143 results about "Underground cavern" patented technology

The invention provides a surrounding rock stability dynamic analysis method and system in the excavation process of an underground cavern group. The method comprises the steps that original surrounding rock monitoring data of the underground cavern group are collected at the self-adaptive sampling frequency through a distributed optical fiber sensor; performing real-time filtering, abnormal value primary screening and data compression processing on the original surrounding rock monitoring data to obtain target surrounding rock monitoring data; performing feature extraction on the target surrounding rock monitoring data to obtain a feature vector corresponding to the target surrounding rock monitoring data; and inputting the feature vector into a pre-constructed surrounding rock stability evaluation model for evaluation to obtain the stability level of the surrounding rock. Therefore, data redundancy can be prevented, key information omission is avoided, and the reliability of a surrounding rock stability evaluation result is improved.

The application provides a kind of stability discrimination method of underground gas storage cavern initial support of compressed air energy storage power station, the method comprises: considering the expansion force and softening of hydrophilic surrounding rock of underground gas storage cavern, draw the response curve of surrounding rock in humidification process;Obtain the supporting parameters of underground gas storage cavern, calculate and draw the characteristic curve of supporting structure;The intersection point of the response curve of surrounding rock and the characteristic curve of supporting structure is determined, and the coordinate of the intersection point is obtained according to the coordinate of the intersection point, and the equilibrium supporting force of hydrophilic surrounding rock and supporting balance is obtained according to the coordinate of the intersection point;According to the safety factor of underground gas storage cavern, the stability of the cavern is judged according to the safety factor.The application fully considers the expansion force, softening and supporting force effect of surrounding rock in the relationship between surrounding rock and support, and uses safety factor as the stability index of underground gas storage cavern, which is suitable for the stability evaluation of underground cavern with complex surrounding rock characteristics, and the discrimination method is accurate and efficient.

The application discloses the technical field of image data processing, and provides a method, device and equipment for risk assessment of an underground cavern and a medium, which comprises the following steps: adopting a geometric segmentation algorithm to process point cloud data of a target cavern acquired in advance, combining a plane fitting technology, identifying a structural surface and extracting structural surface parameters; constructing a geological topology model according to the structural surface parameters, identifying key blocks of the geological topology model based on a preset block identification standard, and randomly disturbing the structural surface parameters to obtain multiple groups of sample parameters; determining the instability probability of each key block based on the sample parameters, performing risk assessment on the key blocks based on the instability probability, and obtaining a risk assessment result of the underground cavern. The application significantly improves the timeliness and efficiency of risk identification of underground engineering, and effectively solves the problems of accurate identification of a structural surface and risk quantization under complex geological conditions.

The invention provides a nonlinear finite element discrimination method for a critical stable section of an underground cavern, and the method comprises the steps: determining a rock mass strain softening parameter according to a rock mass mechanical test in actual engineering; performing secondary development on a rock mass constitutive model based on the rock mass strain softening parameters; based on the rock mass constitutive model after secondary development, a nonlinear finite element initial model for solving the critical stable section of the underground cavern is constructed; the nonlinear finite element model is solved, and the overall tangent rigidity of the model in the calculation time step is obtained; based on the overall tangent stiffness, calculating to obtain a minimum characteristic value of an overall tangent stiffness matrix; and based on the minimum characteristic value, distinguishing the critical stable section of the underground cavern. The method directly and quantitatively reflects the essential characteristics of rigidity loss and bearing capacity reduction of the whole cavern structure system after a complex nonlinear deformation path including strain softening is considered, so that the critical instability moment and the potential instability section are accurately captured.

The invention discloses a compressed air energy storage underground cavern temperature adjusting structure and a manufacturing method thereof, and belongs to the technical field of compressed air energy storage, the temperature adjusting structure is sequentially provided with a flexible sealing layer, a phase change temperature adjusting layer and an optical fiber sensing layer from inside to outside, the optical fiber sensing layer is connected with the inner wall of a concrete lining, and the phase change temperature adjusting layer is used for stabilizing temperature fluctuation in a cavern; the optical fiber sensing layer is used for monitoring temperature and a strain field in real time and is electrically connected with an external fiber grating sensor grid monitoring system through a sensor lead. According to the method, latent heat can be absorbed or released through the phase change process, temperature fluctuation in the cavern can be effectively stabilized, material fatigue and structural damage caused by thermal stress circulation are reduced, real-time monitoring of the temperature and strain of the sealing layer is achieved, and the sensing capacity and the early warning level of the operation state of the cavern are improved.

