An intelligent detection and control method for cavern pressure based on artificial cavern test model

Abstract

The present invention belongs to the technical field of cavern pressure detection, and specifically relates to an intelligent detection and control method for cavern pressure based on an artificial cavern test model. By dividing the cavern space area into key areas and non-key areas, and then redundantly deploying information collection points in the key areas, and solely deploying information collection points in the non-key areas, the pressure risk early warning capability of the key areas can be improved. Meanwhile, the single deployment in the non-key areas can simplify the data acquisition system and reduce unnecessary complexity. At the same time, environmental parameter detection is added when performing pressure detection at the information collection points deployed in the cavern space area. When performing pressure risk identification, not only the risk caused by the pressure reaching the warning level itself can be identified, but also the trend risk caused by excessive environmental increase can be identified. The comprehensiveness and accuracy of risk identification are improved, and targeted processing direction is provided for subsequent pressure control.

Description

Technical Field

[0001] The invention belongs to the technical field of cavern pressure detection, and in particular relates to an intelligent detection and control method for cavern pressure based on an artificial cavern test model. Background Art

[0002] A cavern is a cave or space excavated in rock for a specific purpose during mining or other underground engineering construction. Pressure monitoring and control of caverns is a crucial link in modern mining management. With the increase of mining depth, ground pressure also increases. If effective monitoring and control are not carried out, dangerous events such as rock bursts may occur. Therefore, pressure detection, prediction and control of caverns are key measures to ensure safe production in mines.

[0003] The pressure detection and prediction of the cavern inevitably requires obtaining test data through experiments, but there are high safety risks in conducting experiments in actual mines or underground projects, and under natural conditions, many factors are uncontrollable. Therefore, in order to better control the test conditions and conduct experiments safely, the current pressure detection and prediction of the cavern usually rely on artificial cavern test models for operation, but the current detection control has the following shortcomings: 1. Since the cavern has a certain spatial area, it is necessary to arrange the collection points before conducting pressure detection. The existing layout of pressure collection points is mostly uniform and single, and lacks redundant layout for key areas. This layout method is easily affected by single point failures. Once a collection point fails, it may lead to data loss in the area. For the key area, once the data is lost, it cannot truly reflect the pressure state inside the cavern, which further leads to the lack of sufficient data support for pressure control, making it difficult to take preventive measures in time, which increases the risk of accidents to a certain extent.

[0004] 2. After the pressure test, the next step is risk identification. However, the current risk identification of cavern pressure mainly relies on direct pressure test results, and less consideration is given to the trend pressure risk caused by changes in environmental conditions. When the temperature and humidity in the environment increase significantly, it is easy to cause pressure to increase. If this trend pressure risk is ignored, it will not only lead to a lag in pressure risk identification, affecting the timeliness and accuracy of risk warnings, but also cause the subsequent pressure control to be solidified, making it difficult to provide targeted processing directions for subsequent pressure control, resulting in limited processing effects. Summary of the invention

[0005] In view of this, the present invention aims to propose an intelligent detection and control method for cavern pressure based on an artificial cavern test model, which effectively solves the problems mentioned in the background technology by focusing on the layout of collection points in cavern pressure detection and optimizing risk identification after pressure detection.

[0006] The object of the present invention can be achieved through the following technical scheme: A method for intelligent detection and control of cavern pressure based on an artificial cavern test model, comprising the following steps: S1, redundantly deploying information collection points in the cavern, and setting up detection terminals at the locations of the deployed information collection points, wherein the detection terminals are composed of a pressure gauge and environmental detection equipment.

[0007] S2. Start the detection terminal of each information collection point to detect pressure and environmental parameters to obtain detection data, and extract valid detection data therefrom.

[0008] S3. Identify risk collection points based on the extracted valid detection data.

[0009] S4. Extract construction attribute information from the construction file of the cavern to construct an artificial cavern test model of the cavern, thereby marking the position in the constructed test model based on the location of the risk collection point.

