A Smart Prediction Method for Surrounding Rock Safety Risk of Underground Powerhouse in Pumped Storage Power Station

Abstract

This invention relates to an intelligent prediction method for the safety risks of surrounding rock in underground powerhouses of pumped storage power stations, comprising the following steps: S1, firstly, multiple monitoring devices are installed in the surrounding rock of the underground powerhouse to monitor and collect data on surrounding rock displacement, stress change, fissure opening, and environmental parameters; S2, then, the collected data on surrounding rock displacement, stress change, fissure opening, and environmental parameters are preprocessed using a terminal to complete time synchronization, abnormal data removal, and data standardization, forming a time-series dataset required for surrounding rock safety analysis. This invention realizes real-time monitoring, intelligent prediction, automatic support, and feedback of safety risks in the surrounding rock of underground powerhouses, improves the efficiency of surrounding rock safety management, reduces the risks of manual inspection and emergency operations, and can adapt to the temporary support needs of different surrounding rock types and construction stages.

Description

Technical Field

[0001] This invention relates to the field of risk prediction for underground powerhouses of pumped storage power stations, specifically to an intelligent prediction method for the safety risks of surrounding rock in underground powerhouses of pumped storage power stations. Background Technology

[0002] Pumped storage power stations are generally used for peak shaving, frequency regulation, phase regulation, and emergency backup in power grids. Because pumped storage power stations are mostly underground powerhouse cavern structures with complex geological conditions, long construction sites, and numerous safety hazards in the caverns, high slopes, and deep foundation pits, safety risks exist in the construction processes such as cavern blasting, slope support, and foundation pit excavation. In some pumped storage power stations, multiple underground caverns, such as access tunnels, ventilation and safety tunnels, and tailrace tunnels, are transversely cut by a fault zone with poor geological conditions, making collapses prone to occur during underground cavern excavation. There have been precedents of localized rockfalls within tunnels during the construction of access tunnels for pumped storage power stations. Therefore, analyzing the stability of the surrounding rock and conducting risk prediction are essential tasks in the construction of pumped storage power stations. Surrounding rock deformation monitoring is a crucial aspect of determining surrounding rock stability, as surrounding rock deformation is actually a long-term process. In actual engineering, monitoring instruments and equipment are typically installed to obtain measured data, providing basic information for determining surrounding rock stability and deformation analysis.

[0003] According to the published patent CN117574781B, the design includes: acquiring time-series monitoring data of the surrounding rock of the underground powerhouse of a pumped storage power station; substituting the time-series monitoring data into a modified LSTM-CNN neural network model to predict the surrounding rock deformation, thereby obtaining the predicted surrounding rock deformation results; and performing surrounding rock stability analysis based on the predicted surrounding rock deformation results and surrounding rock stability evaluation indicators, thereby obtaining surrounding rock stability analysis data. This design predicts the safety risks of the surrounding rock of the underground powerhouse of a pumped storage power station in advance, improves the foresight of risk prediction, and greatly avoids potential safety problems.

[0004] In practice, traditional intelligent prediction methods for the safety risks of surrounding rock in underground powerhouses of pumped storage power stations use LSTM-CNN neural networks to predict rock deformation and obtain stability analysis data. However, when a predicted risk is detected, immediate on-site measures cannot be taken, relying instead on manual arrangements or subsequent operations, resulting in a slow risk response. Furthermore, while the method can predict rock deformation, it is not integrated with work zoning and on-site construction operations, leading to a lack of direct guidance for on-site support work and operational information gaps that can negatively impact the effectiveness of post-warning and post-treatment measures. Therefore, a new technical solution is needed to address these issues. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies, adapt to practical needs, and provide an intelligent prediction method for the safety risks of surrounding rock in underground powerhouses of pumped storage power stations. This addresses the current problem that traditional intelligent prediction methods for the safety risks of surrounding rock in underground powerhouses of pumped storage power stations use LSTM-CNN neural networks to predict the deformation of surrounding rock and obtain stability analysis data. However, when a predicted risk is discovered, on-site measures cannot be taken immediately, and the method relies on manual arrangements or subsequent operations, resulting in a slow risk response. Furthermore, although the method can predict the deformation of surrounding rock, it does not integrate with the work zoning and on-site construction operations, leading to a lack of direct guidance for on-site support work and a gap in operational information, which can easily affect the effectiveness of post-warning processing.