The invention provides a block model construction method for underground cavern group surrounding rock stability analysis, which comprises the following steps: generating an initial block according to a topological relation between a fault and a joint in a three-dimensional geologic model, and carrying out geometric reconstruction on a block interface through Boolean operation to eliminate overlapping; automatically matching lithology types of the divided blocks with an engineering attribute database, dynamically distributing elastic modulus, Poisson's ratio and tensile strength, establishing rock attribute mapping, synchronously iteratively solving interaction equations of a mechanical deformation field, a seepage pressure field and a temperature stress field in a unified solver, and performing multi-physics field coupling; a potential sliding surface is automatically identified according to a multi-physics field coupling result, a sliding surface friction coefficient and shear strength are fused to calculate a safety coefficient, non-overlapping block division is carried out by adopting a topological relation of faults and joints, an interaction equation of a mechanical deformation field, a seepage pressure field and a temperature stress field is synchronously iterated through a unified solver, and a non-overlapping block is obtained. And accurate quantitative analysis of surrounding rock stability is realized.

The invention provides an underground cavern group geological analysis method based on a TGS technology and a CT technology, and the method comprises the steps: scanning the geological conditions of an underground cavern group in the horizontal direction through the TGS technology, and obtaining first geological data in the horizontal direction; forecasting the geological conditions of the underground cavern group in the vertical direction through a CT technology to obtain second geological data in the vertical direction; fusing the first geological data and the second geological data to generate a three-dimensional geological model; based on the three-dimensional geologic model, identifying the position and the distribution range of unfavorable geologic bodies in the underground cavern group, and analyzing the geological characteristics of the surrounding rock to obtain an analysis result; according to the analysis result, the spatial distribution information of the unfavorable geologic body and the corresponding geological phenomenon description are output, the misjudgment risk caused by data discontinuity of a traditional method is overcome, and the recognition capacity for complex geological phenomena such as inclined strata and cross fractures is improved.

The invention discloses an underground cavern surrounding rock stability determination method, and relates to the technical field of underground engineering monitoring, and the method comprises the steps: constructing a surrounding rock stability causal diagram according to a surrounding rock fusion feature sequence, and carrying out the causal analysis and delay relation fusion through a causal relation determination method, and obtaining a surrounding rock stability causal network diagram; performing path analysis on the surrounding rock stability causal network diagram to obtain a surrounding rock stability master control path, and establishing surrounding rock stability evaluation indexes for quantitative scoring to obtain a surrounding rock stability scoring time sequence; analyzing the surrounding rock stability scoring time sequence and the change trend and visual displacement abnormity of the preprocessed surrounding rock monitoring data by using a surrounding rock instability space-time early warning model to obtain a surrounding rock early warning grade and an early warning area; according to the method, causal association modeling is performed on the surrounding rock fusion feature sequence, so that the sensitivity and response speed of abnormal evolution are improved, and technical support of reliability and operability is provided for safety management of the surrounding rock of the underground cavern.

The invention relates to the field of underground cavern group blasting construction, and particularly discloses an underground cavern group blasting construction working face dust control intelligent spraying method which comprises the steps that acoustic information and scene sequence images of a construction site are collected, and working face blasting construction procedures are obtained; after it is confirmed that blasting is completed on the construction site, real-time dust concentration space distribution information of the construction working face is obtained according to the scene sequence image; a spraying working position is determined according to the dust concentration space distribution information, and a navigation positioning system is used for autonomous navigation to reach the spraying working position; spraying parameters are determined according to the dust concentration space distribution information, and spraying work is started; the spraying effect is evaluated according to the real-time dust concentration space distribution information, and spraying work is stopped after the dust concentration control condition reaches the standard. The invention further provides a corresponding device. According to the method and the device, the dust concentration can be obtained in real time, and intelligent rapid dust falling autonomous operation can be implemented.