[0010] S5. Collect environmental time series detection data from risk collection points and analyze the environmental change rate.

[0011] S6. Obtain the current environmental parameters of the risk collection point, and thereby combine the current environmental parameters with the environmental change rate to form several test groups, each test group corresponding to independent variable environmental parameter data and fixed environmental parameter data.

[0012] S7. Input the independent variable environmental parameter data and the fixed environmental parameter data corresponding to each test group into the test model for simulation to obtain the simulated pressure of the risk collection point in each test group.

[0013] S8. Classify the test groups with the same independent variable environmental parameters to form a test group set, and classify and analyze the simulated pressure of each test group in the set, thereby determining the risk tendency environmental parameters of the risk collection point.

[0014] S9. Communicate the location of risk collection points and risk tendency environmental parameters to the control center, which will perform pressure control.

[0015] Combining all the above-mentioned technical solutions, the positive effects of the present invention are as follows: 1. The present invention divides the cavern space area into critical areas and non-critical areas, and then redundantly deploys information collection points in critical areas, and single deployment of information collection points in non-critical areas. This can minimize the data loss in critical areas caused by single point failures, which is beneficial to improving the pressure risk warning capability in critical areas and ensuring timely discovery and handling of potential risks. At the same time, single deployment in non-critical areas can simplify the data collection system and reduce unnecessary complexity.

[0016] 2. The present invention also adds environmental parameter detection when performing pressure detection at the information collection points arranged in the cavern space area. When performing pressure risk identification, it can identify not only the risks caused by the pressure reaching the warning level itself, but also the trend risks caused by excessive environmental increase. This not only improves the comprehensiveness and accuracy of risk identification, but also provides a more targeted processing direction for subsequent pressure control, thereby more effectively controlling risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0018] Figure 1 It is a diagram of the steps for implementing the method of the present invention.

[0019] Figure 2 This is a schematic diagram of extracting the center position of a cavern under the regular cavern shape in the present invention.

[0020] Figure 3 Schematic diagram of volume network division of the cavern with irregular shape in the present invention

[0021] Figure 4 It is a schematic diagram of the process of extracting effective detection data in the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] See also Figure 1 As shown, the present invention proposes an intelligent detection and control method for cavern pressure based on an artificial cavern test model, comprising the following steps: S1, redundantly deploying information collection points in the cavern, and setting detection terminals at the deployed information collection points, wherein the detection terminal is composed of a pressure gauge and an environmental detection device, and further the environmental detection device is a temperature sensor and a humidity sensor.

[0024] In the optimization of the above scheme, redundant layout of information collection points in the tunnel is implemented as follows: S11, three-dimensional scanning of the tunnel is performed to obtain a three-dimensional image of the tunnel, and the central area of ​​the tunnel is identified from the image. The specific identification process is to use an edge detection algorithm to extract the contour of the tunnel boundary from the three-dimensional image of the tunnel, and then use a surface fitting method (such as the least squares method) to fit the surface of the tunnel to obtain geometric features, and then match the extracted geometric features with known regular shapes (such as cylinders, cubes, etc.) to determine whether they conform to a certain regular shape, thereby identifying whether the shape of the tunnel is regular. If the shape of the tunnel is regular, the geometric center is extracted from the three-dimensional image of the tunnel as the center position of the tunnel, see Figure 2 As shown, for example, the geometric center calculation function of the 3D modeling software can be used. If the shape of the cave is irregular, see Figure 3 As shown in the figure, the three-dimensional image of the cavern is divided into multiple small volume grids, and the volume of each grid is counted respectively, and then the center position of the cavern is obtained by weighted average according to the volume of each grid. After obtaining the center position of the cavern, the shape of the central area is determined. If the shape is determined to be a circle, the radius is defined according to actual needs. The center position of the cavern is taken as the center of the circle, and a circle is drawn with the defined radius. The area inside the circle is the center area of ​​the cavern. If the shape is determined to be a square, the side length is defined to obtain the center area of ​​the cavern.