[0006] To achieve the objectives of this invention, the technical solution adopted is as follows: A method for intelligent prediction of safety risks of surrounding rock in underground powerhouses of pumped storage power stations is designed, comprising the following steps:

[0007] S1. First, multiple monitoring devices are installed in the surrounding rock of the underground powerhouse to monitor and collect data on surrounding rock displacement, stress change, crack opening, and environmental parameters.

[0008] S2. Next, the collected surrounding rock displacement data, surrounding rock stress change data, fracture aperture data and environmental parameter data are preprocessed using the terminal to complete time synchronization, abnormal data removal and data standardization, forming the time series dataset required for surrounding rock safety analysis.

[0009] S3. Construct a surrounding rock safety risk prediction model based on time series dataset, predict the trend of surrounding rock stability at different spatial locations of underground powerhouse, and output the corresponding risk level and risk spatial coordinates.

[0010] S4. Match the predicted risk level with the pre-divided work zones of the underground plant to determine the target zones with risks and the corresponding risk points.

[0011] S5. When the predicted risk level reaches the preset threshold, trigger the surrounding rock safety response command and send the response command to the scheduling control module.

[0012] S6. The scheduling and control module selects the target small mobile robot from multiple small mobile robots deployed in the underground plant, according to the location of the risk point, the spatial distance of the small mobile robots, and the current working status, and generates a work instruction.

[0013] S7. After receiving the operation instructions, the target small mobile robot autonomously navigates to the location corresponding to the risk point and uses the mechanical arm set on its top to carry out temporary rock support operations in the risk area.

[0014] S8. After the temporary support operation of the surrounding rock is completed, a small mobile robot collects the surrounding rock status data after support and feeds the collected data back to the scheduling and control module to correct the surrounding rock safety risk prediction results.

[0015] Preferably, the monitoring device includes a displacement monitoring unit, a stress monitoring unit, a crack monitoring unit, and an environmental monitoring unit.

[0016] In practical applications, by deploying displacement monitoring units, stress monitoring units, crack monitoring units, and environmental monitoring units in key parts and high-risk sections of the underground powerhouse, it is possible to collect data on surrounding rock displacement, stress changes, crack opening, and environmental parameters in real time. This provides comprehensive and accurate raw data for surrounding rock safety analysis and improves the reliability of risk prediction.

[0017] Preferably, the small mobile robot is equipped with a walking track mechanism at its bottom. The walking track mechanism includes a track frame, track plates, drive wheels, and guide wheels, which are used to move the small robot within the underground factory.

[0018] In practical use, the tracked mechanism enables the robot to move autonomously in complex underground terrain, making it highly adaptable and able to traverse narrow passages and uneven ground, thereby improving the robot's mobility and operational coverage.

[0019] Preferably, during navigation, the small mobile robot calibrates its position by installing external laser positioning probes and utilizing laser positioning, environmental feature recognition, or preset anchor point information.

[0020] In practical use, laser positioning, environmental feature recognition, and anchor point information calibration ensure the robot's positioning in complex underground environments, improving the accuracy of support operations and reducing operational errors and risks.

[0021] Preferably, the underground plant is divided into multiple work zones according to the cavern structure, construction stage, or operating area, and at least one small mobile robot is installed in each work zone.

[0022] In practical use, the work zones are divided according to the cavern structure, construction stage or operating area, and robots are deployed in each zone. This enables zone management and regional scheduling, improving the organizational efficiency of surrounding rock risk management.

[0023] Preferably, when selecting a target small mobile robot, the scheduling control module selects a small mobile robot that is located in the work zone where the risk point is located and is in an idle state, based on priority.

[0024] In practical use, by prioritizing the use of idle robots within the work zone where the risk point is located, the response efficiency can be maximized, and high-risk areas can be dealt with in a timely manner.

[0025] Preferably, the small mobile robot is equipped with a multi-axis robotic arm on its top, which is used to support the surrounding rock of the top or side wall of the underground plant after the small mobile robot stops at a predetermined position. The end of the robotic arm is equipped with a clamping mechanism, which is used to clamp the support components or monitoring units. This allows for the adaptation to different types of temporary support needs of surrounding rock by replacing different support components, or the replacement of different monitoring units to cooperate with the monitoring device to deal with local sudden deformation or cracks.