The invention provides a hydropower station underground cavern position optimization method, and relates to the technical field of hydropower stations, and the method comprises the steps: screening the potential arrangement position of an underground cavern according to the ground stress, geological conditions and hub arrangement conditions of an engineering area; distribution of the workshop structural surfaces under the potential arrangement positions is analyzed, potential unfavorable blocks and the positions of the potential unfavorable blocks are calculated and analyzed, and the positions comprise the internal position of the hydropower station and the structural surface combination; calculating the distance between the potential arrangement position and the plant control fault; from the aspect of block stability, the stability is determined based on the distribution of plant structural surfaces and potential unfavorable blocks, the anti-seepage difficulty is determined based on the distance with a plant control fault, the stability, the anti-seepage difficulty and hub arrangement conditions are comprehensively considered, and the optimal arrangement position of the underground cavern is screened. By the adoption of the scheme, the optimal selection of the arrangement position of the underground cavern of the hydropower station is achieved.

The invention relates to the technical field of hydraulic underground caverns, and discloses an underground cavern support design case matching method and device based on semantic weight, and the method comprises the steps: inputting target text data into a pre-training language model, so as to enable the pre-training language model to output a semantic vector sequence of the target text data; inputting the semantic vector sequence of the target text data into a pre-constructed entity extraction model, so that the entity extraction model outputs a second entity tag sequence of the target text data; performing similarity calculation on the second entity tag sequence and the first entity tag sequence of each historical underground cavern support design case based on target semantic weights corresponding to different entities; and determining a case matching result of the target underground cavern support design case based on the similarity between the target underground cavern support design case and each historical underground cavern support design case, thereby effectively improving the accuracy of the case matching result.

The application provides a kind of underground cavern microstructure surface quantum sensing detection device, including nanoscale piezoelectric sensor array, data acquisition unit, signal processing unit and structure surface identification unit;Nanoscale piezoelectric sensor array is deployed in surrounding rock surface or borehole, and microseismic signal is detected;Data acquisition unit collects and converts signal into digital signal;Signal processing unit pre-processes and extracts feature data;Structure surface identification unit identifies the position, dip angle and length of microstructure surface based on feature data.The application can detect underground cavern microstructure surface with high sensitivity and high precision, provide important basis for rock mass stability evaluation and engineering design, reduce geological disaster risk and ensure the safety of underground engineering.

The application discloses a multi-source data monitoring underground cavern stability evaluation method and system and a medium, relates to the related field of geotechnical engineering monitoring, and the method comprises the following steps: based on the geological survey parameters and engineering design parameters of the target underground cavern, the surrounding rock stability grade of the target underground cavern is output through a pre-trained surrounding rock stability classification model; multi-source time series monitoring data of the target underground cavern during the construction period is collected, a set of dynamic influence factors is extracted therefrom, and the set of dynamic influence factors is input into a pre-trained surrounding rock deformation prediction model to output surrounding rock deformation prediction values in a future period; the surrounding rock stability grade and the surrounding rock deformation prediction values are fused to perform stability evaluation, and decision feedback information for the current construction stage is generated according to the stability evaluation result. The technical problems that the existing underground cavern stability evaluation result is not accurate and unreliable are solved, and the technical effects of improving the accuracy and reliability of the underground cavern stability evaluation are achieved.

The invention discloses a flexible supporting and sealing structure of a compressed air energy storage underground cavern and a construction method thereof.The underground cavern is excavated in surrounding rock, the flexible supporting and sealing structure comprises a polymer sealing layer, a fiber reinforced sprayed concrete layer and a plurality of telescopic hollow anchor rods, the polymer sealing layer covers the inner surface of the surrounding rock, and the fiber reinforced sprayed concrete layer covers the inner surface of the surrounding rock; the fiber-reinforced sprayed concrete layer covers the inner surface of the polymer sealing layer and serves as a primary support, the inner surface of the fiber-reinforced sprayed concrete layer is covered with a concrete lining, and the telescopic hollow anchor rods are grouted and anchored through a flexible polymer grouting material. And a full-anchor flexible supporting system for surrounding rock is formed. According to the flexible support sealing structure, the surrounding rock can be flexibly supported, so that the surrounding rock can adapt to the displacement direction change caused by the internal pressure effect of circulating air, the adaptability and toughness of the overall structure of the cavern are remarkably enhanced, and therefore the service life of the cavern is effectively prolonged under the effect of periodic loads.

