference for stability analysis and pillars resource recovery of similar mine pillar groups.

The mining of open-pit mines destroys a large amount of land resources and causes a series of ecological and environmental problems such as vegetation damage and landslides.Vegetation restoration is the preferred method to solve the ecological environment and natural disasters of open-pit mine dump slopes.Aiming at the tensile properties of the interface between herbaceous plant roots and soil，alfalfa was taken as the research object，and the reinforcement effect of herbaceous plant roots on the shallow soil was used to improve the stability of open-pit mine dump slope. The effects of root diameter（0~2 mm），soil compaction（65%~95%） and root depth （8~15 cm） on the pull-out resistance of root-soil interface were analyzed by carrying out indoor single-root tensile and pull-out tests，and the mechanical properties and failure mechanism of alfalfa roots under pull-out were explored.The results show that：（1） The tensile and pull-out force of alfalfa increase with the increase of root diameter，while the tensile and pull-out strength decrease with the increase of root diameter.There is an obvious non-linear relationship between pull-out strength and root diameter，and the correlation coefficient is more than 0.9.The pull-out failure occurr when the pull-out failure reach the critical value of 0.73 mm.（2）With the increase of the degree of compaction，the pull-out resistance of alfalfa shows a trend of first increasing and then smoothly，indicating that after the peak of pull-out resistance，the friction force at the root-soil interface is greater than the tensile force of the root system，and the root system is broken.The greater the degree of compaction，the higher the probability of tensile failure.（3）The influence of buried depth on the pull-out resistance of alfalfa roots is related to the degree of compaction and root diameter.The greater the degree of compaction，the smaller the influence of buried depth on the pull-out resistance，and the greater the root diameter，the greater the influence of buried depth on the pull-out resistance.In Summary，root diameter，soil compaction and root burial depth are the main factors affecting the pull-out resistance of the root-soil interface.By studying the influence of root diameter，soil compaction and root buried depth on the pull-out resistance of root-soil interface can improve the stability of open-pit dump mine slope.

At present，traditional evaluations of green mine construction mostly use a direct scoring method to evaluate the effectiveness of green mine construction.This method ignores the relationship between evaluation indexes and fails to fully consider the input-output efficiency of mining enterprises.This paper selected 24 specific indicators from four aspects of resource utilization，environmental protection，enterprise society，and technological economy to build an evaluation index system for the effectiveness of green mine construction in the gold industry.In addition，the input-output efficiency evaluation index system of green mine construction in gold industry was constructed by optimizing and screening efficiency index from input-output aspect.The interval analytic hierarchy process（IAHP） and entropy weight method（EWM） were used to obtain the subjective and objective combined weights of evaluation indicators，and the TOPSIS method was combined to build the green mine construction effect evaluation model for the gold industry.The input-output efficiency evaluation of green mine construction in the gold industry was analyzed by using the BCC model of DEA method.On the basis of effect and efficiency measurement，the efficiency-effect evaluation matrix was designed to facilitate the analysis of the degree of green mine construction and the benefits of gold mining enterprises.It helps to judge whether gold mining enterprises take efficiency into account during the process of green mine construction，and the results reflect the actual overall construction situation of gold mining enterprises.Through the case analysis of ten gold mining enterprises of the same type in Shaanxi Province，the results show that the effectiveness and efficiency of the ten gold mining enterprises are fluctuating and rising.The overall construction situation is good，and the input-output efficiency is superior to the construction’s effect.Among the ten gold mining enterprises，except for M2 and M9，the green mine construction of the other mining enterprises has reached a high status.The four mining enterprises，M1，M4，M5 and M6，have given consideration to the effect and efficiency，and the other mining enterprises should pay attention to the development of both in the future construction.The index system constructed in this paper considers the characteristics of green mine construction in the gold industry，and the economic and quantitative indicators improve the accuracy of the evaluation results.The effect-efficiency evaluation model strengthens the connection between indicators，and can evaluate and analyze the overall situation of green mine construction in gold mining enterprises.It provides a scientific basis for the improvement of gold mining enterprises.