[0026] In practical use, the multi-axis robotic arm can flexibly operate the support components or monitoring units after the robot stops. The gripping mechanism can be replaced with different tools to achieve temporary support or local monitoring, thereby improving the support adaptability and emergency response capability.

[0027] Preferably, the support components include an expansion anchor launcher, a small shotcrete spray gun, an emergency fiber mesh installer, and a miniature high-pressure grouting head;

[0028] Among them, the expansion bolt launcher is used to press the expansion bolt into the rock mass, and the tail of the bolt expands and locks, thus fixing the rock mass with the bolt.

[0029] Small shotcrete guns are used to spray quick-setting concrete to form a thin protective layer to prevent weathering or further spalling.

[0030] Emergency fiber mesh installers are used to deploy fiber mesh to cover risk areas and secure the four corners of the mesh with edge clips or anchors to reinforce designated locations.

[0031] Miniature high-pressure grouting heads are used to inject quick-setting grout to fill cracks, block water flow, and reinforce the rock mass when there are obvious cracks in the top surrounding rock and slight water seepage.

[0032] In practical use, through diverse support methods, appropriate support measures can be selected according to different surrounding rock conditions to quickly reinforce risk areas, block water seepage, control crack expansion, and improve the overall safety of the surrounding rock of underground powerhouses.

[0033] Preferably, the small mobile robot completes the temporary rock support operation according to the pre-stored operation zoning information and risk response rules, and uploads the operation data after communication is restored.

[0034] Preferably, when the small mobile robot encounters problems while performing temporary rock support work, the scheduling and control module selects another small mobile robot to move to the target location to perform the support work.

[0035] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0036] 1. This invention, by setting up monitoring devices at different locations in the surrounding rock of an underground powerhouse and equipping a small mobile robot with a robotic arm and support components, enables the prediction of surrounding rock safety risks and on-site temporary support. Compared with traditional methods that only perform predictions, this invention can automatically or manually operate and schedule the robot to navigate to the target area when a risk occurs. The top robotic arm can quickly implement various types of support operations, such as expansion anchor bolts, shotcrete, fiber mesh, or high-pressure grouting. At the same time, it collects surrounding rock status data after support and feeds it back to the scheduling and control module to correct the risk prediction results, thereby improving the real-time performance, accuracy, and operability of surrounding rock safety risk protection.

[0037] 2. By setting up at least one small mobile robot in each work area, this invention can achieve cross-area support and rapid replacement of backup robots. At the same time, the gripping mechanism of the small mobile robot can hold different monitoring units. By combining the monitoring device with the mobile acquisition of the monitoring unit of the small robot, it can achieve dual acquisition and monitoring of fixed and mobile data, which improves the accuracy of data acquisition and can cope with sudden local deformation or cracks. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the process of the present invention;

[0039] Figure 2 This is a schematic diagram of the monitoring device of the present invention;

[0040] Figure 3 This is a schematic diagram of the small mobile robot of the present invention;

[0041] Figure 4 This is a schematic diagram of the support component of the present invention. Detailed Implementation

[0042] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0043] A method for intelligent prediction of surrounding rock safety risks in underground powerhouses of pumped storage power stations, see [link to relevant documentation]. Figures 1 to 4 This includes the following steps:

[0044] S1. First, multiple monitoring devices are installed in the surrounding rock of the underground powerhouse to monitor and collect data on surrounding rock displacement, stress change, crack opening, and environmental parameters.

[0045] Specifically, displacement, stress, crack and environmental monitoring units are installed at key or high-risk locations in the surrounding rock of the underground powerhouse. Sensors collect real-time data on changes in surrounding rock displacement, stress distribution, crack opening, and environmental parameters such as temperature, humidity and seepage. The sensors convert the collected analog signals into digital signals and upload them to the data processing terminal via wired or wireless networks.

[0046] S2. Next, the collected surrounding rock displacement data, surrounding rock stress change data, fracture aperture data and environmental parameter data are preprocessed using the terminal to complete time synchronization, abnormal data removal and data standardization, forming the time series dataset required for surrounding rock safety analysis.

[0047] Specifically, the collected time-series data is synchronized, outliers and noise are removed, and the data units and numerical ranges are standardized through standardization methods. The processed data forms a complete time-series dataset that can be directly input into the risk prediction model.