[0025] It is important to note that the above radius and side length definitions must be reasonable and must be able to cover the core areas that require attention.

[0026] S12. Using an edge detection algorithm, identify the edge contour of the image from the three-dimensional image of the mine.

[0027] S13, pre-generate a template of a known supporting structure, such as a steel arch frame, anchor rod, prestressed anchor cable, etc.), and then use a template matching algorithm to search for an area matching the template in the three-dimensional image, and annotate the searched area to obtain a supporting structure area.

[0028] S14. Evenly and redundantly arrange information collection points in the central area and supporting structure area of ​​the cavern, and arrange information collection points singly in other areas except the central area and supporting structure area.

[0029] It should be understood that the central area of ​​the cavern is selected for redundant layout of information collection points because of the first point: the central area is usually the place where stress is most concentrated. Since the cavern is subjected to forces on all four sides, the stress borne by the central area is often large, and it is easy to become an area of ​​stress concentration. The second point: the central area is the core part of the cavern structure. The stress state of the central area directly affects the stress distribution of the surrounding rocks. Any instability in the central area may cause the instability of the entire cavern structure. The third point: the central area is very sensitive to deformation. Any slight deformation may lead to stress redistribution, and the deformation of the central area is easily transmitted to the surrounding areas, affecting the uniformity of the overall structure. It is precisely in view of the above characteristics of the central area that the central area of ​​the cavern needs to be used as a key area for redundant layout of information collection points.

[0030] It needs to be further explained that the reason why the supporting structure area is selected for redundant layout of information collection points is that the first point is that the supporting structure is the main load-bearing component of the cavern, and its strength and stability are directly related to whether the cavern can safely bear the predetermined load; the second point is that the supporting structure area is often an area with higher safety risks, and any problem with the supporting structure may lead to serious safety accidents. Therefore, it is also necessary to regard the supporting structure area in the cavern as a key area for redundant layout of information collection points.

[0031] In the implementation of the above plan, when redundantly deploying information collection points in the central area and supporting structure area of ​​the cavern, the number of deployment points should be reasonably determined based on the area. Specifically, for larger areas, the number of information collection points should be increased to ensure the breadth and reliability of data coverage, taking into account factors such as geological conditions, support types and monitoring purposes; while for smaller areas, the number of information collection points can be appropriately reduced to achieve efficient use of resources and ensure effective monitoring and data collection in areas of different sizes.

[0032] It should be noted that the purpose of setting up the information collection point is to set up the detection terminal in the information collection point, so the information collection point needs to be a physical point, but the central area of ​​the cavern may be a cavity area, and the information collection point set up in this area is a virtual positioning point, which represents the location that needs to be monitored. Although there is no physical support, the detection terminal can be fixed at this location by other methods to ensure accurate data collection. For example, a bracket can be set around the cavity, and the detection terminal can be suspended in the air by the bracket; for another example, the suspension device can be fixed to the solid rock or structure around the cavity using expansion bolts.

[0033] S2. Start the detection terminal of each information collection point to detect pressure and environmental parameters to obtain detection data, where the environmental parameters are temperature and humidity, and extract valid detection data from them.

[0034] It should be understood that the reason for using temperature and humidity as environmental parameters is that temperature and humidity will have a certain impact on pressure, which is specifically reflected in the following aspects: when the temperature rises, the movement of gas molecules in the cavern space intensifies, resulting in an increase in the frequency of collisions between gas molecules and the container wall, which in turn leads to an increase in pressure; in addition, temperature changes will cause thermal expansion and contraction of materials. In a cavern environment, if the surrounding rock or supporting materials are deformed, the volume of the enclosed space may change, thereby affecting the internal pressure. For example, when the temperature rises, the expansion of rock or supporting materials may reduce the space, resulting in an increase in pressure. Increased humidity means an increase in the water vapor content in the air, which in turn leads to an increase in the partial pressure of water vapor, thereby increasing the total air pressure in the cavern; in addition, some materials are hygroscopic, and increased humidity will cause these materials to absorb moisture and expand, changing their physical state, which in turn affects their bearing capacity and pressure distribution. For example, some rocks absorb moisture and expand in a high humidity environment, which may cause an increase in pressure.

