The prediction of goaf stability is a critical component of mine safety management.Accurate forecasting of goaf stability during mining operations is essential for ensuring both the safety and environmental integrity of mining activities.Due to the intricate nonlinear relationship between goaf stability and its influencing factors，traditional prediction methods often fall short in delivering precise results.Machine learning，however，is well-equipped to address this issue.To overcome the challenges of algorithms becoming trapped in local optima and exhibiting suboptimal convergence speed and accuracy，a stability prediction model for goaf，based on the SIDBO-BP algorithm，was proposed to effectively classify and predict the stability of mine goafs.In light of the internal and external factors influencing goaf stability，nine specific factors were identified based on the unique conditions of the mine：Burial depth （X1），exposed roof area （X2），high collapse ratio （X3），goaf volume （X4），goaf dip angle （X5），conditions of adjacent goafs （X6），rock compressive strength （X7），geological structure （X8），and rock mass structure （X9）.These factors were utilized to classify the stability status of four goafs，which served as output levels in the construction of a goaf stability prediction model. The study utilized ninety-three datasets from a lead-zinc mine in Yunnan as the research object.Correlation and visualization analyses of the goaf data were performed using graphical correlation analysis and correlation coefficient graph methods，thereby validating the appropriateness of the selected factors.The goaf data were normalized using MATLAB.Subsequently，the SIDBO algorithm was applied for optimization，yielding the optimal position and fitness，which were then employed in a BP neural network to determine the optimal threshold weight for prediction.The macro-average method was employed to compute the metrics of accuracy，precision，specificity，recall，and F1 score for the predictive outcomes.The model’s predictions were assessed and benchmarked against those generated by other algorithmic models.The findings revealed that only a single prediction from the SIDBO-BP model was inaccurate.The SIDBO-BP model demonstrated superior accuracy，precision，specificity，recall，and F1 score compared to the other algorithmic models，aligning closely with the actual results.Additionally，its overall performance exceeded that of the other five models.These results suggest that the SIDBO-BP model provides substantial advantages in terms of prediction accuracy，convergence speed，and the avoidance of local optima.

Underground metal mining operations are complex，involving dynamic events，parallel processes，and strict time-space constraints.These operations require the coordination of multiple independent mining equipment，，and process entities to ensure efficient ore extraction.Accurately modeling these operations and replicating the scheduling process iscrucial for optimization.Converting the physicalmining processes into a virtual environment is vital for effective simulation and modeling.This paper introduces a Petri net-based method for modeling and simulation，creating a framework that maps underground mining systems to a simulation space，facilitating dynamic scheduling of mining equipment.The construction of the mining system employs a five-layer architecture for hierarchical planning，establishing a structured framework for the mining operation system.This framework encompasses elements such as thematic domain grouping，thematic domains，business objects，logical entities，and attributes.Fundamentally，the mining operation system is characterized by the dynamic interaction of multiple entity objects across three primary thematic domains：Mining operation space，mining operation equipment，and mining operation processes.Utilizing Petri nets，the interrelationships between theoverall system and its constituent parts，as well as the interactions among components within localized areas，arethoroughly delineated.This approach facilitates the demonstration of event sequences，parallelism，synchronization，and asynchronous characteristics during the mining operation process.It also highlights conflicts，mutual exclusion，uncertainties，and potential deadlock issues within the system，thereby enabling a comprehensive analysis and evaluation of the system.The Generalized Stochastic Petri Net（GSPN） methodology is utilized to model transitions as random variables governed by mathematical distributions，thereby addressing the intricate coupling issues among space，equipment，and processes within the mining operation system.This approach facilitates the assessment of the system’s performance and reliability，yielding key production evaluation metrics such as production efficiency，equipment utilization，and effective working time.A practical simulation，based on an idealized data set constructed in accordance with the standard design stope of the Chambishi copper mine，serves to validate the mining operation system.The simulation results indicate that the total operation progress completion time is 213.77 days，during which 656 mining operation records，are generated.Additionally，a mining operation Gantt chart is constructed to illustrate the operation cycle and strip succession.The findings of this research provide robust support for dynamic scheduling decisions in underground metal mining operations，contributing to the achievement of efficient，safe，and intelligent mine production management.