[0048] S3. Construct a surrounding rock safety risk prediction model based on time series dataset, predict the trend of surrounding rock stability at different spatial locations of underground powerhouse, and output the corresponding risk level and risk spatial coordinates.

[0049] Specifically, based on the processed time-series data, machine learning or statistical modeling methods, such as time-series analysis and LSTM-CNN hybrid models, are used to predict the stability state of the surrounding rock at different spatial locations, calculate the potential instability probability of the surrounding rock, and generate low, medium, and high risk levels and their corresponding spatial coordinates.

[0050] S4. Match the predicted risk level with the pre-divided work zones of the underground plant to determine the target zones with risks and the corresponding risk points.

[0051] Specifically, the risk levels predicted by the model are matched with the pre-planned work zones of the underground plant, and the locations of high-risk zones and specific risk points are determined by coordinate positioning.

[0052] S5. When the predicted risk level reaches the preset threshold, trigger the surrounding rock safety response command and send the response command to the scheduling control module.

[0053] Specifically, when the risk level reaches a preset threshold, the terminal automatically generates a safety response command, which is then sent to the designated robot or management terminal through the scheduling and control module.

[0054] S6. The scheduling and control module selects the target small mobile robot from multiple small mobile robots deployed in the underground plant, according to the location of the risk point, the spatial distance of the small mobile robots, and the current working status, and generates a work instruction.

[0055] Specifically, the scheduling module selects the most suitable robot and generates work instructions based on the spatial coordinates of the risk point, the robot's current position and status.

[0056] S7. After receiving the operation instructions, the target small mobile robot autonomously navigates to the location corresponding to the risk point and uses the mechanical arm set on its top to carry out temporary rock support operations in the risk area.

[0057] Specifically, the robot moves within the underground plant using a walking track mechanism. Navigation, combined with laser positioning, environmental feature recognition, or preset anchor point information, enables it to autonomously reach risk points. The multi-axis robotic arm operates support components such as expansion anchors, shotcrete, fiber mesh, and grouting according to work instructions to complete temporary support.

[0058] S8. After the temporary support operation of the surrounding rock is completed, a small mobile robot collects the surrounding rock status data after support and feeds the collected data back to the scheduling and control module to correct the surrounding rock safety risk prediction results.

[0059] Specifically, after the robot completes the support operation, it collects data on the surrounding rock condition, such as displacement and stress, through the monitoring unit, and uploads the data to the scheduling and control module to correct the risk prediction model and achieve optimization.

[0060] Specifically, the monitoring device includes a displacement monitoring unit, a stress monitoring unit, a crack monitoring unit, and an environmental monitoring unit.

[0061] Furthermore, the small mobile robot is equipped with a walking track mechanism at its bottom. The walking track mechanism includes a track frame, track plates, drive wheels, and guide wheels, which are used to move the small robot in the underground plant. During navigation, the small mobile robot completes its own position calibration by installing laser positioning probes on its exterior, using laser positioning, environmental feature recognition, or preset anchor point information. The underground plant is divided into multiple work zones according to the cavern structure, construction stage, or operating area, and at least one small mobile robot is set up in each work zone.

[0062] It is worth noting that when selecting target small mobile robots, the scheduling and control module selects small mobile robots that are located in the work zone where the risk point is located and are in an idle state, based on priority.

[0063] It is worth noting that the small mobile robot is equipped with a multi-axis robotic arm on its top, which is used to support the surrounding rock of the top or side walls of the underground plant after the small mobile robot stops at the predetermined position. The end of the robotic arm is equipped with a clamping mechanism, which is used to clamp the support components or monitoring units. This allows for the adaptation to different types of temporary support needs of surrounding rock by replacing different support components, or to replace different monitoring units to cooperate with the monitoring device to deal with local sudden deformation or cracks. The support components include an expansion anchor bolt launcher, a small shotcrete spray gun, an emergency fiber mesh layer, and a miniature high-pressure grouting head.