The invention relates to the technical field of TBM equipment, and discloses a pumped storage power station underground cavity TBM applicability evaluation method, which comprises the following steps: S1, firstly carding an existing framework of pumped storage power station underground cavity TBM applicability evaluation, and defining core technical elements and a protection range of the framework; an evaluation framework of non-AHP driving, dynamic multi-source data fusion and machine learning enabling is provided. According to the method, an evaluation framework of non-AHP driving, dynamic multi-source data fusion and machine learning enabling is provided, an index system is constructed, the index system is divided into static indexes (in a design stage, cavern burial depth, surrounding rock category and turning radius) and dynamic indexes (in a construction stage, TBM attitude error, cutter wear rate, surrounding rock deformation rate and ventilation efficiency), a real-time monitoring system is integrated, and the dynamic index is divided into a dynamic index and a dynamic index; and data such as TBM attitude, cutter wear and surrounding rock deformation are input into the model in real time, and the applicability score is dynamically updated, so that the method has the advantages of high pertinence, comprehensive index system and the like.

The present application relates to the field of compressed air energy storage, and particularly relates to a compressed air energy storage underground cavern lining test device with less required filling gas and more stable gas pressure, comprising a support cylinder, which is sleeved with a ring-shaped lining, and a sealed cavity is formed between the side wall of the support cylinder and the elastic sealing layer of the side wall of the ring-shaped lining; and an air inlet is arranged to connect the sealed cavity with the outside. The present application is particularly suitable for use in the compressed air energy storage underground cavern lining test.

This invention discloses a three-dimensional finite element simulation method for underground caverns that considers the dynamic evolution of surrounding rock deformation parameters. Based on triaxial loading and unloading tests of the surrounding rock in an underground cavern, the method first obtains the evolution equations of the surrounding rock deformation parameters under different stress paths, then constructs a three-dimensional finite element calculation model of the underground cavern. Through secondary development technology of finite element software, the evolution equations are embedded into the simulation process, realizing real-time dynamic adjustment of the deformation modulus and Poisson's ratio of the surrounding rock as the stress state changes during excavation. During the simulation, the method cyclically executes "stress monitoring - path judgment - parameter calculation - model assignment - convergence calculation" according to the excavation step sequence until the cavern excavation is completed. This invention restores the deterioration law of surrounding rock parameters under unloading during underground cavern excavation, improves the fit between numerical simulation results and engineering reality, and provides a reliable basis for the stability analysis of the surrounding rock and the design of excavation and support schemes for underground caverns.

The invention provides an underground cavern group analysis method combined with a disturbance effect, a program product and equipment, and relates to the technical field of engineering data processing. The method comprises the following steps: acquiring monitoring data of a plurality of indexes of an underground cavern group in a construction process; according to the stress parameter of the target cavern disturbed by the adjacent cavern and the working procedure crossing duration of the target cavern and other caverns, the comprehensive disturbance effect intensity is determined; taking the monitoring data and the comprehensive disturbance effect intensity as first time-varying risk factors, screening out second time-varying risk factors from the first time-varying risk factors based on the correlation between the first time-varying risk factors and time-varying progress risks, and forming a time-varying risk factor sequence according to the second time-varying risk factors of different processes; and processing the time-varying risk factor sequence by using a time sequence model with an attention mechanism to obtain an analysis result of the target cavern. The accuracy of the analysis result of the underground cavern group is improved.

The invention relates to the technical field of geotechnical engineering support, and discloses a slowly-inclined thin-layer soft rock advanced small guide pipe hole row matching grouting support method and an advanced small guide pipe. The grouting supporting method comprises the steps that the true rock stratum thickness of the underground cavern vault surrounding rock and the included angle between the rock stratum surface and the tunneling direction are obtained; the axial distance between every two adjacent rows of grouting holes is determined according to the outer inserting angle, the true rock stratum thickness and the included angle designed by the small guide pipe; a plurality of rows of grouting holes are machined in the surface of the small guide pipe at intervals, and the plane where each row of grouting holes are located coincides with the design of a rock stratum face; and the machined small guide pipe is installed in the drill hole for grouting operation. According to the method, the geometric matching relation between the intervals of the grouting hole rows and the rock stratum structure is established, and the grouting hole rows are aligned with the main weak surface of the bedding surface, so that grout can preferentially enter the bedding surface and diffuse along the bedding surface. The grouting reinforcement effect on the slowly-inclined thin-layer soft rock is improved, and deformation and damage along the bedding surface after excavation are effectively restrained.