[0035] Applied to the above scheme, see Figure 4 As shown, the effective detection data extraction process is as follows: S21, dividing the cavern space area into a single layout area and a redundant layout area according to whether the information collection points are redundantly arranged.

[0036] S22. Use the detection data of the detection terminals in a single deployment area as valid detection data.

[0037] S23, comparing the detection timestamps of multiple detection terminals in the redundant deployment area, if the detection timestamps are consistent, execute S24-S26, otherwise execute S27.

[0038] It should be pointed out that in each detection terminal, the design of the pressure gauge and the environmental detection equipment realizes the synchronous detection function, that is, the pressure data and the environmental parameter data are obtained at the same detection time. Therefore, when verifying the time consistency of the data obtained by the detection terminal, there is no need to compare the pressure detection time and the environmental parameter detection time inside each detection terminal separately. Instead, it is only necessary to compare the overall detection timestamps of each detection terminal to confirm the time consistency of the data of each detection terminal.

[0039] The above-mentioned detection timestamp consistency comparison can specifically compare the detection timestamps of multiple detection terminals, select the earliest timestamp and the latest timestamp, and then obtain the detection interval duration based on the earliest timestamp and the latest timestamp, and then compare the detection interval duration with the allowed detection interval duration. If the detection interval duration is less than or equal to the allowed detection interval duration, where the allowed detection interval duration can be 0.1s, then the identification detection timestamp is consistent; if the detection interval duration is greater than the allowed detection interval duration, then the identification detection timestamp is inconsistent.

[0040] S24. Calculate the standard deviation of the detection data of multiple detection terminals, and compare the calculation result of the standard deviation with the defined threshold. If the calculation result of the standard deviation does not reach the defined threshold, it indicates that the differences between the detection data of multiple detection terminals are small and the distribution is relatively concentrated. At this time, execute S25. On the contrary, it indicates that the differences between the detection data of multiple detection terminals are large and the distribution is relatively differentiated. At this time, execute S26.

[0041] S25. Take the average data from the detection data of multiple detection terminals as the valid detection data.

[0042] S26. Number each detection terminal, and sequentially extract each detection terminal as the target detection terminal according to the number order. Calculate the standard deviation of the detection data of other detection terminals and the detection data of the target detection terminal to obtain the calculation results of the standard deviation with each detection terminal as the target detection terminal. Among them, the detection data corresponding to the detection terminal with the smallest standard deviation calculation result is the closest to the data of other detection terminals and has high representativeness. Therefore, take the detection data corresponding to the detection terminal with the smallest standard deviation calculation result as the valid detection data.

[0043] S27. Compare the detection timestamps of multiple detection terminals to identify the reference detection timestamp. The reference detection timestamp is the detection timestamp in the middle. Furthermore, align the detection data of the detection terminals whose detection timestamps are not the reference detection timestamp to the reference detection timestamp using linear interpolation to obtain the interpolated alignment data. At the same time, assign different weights to the detection data of different detection terminals according to the proximity of the detection timestamp to the reference detection timestamp. Therefore, perform weighted average calculation on the detection data or interpolated alignment data of each detection terminal combined with the assigned weights to obtain the valid detection data.

[0044] The above method can effectively process the data of multiple detection terminals by selecting the middle timestamp as the reference when the detection timestamps of multiple detection terminals are inconsistent, and combining linear interpolation and weighted average with weights to ensure the consistency and accuracy of the data.