Aiming at the unfavorable blasting effects of the open-air step stripping blasting，such as large blocks and many root bottoms，the blasting effect of the open-air step was optimized by adopting the two-stage charge structure of water medium interval at the bottom of the blast hole and the middle part.To obtain the optimal two-stage charge structure of water medium interval in step stripping blasting of the open-pit mine，the blast hole pressure of the air-medium and water-medium interval charging were discussed from theoretical analysis，the unit consumption of water medium interval charge structure in Wujiata open-pit mine was optimized by experiments.On the basis of the selected optimal unit consumption，three groups of three-dimensional models of different water medium interval charge structure were established by using ANSYS/LS-DYNA numerical simu-lation software.The optimal water medium interval charge structure was selected by analyzing the stress change curve of rock mass，rock mass damage range，and free surface tensile crack area ratio at the bottom of the blast hole，the middle of the lower charge，and the upper position of the blast hole.The blasting effect was evaluated from the index of boulder yield by the blasting comparative experimental with the charge structure of air medium interval.The research results show that blasting with a charge structure of a water medium interval can significantly increase the blasting pressure and improve the rock-crushing effect.The optimal unit consumption of water medium interval charge structure blasting in Wujiata open-pit mine is 0.33~0.34 kg/m³，saving 0.03~0.04 kg/m³ unit consumption of explosives than before. The blasting of a charge structure with an upper water medium interval of 1 m and the lower water medium interval of 1 m has more stress on the rock mass and has a long duration，the damage is evenly distributed along the blast hole，the range is regular，and the free surface tensile crack area accounts for the highest proportion，and its blasting effect optimal.The blasting fragmentation indexes of the optimal water medium interval charge structure all show a good proportion.The research results can provide a reference for the application of water medium interval charge structure in open-pit step blasting.

There are few rugged plains in southwest China，and tunnel construction plays an important role in southwest China.However，when the tunnel is excavated by blasting in grade Ⅳ and Ⅴ weak surrounding rock，the traditional continuous charge smooth blasting technology often has problems such as over-excavation，low footage rate and large fragmentation after blasting.In order to solve such problems，based on the principle and application status of shaped water pressure smooth blasting technology，an axial porous shaped charge tube was designed.The integrated shaped water pressure structure was formed by alternately arranging the water bag and cartridge in the tube with detonating cord and digital electronic detonator.Four kinds of tests were carried out in the Ⅳ surrounding rock area of Mengsheng tunnel in Yunnan Province，including continuous charge smooth blasting，water bag interval smooth blasting，air interval smooth blasting and shaped water pressure smooth blasting.In order to maintain comparability，15 holes at the vault position were selected for the four tests.The blasting effect was checked one hour after blasting，and the blasting technical indexes such as single cycle footage，half-hole rate and overbreak and underbreak were measured by using range finder，tape and benchmark.The test results show that the half-hole rates of continuous charge smooth blasting，water bag interval smooth blasting and air interval smooth blasting are 13%，45% and 38% respectively，and the footage of one cycle is 3.40 m，3.50 m and 3.45 m respectively.The surrounding rock of the contour surface is broken and there is overbreak and underbreak phenomenon，while the half-hole rate of the shaped water pressure smooth blasting with integrated charge structure is 62%.The contour of a test area with a cycle footage of 3.60 m is relatively flat，there is no obvious over-excavation phenomenon，and the gravel after blasting is small.The comparison between the control test groups verifies the superiority of the energy-gathering structure based on plastic pipe research and development.The field application shows that the new shaped water pressure blasting has a significant effect on the flatness control of the surrounding contour of the weak surrounding rock tunnel.While reducing the number of boreholes and the amount of explosives，it maintains the integrity of the surrounding rock.It is an environmentally friendly and energy-saving controlled blasting technology.