The invention discloses an assembly type pipe shed system for a water-rich stratum deeply-buried underground cavern, which comprises a plurality of pipe shed units, the plurality of pipe shed units are linearly arranged, friction yielding connecting units are fixedly connected among the plurality of pipe shed units, and the plurality of friction yielding connecting units connect the plurality of pipe shed units in series. The invention further discloses a construction method of the fabricated pipe shed system for the water-rich stratum deeply-buried underground cavern. According to the fabricated pipe shed system for the water-rich stratum deeply-buried underground cavern and the construction method, the problems that a traditional pipe shed system is poor in adaptability and has split functions under complex geological conditions are solved. The pipe shed is reasonable in design, the functions of supporting, energy dissipation, grouting, monitoring and maintenance are integrated, the technical defects of a traditional pipe shed are effectively overcome, the safety and economical efficiency of complex surrounding rock tunnel construction are improved, and remarkable engineering value and popularization prospects are achieved.

The application discloses a kind of underground cavern group construction ventilation network energy consumption minimization method, it is related to ventilation network optimization and energy consumption control technical field, this method includes: constructing dynamic ventilation network perception and reconstruction model during construction, and the topological structure is dynamically updated by multi-source sensor and construction progress log;Based on Hardy Cross method, the theoretical air flow distribution satisfying physical conservation is solved;Fusion theoretical air flow and real-time data, construct Hardy Cross-GB hybrid prediction model, train gradient boosting regression to correct prediction bias, output air flow and energy consumption prediction value;Based on the model, the system total power minimization objective function is constructed, the frequency conversion strategy of fan is optimized under the constraint of air demand and the control instruction is output.The application realizes dynamic ventilation network accurate simulation and real-time energy efficiency optimization, significantly reduces the energy consumption of ventilation system, guarantees construction safety and control stability.

The invention relates to the technical field of image data acquisition in a live-action three-dimensional reconstruction technology, and provides an underground cavern live-action modeling image shooting pose planning method and device and a storage medium, and the method comprises the steps: shooting a section in a current cavern, controlling a pan-tilt shooting device to sequentially carry out the horizontal alignment and adaptive vertical sequence shooting of two side walls, and carrying out the self-adaptive vertical sequence shooting of the two side walls; cross coverage shooting of the opposite side walls is completed; dynamically calculating the target position of the next shooting section according to the camera field angle and a preset image overlapping ratio requirement, and moving the pan-tilt shooting equipment; and repeating the process until any point of the excavated wall surface meets the requirement of the target image coverage number. According to the invention, through self-adaptive pose planning and dynamic back calculation of the pitch angle and the advancing distance, it is ensured that the image overlapping rate meets the three-dimensional reconstruction requirement, automatic control over the image shooting process is achieved, manpower can be saved, the shooting efficiency can be greatly improved, the image shooting quality can be improved, and the problems of image blurring and the like are avoided.

This invention relates to the field of geotechnical engineering physical simulation technology, and more particularly to a physical simulation test device and method for simulating the sectional unloading and disturbance excavation of underground caverns. The test device includes a loading unit, a support and fixing unit, a modular cavern simulation unit, an unloading unit, an impact disturbance unit, and a monitoring unit. This test device can accurately realize the sectional unloading and disturbance excavation of large caverns, and through micro-adjustment, it can simulate the gradual stress release process of each excavation layer. In terms of disturbance and impact, multi-level clamps and spring devices can control the number of impacts and the magnitude of the impact force to simulate the impact of tunnel blasting disturbance excavation on the stability of the surrounding rock. This invention can accurately simulate the sectional and sectional gradual blasting excavation disturbance and unloading process of underground caverns, and can provide a reliable experimental platform for the mechanical behavior and stability analysis of underground cavern construction.

The invention relates to the technical field of peripheral hole drilling equipment, in particular to an angle control device for drilling peripheral holes in an underground cavity, a rotating plate is hinged to a fixed plate, a hinge seat is arranged at the other end of the rotating plate, and an overturning assembly penetrates through the fixed plate, then is connected to the hinge seat and is used for driving the rotating plate to rotate; the first lifting assembly is used for driving the table top at the first bearing base to ascend and descend, and the second lifting assembly is used for driving the table top at the second bearing base to ascend and descend. The table top is supported and fixed through the support, the drill rod is fixed to the table top through the first bearing base and the second bearing base, the table top and the rotating plate are connected through the lifting assembly, the rotating plate is driven by the overturning assembly to rotate, and the position of the table top is close to peripheral holes, so that drilling is convenient; meanwhile, the first lifting assembly and the second lifting assembly can drive the two opposite sides of the table top to ascend and descend to form a height difference, the table top is controlled to incline to meet the drilling requirement, and the problems that in the drilling process, the position close to peripheral holes is difficult to adapt, and the drilling angle is difficult to control are solved.