The extraction of mineral resources through techniques such as the room-and-pillar method in underground metal mines results in the formation of numerous goafs，which can readily lead to pillar deformation and roof collapse，thereby creating significant safety hazards.Furthermore，the collapse of goafs can induce mining-related seismic events，water inrush，and mud inrush，posing substantial risks to both personnel and equipment.Conducting precise risk assessments of goaf stability is fundamental for comprehending and managing these risks，which is essential for effective disaster prevention and mitigation.Consequently，performing precise stability risk assessments for goaves is crucial for effective risk management.Addressing the complexities posed by numerous influencing factors and the challenges associated with quantitative assessment in goaf stability analysis，this study introduces a fuzzy comprehensive evaluation approach that integrates the Decision-Making Trial and Evaluation Laboratory （DEMATEL） and Interpretive Structural Modeling （ISM） methodologies.Initially，the factors influencing goaf stability were identified through a comprehensive approach involving field investigations，literature review，and consultations with experts.An initial screening process identified fifteen risk factors associated with rock mass structure，goaf morphology，and environmental impact.Subsequently，twelve critical risk factors were determined，encompassing geological structure，rock mass properties，in-situ stress，groundwater conditions，goaf volume，shape，span，temporal effects，effects of adjacent goafs，mining disturbances，support conditions，and natural disasters.This led to the development of a systematic framework for evaluating goaf stability.Utilizing DEMATEL and ISM analysis，this study investigated the intricate interrelationships among influencing factors to develop a hierarchical framework for goaf stability determinants.An oriented hierarchical structure diagram was constructed to facilitate the classification and weighting of these factors，enabling the calculation of each factor’s significance within the evaluation system.Subsequently，these weights were integrated into a fuzzy comprehensive evaluation system，culminating in the formulation of an innovative model for assessing and grading goaf stability.This model provides management guidelines from an intrinsic safety standpoint，with the objective of fundamentally mitigating the risk of accidents. Through a case study of the 36-2 goaf at the 930-meter level of a lead-zinc mine，the primary influencing factors and direct causes of goaf instability were identified，and the complexity of goaf incidents was thoroughly analyzed.Consequently，the goaf received a stability evaluation rating of “good”. In conclusion，the stability of the goaf was analyzed using the FLAC^3D numerical simulation software.A mine-scale FLAC^3D model was developed to assess the differential stress distribution and failure mechanisms of the rock mass above the goaf，thereby validating the accuracy of the goaf stability classification model.The findings indicate that the goaf stability evaluation model introduced in this study demonstrates high applicability and accuracy.This method mitigates subjectivity in the weight determination process inherent in traditional fuzzy comprehensive evaluation techniques，while simultaneously accounting for a comprehensive range of factors influencing goaf stability.Consequently，this research offers a novel approach to evaluating goaf stability，which may serve as a valuable reference for the stability assessment of goafs in metal mining contexts.

To address the challenges associated with the probability distribution and weight allocation processes of evaluation indicators for the stability of artificial pillars （backfill） using the segmented empty-field subsequent filling method，an evaluation indicator system was developed.This system considers the influences of five categories of information sources：Engineering geological environment，rock mass quality，mechanical response indicators，acoustic emission parameter characteristics，and geometric shapes.The resulting evaluation framework encompasses 17 factors that influence stability.Based on this analysis，the recognition framework and evaluation index grading standards for the stability of artificial pillars （backfill） were established utilizing Dempster-Shafer （D-S） evidence theory.By incorporating the normal membership function and an enhanced Kullback-Leibler （K-L） distance，the probability distribution and weight allocation of indicator factors were optimized.Consequently，an improved multi-source information fusion model，grounded in D-S evidence theory，was developed for the assessment of artificial pillar （backfill） stability.The applicability of the model was verified through five typical artificial pillar（backfill） in a mine with segmented open stoping subsequent filling mining.The results indicate that the stability evaluation levels for multi-source information fusion were ranked as follows：AP-2=AP-3=AP-5（Level Ⅱ）>AP-4（Level Ⅲ）>AP-1（Level Ⅳ）.These stability evaluation out-comes are largely consistent with previous assessments based on displacement，acoustic emission on-site monitoring，and numerical simulation analysis.This congruence validates the scientific efficacy of the evaluation model in practical applications.Consequently，this model offers a more comprehensive approach for analyzing the stability of artificial pillars （backfill）.