With the booming development of big data and Internet of Things technology，traditional mines have developed to smart mines and intelligent mines，and unmanned technology has been gradually applied to mining areas.In order to solve the problem that the rockfall detection of unstructured road in open-pit mine area has complex environment，large difference in rockfall size and similar color between rockfall and unstructured road surface，a rockfall detection model of mining road based on weighted bidirectional feature fusion was proposed.First，the SimAM attention mechanism is added to the backbone network，this attention mechanism is different from the previous channel attention mechanism and spatial attention mechanism，it can effectively eliminate the interference of the background environment without adding additional parameters，so that the model can focus more on the target characteristics of rockfall.Second，the weighted bidirectional feature pyramid（BiFPN）structure was used to realize multi-scale feature fusion in the neck.Since the PANet structure in the YOLOv5s network model only adds or splice the characteristics of the pyramid structure in the melting process，the bidirectional feature weighting was combined with the bidirectional feature of the weight and adaptive adjustment to ensure that the network model attaches proper importance to the rock ebaissees different sizes and different levels and realizes the addition between the low-level position information and high-level semantic information for multiple cross-layer weighted feature fusion，thus enhancing the feature extracion ability of the model for rockfall of different sizes.Finally，the lightweight convolution GSConv module was introduced into the col，which can be used to process function cards at this time，not only reducing redundant information，but also avoiding compression.The GSConv lightweight convolution module is based on deep separable convolution（DSC），ordinary convolution（SC） and channel shuffle operation，which improves the detection speed of the model by effectively reducing the complexity of the model.The experimental results show that the average detection accuracy of this algorithm reaches 92.8%.and the detection speed reaches 63.1FPS.Compared with the current fastest R-CNN，YOLOv4-tiny，YOLOv7 and YOLOv5s algorithms，the average detection accuracy is increased by 17.0，13.6，3.4 and 2.5 percentiles，and the detection speed is increased by 32.2，1.4，14.6 and 2.6 FPS，respectively.Moreover，the model size of the algorithm is only 12.9 MB，which is easy to deploy on mobile devices.Therefore，the algorithm can realize the real-time and accurate detection of unstructured road rockfall in mining area，and ensure the safe driving of unmanned mining card.

The “cloud-edge-terminal” system of smart mines under the mode of industrial internet of things（IIoT） has become the focus of development of mining enterprises.At present，the digitalization and intelligence at the “cloud” and “end” levels have began to take shape.However，the construction at the “edge” level，namely the edge layer，is relatively weak，which restricts the production efficiency of smart mines.Therefore，the concept of “mine-edge intelligence” were proposed，which was defined as an edge intelligence technology system specifically designed for mining production environments，with multi-dimensional characteristics in geography，network and technology.The key technologies include data fusion，information communication and edge security.Data fusion not only integrates data from different sources，but also solves the issue of equipment compatibility.The equipment in the mining environment comes from various vendors，utilizing different communication protocols and data formats.It is very important to establish a data fusion platform that can handle heterogeneous data and ensure compatibility among devices.This necessitates the conversion of data formats，so that the data in the system circulation，coupled with data cleaning，processing，and analysis.The information and communication capacities focus on establishing an efficient and reliable communication network in the mining environment.This includes the adoption of advanced communication technologies and network architectures to optimize network configuration and management，and ensure rapid and secure data transmission between the “cloud”，“edge” and “end”.The edge security technology focuses on protecting edge layer devices and data，including encryption，authentication，and access control methods to enhance security.Moreover，it is necessary to emphasize the deep integration of mining technology with intelligent technology.This includes process reshaping driven by software requirements，as well as innovations in process decision-making，network architecture and environmental sensing.In summary，as the core of the smart mine’s edge layer，“mine-edge intelligence”needs to strengthen data fusion，information communication and edge security technology.Future development should focus on the deep integration of mining technology and intelligent technology to enhance production efficiency and safety.

Promoting the digital transformation of resource-based enterprises is an effective path to crack the development dilemma of enterprises and achieve high-quality development.Due to the significant differences in resource-based enterprises’ resource endowments and transformation capabilities，the degree of digital transformation also shows significant differences.So，what combination of influencing factors can effectively promote the digital transformation of resource-based enterprises？What are the differentiated transformation paths that different resource-based enterprises should follow？At the same time，the digital transformation of resource-based enterprises is a complex systemic project，and previous studies are limited in exploring it only from a single dimension.This paper took resource-based enterprises in China as the research object，from the perspective of “pressure-capacity-drive”，it refined into six antecedent conditions：market competition，growth capability，resource integration capability，innovation capability，enterprise size and financial support.Using NCA and fsQCA methods，it analyzed the relationship between the six antecedent conditions and the NCA and fsQCA methods were used to analyze the complex causal relationships between the six antecedent conditions and the digital transformation of resource-based enterprises.The study found that：The digital transformation of resource-based enterprises is not driven by a single condition，but it is the result of the synergy of multiple conditions.There are three types of groupings that generate high-level digital transformation，namely，the innovation capability-driven type that responds to pressure，the government-led type under pressure and the comprehensive-driven type，mainly by SMEs.There are two types of groupings that generate non-high-level digital transformation，namely，the settled type and the type with both weak innovation capability and drive.And market competition plays a key role in the digital transformation of resource-based enterprises.The results of the study reveal the interaction between various factors in the digital transformation of different resource-based enterprises，which can provide theoretical guidance and path references for resource-based enterprises to realise digital transformation.