This utility model provides a telescopic support device for underground cavern excavation, comprising at least two support component units and telescopic connecting rods connecting adjacent support component units. Each end of the telescopic connecting rod is fixedly connected to one of the support component units. The extension and retraction of the telescopic connecting rods drives the adjacent support component units to move closer or further apart. This support device provides timely support for underground caverns. During support, the device is installed inside the cavern, and the telescopic supports of the support component units extend to tighten the arch support beam and sidewall support beams respectively. Using multiple support component units provides support force to the surrounding rock, increasing the excavation advance per cycle, improving excavation efficiency, and shortening the construction period. The combined use of the telescopic connecting rods and telescopic supports allows for movement within the underground cavern, and the support device is reusable, saving costs.

The application relates to the technical field of acoustic wave testing, in particular to an acoustic wave testing equipment for underground caverns; the equipment comprises a tester and a transducer connected with the tester through a wire; the transducer transmits waves and transmits signals to the tester through the wire after receiving the waves; the transducer is fixedly connected with an elastic pipe through a transducing seat; an elastic liquid inlet pipe and an elastic gas outlet pipe pass through the elastic pipe; one end of the liquid inlet pipe is connected with the outer wall of the transducing seat, and the other end is connected with a water pump; in the process of approaching the testing hole, the two pushing chains are turned inward and combined to form a pushing piece; in the process of moving away from the testing hole, the two pushing chains are turned outward and separated; thus, the testing equipment can meet the testing requirements of testing holes with different depths, the pushing piece can be combined and separated quickly, the testing efficiency is high, and the equipment can be used in the space-limited surroundings of some testing hole orifices.

A personnel positioning system and method for a large underground cavern complex based on a 5G communication network is disclosed. The system includes: a 5G communication network module transmitting the unique NR-PCI code and antenna deployment information of each base station to a GIS information system module; the GIS information system module providing an indoor geographic database containing the binding relationship between NR-PCI codes and geographic locations to a positioning engine module; the positioning engine module calculating the location of the user equipment based on the measurement data, NR-PCI codes, and GIS geographic information reported by the user equipment, and sending the positioning result to a data management server module; the data management server module storing the positioning result, personnel trajectory, and system configuration data, and supporting access by upper-layer applications; and the user equipment accessing the 5G network, collecting positioning measurements such as reference signal received power, signal arrival time, and signal arrival angle, and uploading the data to the positioning engine module to report its own location. This system addresses the problems of insufficient positioning accuracy, reliance on additional base stations, and poor adaptability in existing technologies.

The utility model relates to tunnel engineering construction technical field discloses a kind of tunnel blasting smoke exhaust intelligent control system, including spraying subsystem, multi-sensor detection module and remote control terminal with display;Spraying subsystem includes several spray heads and booster water pump;Remote control terminal with display receives the information of multi-sensor detection module, and controls booster water pump;Remote control terminal with display shows the data detected by multi-sensor detection module;Multi-sensor detection module includes smoke inductor, sound inductor, vibration inductor and air quality detector;The utility model is applied in tunnel or underground cavern group, especially in the environment of poor ventilation and continuous blasting construction needs, can effectively reduce dust in air, dust generated by blasting will not affect operation construction after passing through multiple atomization spraying, can effectively save spraying water resource simultaneously, with good economic benefit and social benefit, suitable for promotion and use.

The invention provides an underground cavern stability evaluation method and device, electronic equipment and a storage medium, and relates to the technical field of underground engineering and rock mass mechanics, structural plane parameters of surrounding rock where a target underground cavern is located are collected, sample expansion is performed on the structural plane parameters to obtain multiple groups of working condition samples, and for each group of working condition samples, the corresponding working condition sample is selected; determining a corresponding safety coefficient, preprocessing the collected structural plane parameters and multiple groups of working condition samples, and training a pre-constructed prediction model by taking the safety coefficient corresponding to each group of working condition samples as a supervision label based on the preprocessed structural plane parameters and multiple groups of working condition samples, and inputting the structural plane parameters of the to-be-evaluated underground cavern into the trained prediction model to obtain a safety coefficient corresponding to the to-be-evaluated underground cavern output by the prediction model, and finally realizing efficient and batch evaluation of the stability of any to-be-evaluated underground cavern.