[0064] Among them, the expansion anchor bolt launcher is used to press the expansion anchor bolt into the rock mass, and the tail of the anchor bolt expands and locks, thus fixing the rock mass. When the deformation of the surrounding rock increases, such as when there is slight detachment of the top or side wall surrounding rock, the anchor bolt is quickly implanted to fix the rock mass. Specifically, after the robotic arm is aligned with the warning position, the laser alignment device calibrates the drilling point, and the launcher is activated to press the expansion anchor bolt into the rock mass, and the tail of the anchor bolt expands and locks.

[0065] Small shotcrete guns are used for temporary sealing of areas with local rock collapse risk or dense cracks, such as when there are loose rock blocks or multiple small cracks at the warning location. They spray quick-setting concrete to form a thin protective layer to prevent weathering or further collapse. Specifically, the robotic arm adjusts the angle of the gun, the laser rangefinder controls the distance between the nozzle and the surrounding rock, and the air pump is started to spray the quick-setting material evenly.

[0066] The emergency fiber mesh layer is used to quickly lay fiber mesh to cover and prevent small rocks from falling when there is a large area of ​​loose surrounding rock, such as local weathering and loosening of the rock mass on the side wall of the main plant. Specifically, the robotic arm moves to the target area, unfolds the fiber mesh to cover the risk surface, and fixes the four corners of the mesh with edge clips or anchors.

[0067] Miniature high-pressure grouting heads are used to fill cracks caused by underground water seepage in surrounding rock fissures, such as when there are obvious cracks in the top surrounding rock with slight water seepage. They inject quick-setting grout to fill the cracks, block water flow, and reinforce the rock mass. Specifically, the robotic arm is positioned to the crack or pre-embedded grouting interface through visual positioning, starts the grouting pump, and the flow sensor provides real-time feedback on the grouting volume. It automatically stops after reaching the preset value.

[0068] It is worth mentioning that the small mobile robot completes the temporary rock support operation based on the pre-stored work zoning information and risk response rules, and uploads the work data after communication is restored. When the small mobile robot encounters problems in completing the temporary rock support operation, the scheduling and control module selects another small mobile robot to move to the target location to perform the support operation.

[0069] Example 1

[0070] First, monitoring devices, including displacement monitoring units, stress monitoring units, crack monitoring units, and environmental monitoring units, are deployed in key parts and high-risk sections of the underground powerhouse to collect real-time data on surrounding rock displacement, stress changes, crack opening, and environmental parameters. The collected data is preprocessed, including time synchronization, abnormal data removal, and standardization, to form a complete time-series dataset for subsequent risk analysis.

[0071] Next, a surrounding rock safety risk prediction model is constructed based on time series data to predict the trend of surrounding rock stability at different spatial locations of the underground powerhouse, outputting the corresponding risk level and risk spatial coordinates. The prediction results are matched with the pre-divided work zones to determine the target zones with risks and the specific risk points. When the predicted risk level reaches the preset value, a surrounding rock safety response command is triggered and sent to the scheduling and control module.

[0072] The scheduling and control module prioritizes idle small mobile robots to perform support tasks based on the location of the risk point, the spatial distance between the work zones of each small mobile robot and the current work status. The selected robot's bottom walking track mechanism drives the robot to autonomously navigate to the risk point location. During navigation, the robot uses laser positioning probes, environmental feature recognition and preset anchor point information for position calibration. After reaching the target location, the multi-axis robotic arm on the top of the robot unfolds to perform the operation. The clamping mechanism at the end of the robotic arm can be replaced with an expansion bolt launcher, a small shotcrete spray gun, an emergency fiber mesh layer or a mini high-pressure grouting head, which are used to press in the anchor bolts to fix the rock mass, spray thin-shell concrete, lay fiber mesh or inject quick-setting grout to reinforce the surrounding rock in the risk area.

[0073] After the support operation is completed, the robot can collect the surrounding rock status data through the clamping monitoring unit and feed the data back to the scheduling control module to correct the risk prediction model. If a robot failure or support abnormality occurs during the support process, the scheduling module will select other idle robots to complete the support, so that the risk points can be dealt with in a timely manner.

[0074] This embodiment enables real-time monitoring, intelligent prediction, automatic support, and feedback of safety risks in the surrounding rock of underground powerhouses, improving the efficiency of surrounding rock safety management, reducing the risks of manual inspections and emergency operations, and adapting to the temporary support needs of different surrounding rock types and construction stages.