Following the instability incident in the upper middle section of the filling stope at the West No.2 mining area of Longshou mine in Jinchuan，the original downward-layered consolidated filling method was replaced with the non-pillar sublevel caving method.To investigate critical technical challenges，such as cover layer formation and stope stability associated with the practical application of this method，numerical simulation techniques were employed for the research.The research findings suggest that as the sublevel stope area within a caving stope expands，the composite roof experiences a sequence of progressive failure stages.These stages include the initiation of failure cracks，dispersed bulk caving，arch batch caving，and plunger integral caving.By the conclusion of the first sublevel mining，the height of the roof collapse is expected to exceed 30 meters，thereby forming a sufficiently thick cover layer for the caving method stope.Concurrently，during the mining process，plastic zones are generated on the surface of the roof along the mining approach.The support structure serves a critical function in anchoring and stabilizing the majority of the plastic zones，thereby contributing to the overall stability of the mining approach.Despite the minimal displacement observed in the access roof，there is a pronounced stress concentration within 5 meters behind the working face，necessitating continued attention.An industrial experiment was subsequently conducted on-site，employing the non-pillar sublevel caving method and utilizing induced caving technology to establish a cover layer.To mitigate the risk of large-scale collapse of the composite roof，a technical strategy involving“stepped mining and uniform expansion of goaf combined with total ore extraction control”was implemented on-site.Microseismic monitoring and on-site tracking data revealed that during the initial sublevel mining phase，the actual caving height of the roof in the mining area surpassed 20 meters，resulting in the formation of a thick cover layer approximately 30 meters thick，inclusive of the reserved ore layer.Throughout the entire mining process，the stope remained in a stable condition.To address the issue of significant damage to the access road in the fractured ore and rock zone，a comprehensive technical scheme incorporating multiple support structures was proposed.For the problem of medium and deep hole damage，a systematic approach involving hole inspection，hole dredging，and hole filling was established.Additionally，relevant equipment was promptly introduced to mitigate operational intensity and enhance the applicability of the non-pillar sublevel caving method under the soft，fractured ore and rock conditions prevalent in the Jinchuan mining area.

In the extraction of overlying orebodies in complex environments，it is imperative to prioritize safety and efficiency while preserving the stability of rock strata to prevent surface collapse and safeguard structures.To accomplish these objectives，it is essential to develop effective strategies for managing rock strata movement by comprehensively assessing the impact of underground backfill mining.An initial study was conducted on the impact of various mining field structural parameters，including roof thickness，original rock stress，and backfill ratio，on overlying rock movement at the Panlong lead-zinc mine.This analysis utilized a nonlinear elastic foundation beam mechanical model.Subsequent numerical simulations were performed to determine overlying rock movement values under varying backfill ratios.The results were compared to those obtained from the mechanical model，revealing differences ranging from 3% to 9%，thus validating the reliability of the mechanical model’s results.A ground mine backfill mining technology for controlling settlement of overlying rock was proposed and implemented in industrial experiments.Research findings suggest that the primary factors influencing overlying rock movement，in order of significance，are the backfill ratio，structural parameters of the mining field，roof thickness，and original rock stress.Increasing the backfill ratio is crucial for managing overlying rock displacement，while adjusting anchor support spacing and backfill ratio according to the mining field structure are effective methods for controlling overlying rock displacement.

The block caving method is known to result in a specific range of surface subsidence，which is a major safety risk in mining operations.UAV aerial survey technology offers numerous advantages，including enhanced safety，simple operation process and high precision in monitoring large and complex surface areas. Therefore，this study concentrates on the largest block caving mine in China，employing UAV aerial survey techniques to remotely monitor surface subsidence caused by block caving mining methods.The development of UAV aerial survey technology has led to the proposal of a 3D modeling method for surface subsidence based on UAV aerial survey.An accuracy analysis method for the 3D modeling process was also presented.By cons-tructing 3D models，analyzing the extent of surface subsidence affected areas，examining profile data of subsidence affected areas，and conducting theoretical analysis，this study investigates changes in the subsidence area of the mine from March 2022 to March 2023. Based on an analysis of the subsidence range and orebody profile within the vein，as well as predictions of surface subsidence using the probability integral method （PIM），it can be inferred that the extent of active subsidence is influenced by the distribution of active drawpoints. Furthermore，by utilizing PIM to forecast subsidence values，a comparison between the calculated and actual subsidence amounts at the Pulang copper mine，the maximum error is less than 0.18 m. The prediction of surface subsidence patterns in this mine can be achieved through the application of the probability integral method，which is grounded in the principles of stochastic medium theory. Utilizing the PIM allows for the anticipation of surface subsidence regularity，thereby facilitating effective surface treatment and mitigation of debris flow within the mine. By integrating 3D modeling with surface subsidence patterns，various recommendations can be formulated for the management of subsidence pits and the development of a drawing plan for the mine.