[0075] In addition, all components designed in this invention are general standard parts or components known to those skilled in the art. Their structure and principle can be known to those skilled in the art through technical manuals or conventional experimental methods. Those skilled in the art can fully implement them, so there is no need to elaborate. The content protected by this invention does not involve improvements to the internal structure and method.

Claims

1. A method for intelligent prediction of safety risks of surrounding rock in underground powerhouses of pumped storage power stations, characterized in that, Includes the following steps: S1. First, multiple monitoring devices are installed in the surrounding rock of the underground powerhouse to monitor and collect data on surrounding rock displacement, stress change, crack opening, and environmental parameters. S2. Next, the collected surrounding rock displacement data, surrounding rock stress change data, fracture aperture data and environmental parameter data are preprocessed using the terminal to complete time synchronization, abnormal data removal and data standardization, forming the time series dataset required for surrounding rock safety analysis. S3. Construct a surrounding rock safety risk prediction model based on time series dataset, predict the trend of surrounding rock stability at different spatial locations of underground powerhouse, and output the corresponding risk level and risk spatial coordinates. S4. Match the predicted risk level with the pre-divided work zones of the underground plant to determine the target zones with risks and the corresponding risk points. S5. When the predicted risk level reaches the preset threshold, trigger the surrounding rock safety response command and send the response command to the scheduling control module. S6. The scheduling and control module selects the target small mobile robot from multiple small mobile robots deployed in the underground plant, according to the location of the risk point, the spatial distance of the small mobile robots, and the current working status, and generates a work instruction. S7. After receiving the operation instructions, the target small mobile robot autonomously navigates to the location corresponding to the risk point and uses the mechanical arm set on its top to carry out temporary rock support operations in the risk area. S8. After the temporary support operation of the surrounding rock is completed, a small mobile robot collects the surrounding rock status data after support and feeds the collected data back to the scheduling and control module to correct the surrounding rock safety risk prediction results.

2. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, The monitoring device includes a displacement monitoring unit, a stress monitoring unit, a crack monitoring unit, and an environmental monitoring unit.

3. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, The small mobile robot is equipped with a walking track mechanism at its bottom, which includes a track frame, track plates, drive wheels and guide wheels, and is used to move the small robot in the underground factory.

4. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, During navigation, the small mobile robot calibrates its position by installing external laser positioning probes and utilizing laser positioning, environmental feature recognition, or preset anchor point information.

5. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, The underground plant is divided into multiple work zones according to the cavern structure, construction stage, or operating area, and at least one small mobile robot is installed in each work zone.

6. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, When selecting a target small mobile robot, the scheduling and control module selects a small mobile robot that is located in the work zone where the risk point is located and is in an idle state, based on priority.

7. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, The small mobile robot is equipped with a multi-axis robotic arm on its top. After the small mobile robot stops at a predetermined position, it is used to support the surrounding rock of the top or side walls of the underground plant. The end of the robotic arm is equipped with a clamping mechanism, which is used to clamp the support components or monitoring units. This allows for the replacement of different support components to meet different types of temporary support needs for surrounding rock, or the replacement of different monitoring units to cooperate with the monitoring device to deal with sudden local deformation or cracks.

8. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 7, characterized in that, The support components include an expansion anchor launcher, a small shotcrete spray gun, an emergency fiber mesh installer, and a mini high-pressure grouting head; Among them, the expansion bolt launcher is used to press the expansion bolt into the rock mass, and the tail of the bolt expands and locks, thus fixing the rock mass with the bolt. Small shotcrete guns are used to spray quick-setting concrete to form a thin protective layer to prevent weathering or further spalling. Emergency fiber mesh installers are used to deploy fiber mesh to cover risk areas and secure the four corners of the mesh with edge clips or anchors to reinforce designated locations. Miniature high-pressure grouting heads are used to inject quick-setting grout to fill cracks, block water flow, and reinforce the rock mass when there are obvious cracks in the top surrounding rock and slight water seepage.

9. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, The small mobile robot completes temporary rock support work according to pre-stored work zoning information and risk response rules, and uploads work data after communication is restored.

10. The intelligent prediction method for the safety risk of surrounding rock in the underground powerhouse of a pumped storage power station as described in claim 1, characterized in that, When the small mobile robot encounters a problem while performing temporary rock support work, the scheduling and control module selects another small mobile robot to move to the target location to perform the support work.

Images