Based on the limitation of mine wastewater treatment process and the large demand for water resources in phosphogypsum backfill process，the feasibility of using sulfate-containing mine wastewater as phosphogypsum backfill water was discussed.Results show that the sulfate present in the wastewater promoted the early hydration of the phosphogypsum-based cemented paste backfill，consequently enhancing its early strength.Specifically，as the sulfate concentration increased from 0 to 30 000 mg/L，the 7-day strength of the backfill samples increased from 0.46 MPa to 0.63 MPa，representing a 37% increase.The 14-day strength also exhibited a significant increase from 0.64 MPa to 1.03 MPa，indicating a 61% increase.This can mainly be attributed to the formation of hydration products such as ettringite and C-S-H.However，with the increase of the curing time，the 28-day，60-day，and 90-day strength of the backfill samples reach the highest values at a sulfate concentration of 15 000 mg/L.Excess sulfate ions can interfere with the hydration and hardening processes，and the long-term strength of the backfill is affected by the expansion caused by AFt and the recrystallization of Na₂SO₄ and other influences，resulting in a decrease in the strength of backfill samples.The results can provide a new insight into mine wastewater treatment and phosphogypsum-based cemented paste backfill technology.

The national sustainable development strategy attaches great importance to ecological and environmental protection issues，and the construction of green mines has become an inevitable trend in the development of mining industry.Solid waste filling is an effective measure to achieve green mines.In order to achieve the recycling of solid waste and reduce the solid waste storage of tailings and nickel slag，the mixed aggregate ratio test of nickel slag-full tailings was carried out to provide technical support for the industrial verification of nickel slag full tailings full solid waste filling.In order to study the optimal mix ratio of nickel slag-tailings paste slurry，the particle size distribution and chemical composition of nickel slag-tailings were analyzed by scanning electron microscopy （SEM），X-ray diffraction （XRF） and other test methods，and the optimal nickel slag-tailings ratio range of 4∶6，5∶5，6∶4 was determined by the aggregate packing density theory.Considering the fluidity，workability and backfill strength of slurry，the optimal mix ratio of nickel slag-total tailings is finally recommended as 5∶5，the mass concentration of slurry is 79%~81%，and the reasonable cement content is 310 kg/m3.For nickel slag-total tailings pumping filling in Jinchuan mine，the maximum utilization of nickel slags is not more than 50%，the maximum mass concentration of nickel slags can be pumped and filled is 81%，the corresponding shear stress is 112.634 Pa，and the viscosity is 0.938 Pa·s.

In order to make the residual ore resources of Yunxi Old Factory Branch get better recycling，in view of the situation that there are a large number of residual pillars in the panel area of Gaofengshan ore section，after the completion of the on-site engineering geological investigation，representative rock samples were selected to conduct the rock mechanics test，and the result parameters were obtained.FLAC^3D numerical simulation method was used to analyze the stability of the stope before the recovery of the pillars，and the law of ground pressure appearance in each panel area was obtained.Combined with the actual situation of the mine，it is concluded that the two methods of extracting the mining pillar and replacing the pillar are suitable for the recovery of the pillar in the mining area.On this basis，six pillar recovery schemes were considered under ideal conditions，and after comparative analysis，only the two-step collodion-filled room-and-pillar mining method is suitable for recovering the security ore in this mine.Through the filling mechanical parameter experiments，the filling body parameters required for two-step spaced back mining were derived after comparing multiple schemes.At the same time，the pillar recovery process was discussed in detail，and the industrial test of pillar recovery was carried out in areas with different degrees of geopressure manifestation.From the experimental results，the top plate of the pillar can be kept stable during and after the pillar recovery process.Based on the actual situation，the 13 pillars in the mining section were analyzed in terms of the resource grade of the pillars，the stability of the surrounding empty area，the need for protection works above the pillars，the economic benefits of mining and the price break-even.After the study，it is concluded that through the results of industrial test，the two-step cemented filling mining method with the specifications of 4 m×4 m can be used for the pillar mining，with the strength of one-step cemented filling is greater than 3.52 MPa and the strength of two-step filling is greater than 1.08 MPa.There are four pillars in the section with relatively intact，large amount of retained resources and high grade，which are in good condition for mining.The 4# pillar panel in the south plate area should be mined when the market metal price of tin concentrate is more than 244 800 yuan/t.The remaining 3 ore pillars can be mined when the market metal price of tin concentrate is more than 100 500 yuan/t.

In recent years，with the decrease of land mineral resources，the development of marine minerals has become a global emerging industry，especially the mining of coastal bedrock deposits under the sea，has been the focus of mining development in various countries.For large-scale mining under the sea，the movement and deformation of the submarine rockbody poses a major threat to mining safety.It is important to pay attention to the deformation damage of the surrounding rock，which is important to realize the safe and efficient production of submarine mines. In order to study the deformation and damage characteristics of the surrounding rocks of the underground mine body in the Xinli mining area，the 55 exploration line perpendicular to the mine body was used as the monitoring profile，and the roadway roofs of -200 m，-240 m，-320 m，-400 m，-480 m，and -600 m sublevels were monitored by the underground four-level monitoring system that has been deployed since December 2015.The vertical displacements of the surrounding rocks in sublevels of the roadway at -200 m，-240 m，-320 m，-400 m，-480 m and -600 m were monitored for a long time，and the time series data of the settlement of the roadway roof in each sublevel were obtained.The deformation of the surrounding rock caused by metal ore mining is a complex mechanical problem.By analyzing the multi-year monitoring results，several significant characteristics of underground settlement are revealed.（1）The deformation caused by mining of the orebody in sublevel at different depths of the seabed shows a large influence range on the rock mass of the fault hanging wall，while the influence range on the footwall rock mass is small.（2）The closer to the orebody（or the ore-control fault F₁） the greater the subsidence of the rock body，forming the feature that the slope of the curve to the left of the maximum settlement point in the settlement curve is larger in absolute value，while the slope of the curve to the right is slightly smaller in absolute value.（3）The final subsidence curves of the six sublevels in the line 55 profile have similar shapes and are generally asymmetric funnel-shaped，with the -200 m sublevel and the -240 m sublevel having a gentle bottom of the cumulative subsidence curve，which is similar to a “pot”，while the -320 m，-400 m，-480 m and -600 m sublevels have a “funnel” shaped subsidence curve.This phenomenon is related to the mining activities in the mine area.（4）The single settlement value in any monitoring period in the sublevel at any depth has the characteristics of up and down fluctuation，which reflects the non-linear characteristics of settlement deformation to a certain extent.The practice of seabed mining in Sanshandao gold mine confirms that：Since 2005，with the expansion of mining scale，the increase of mining intensity and the extension of mining years，the deformation of some of the underground tunnels is serious，and the movement and deformation of the underground surrounding rocks may cause seawater to gush into the tunnels along the damaged rocks，thus threatening the life safety of mining personnel.The study concluded that the thickness of the orebody，mining depth，mining intensity，surrounding lithology and rock structure as well as filling effect in the Xinli mining area are potential factors affecting the deformation of the surrounding rock in the mining of the inclined orebody.Among them，the presence of the controlling fault F₁ in the mine area directly affects the shape of the surrounding rock deformation curve，and this deformation feature should be considered in future production work as well as safety maintenance work（roadway repair work） to prevent from affecting productivity or even generating safety accidents.

With the rapid development of digital economy in the world，how to realize the rapid optimal allocation of underground mine production equipment has become the key to the continuous advancement and in-depth application of digital mine.In view of the characteristics of underground mines such as limited space，limited equipment resources，and large production tasks，an optimization model was constructed for production planning of the follow-up filling mining method in the open pit using pre-controlled roof medium-and deep-hole and sublevel open-stopping and subsequent filling method.The model aims at minimizing the interval time between adjacent processes and the total production time，and the above issue is solved using genetic algorithms.The genetic algorithms used for solving the problem include traditional genetic algorithms and optimized genetic algorithms.Taking the actual data of a copper mine test stope in Zambia as an example，it can be seen from the iterative results that all genetic algorithms can solve the model，and the optimized genetic algorithm converges faster than the ordinary genetic algorithm.The genetic algorithm accelerated by heuristic algorithm has the fastest convergence speed.Therefore，the heuristic genetic algorithm is used to solve the multi-objective optimization model and the results are visualized.After analyzing the solution results，it is found that the average utilization rate of equipment is only 49.16%，and the utilization rate of some equipment is low，so the number of equipment is optimized.After the number of equipment was optimized and solved again，the average utilization rate of mine equipment increased to 64.8%，basically meeting the requirements of the mine.In terms of production，the daily average ore output is 3 631.19 t/d，which meets the mining demand and effectively shorts the operation time interval to ensure the requirements of mining safety.In addition，copper and cobalt sunrise ore grade fluctuation is small，easy to concentrate.Therefore，the algorithm and model can quickly and effectively solve the problem of multi-equipment coordination in a copper mine in Zambia，improve production efficiency and safe mining.