[0045] In the example of the above solution, assume that there are three detection terminals, and the detection timestamps of each detection terminal are T1, T2, and T3 respectively, where T1 < T2 < T3. At this time, the reference detection timestamp is T2. Furthermore, use linear interpolation to align the detection terminals with detection timestamps T1 and T3 to T2 to obtain the interpolated alignment data. At the same time, the proximity degrees of each detection terminal to the reference detection timestamp are respectively 、1、 ,where represents the detection interval duration.

[0046] It should be noted that the linear interpolation method is chosen in the above because the concept is intuitive, easy to understand and explain, and the mathematical formula is simple and easy to implement without complex algorithms. In addition, through linear interpolation, data with different timestamps can be aligned to the same time point to ensure the time consistency of the data, and can provide reasonable results in most cases.

[0047] S3. Identify risk collection points based on the extracted effective detection data. The specific identification process is: extract pressure data from the effective detection data of the single layout area and the redundant layout area, and compare it with the warning pressure, where the warning pressure can be extracted from the construction design file of the cavern. If the pressure data of a certain area reaches the warning pressure, the information collection points deployed in the area are used as the main risk collection points, and the detection time of the effective detection data is used as the risk detection time.

[0048] Environmental parameter data is extracted from the effective detection data of the single deployment area and the redundant deployment area, and the environmental parameter data at the earliest detection time is obtained from several previous detection times as the previous environmental parameter data. At this time, the extracted environmental parameter data is subtracted from the previous environmental parameter data and divided by the previous environmental parameter data to obtain the analysis environment increase, which is then compared with the set critical value. Exemplarily, the critical value is 0.6. If the environmental increase in a certain area reaches the critical value, the information collection points deployed in the area are used as trend risk collection points, and the detection time of the effective detection data is used as the risk detection time.

[0049] It should be added that the previous several detection moments mentioned above are a pre-set time window. For example, the length of the time window is 20 minutes, and then the detection moments existing in the time window are taken as the previous several detection moments.

[0050] It should be pointed out that since the environmental parameters include temperature and humidity, the environmental increase rates analyzed are the temperature increase rate and the humidity increase rate. As long as either the temperature increase rate or the humidity increase rate is higher than the critical value, the information collection points deployed in the area will be used as trend risk collection points.

[0051] S4. Extract construction attribute information from the construction file of the cavern to construct an artificial cavern test model of the cavern, thereby marking the position in the constructed test model based on the location of the risk collection point.

[0052] Applied to the above scheme, the construction attribute information extracted from the construction file includes but is not limited to geometric parameters: the size (length, width, height), shape, structure, etc. of the cavern. Geological parameters: rock type, rock thickness, crack distribution, groundwater conditions, etc. Physical and mechanical parameters: elastic modulus, Poisson's ratio, compressive strength, shear strength, etc. of rock. Construction parameters: support method (such as anchor rods). When constructing an artificial cavern test model based on construction attribute information, first select appropriate numerical simulation software, then draw a three-dimensional model of the cavern in the software according to the extracted geometric parameters, and then set the properties of rocks and other materials in the drawn three-dimensional model according to the geological parameters and physical and mechanical parameters, thereby constructing an artificial cavern test model.

[0053] It should be noted that the artificial chamber test model constructed by the present invention is a virtual model. The advantage of using a virtual model is that it can provide more detailed simulation results, and multiple simulation experiments can be quickly carried out under different conditions, saving time and cost.

[0054] S5. Collect environmental time series detection data from risk collection points and analyze the environmental change rate.

[0055] In an optional method, the environmental time series detection data of the risk collection point is collected as follows: the risk detection moment of the risk collection point and several previous detection moments constitute a time series set, and then the environmental parameter data is extracted from the effective detection data of the risk collection point at each detection moment in the time series set to constitute the environmental time series detection data of the risk collection point.

[0056] In the above example, assume that the risk detection time is t5, and the previous detection times are t1, t2, t3, and t4. t1, t2, t3, t4, and t5 constitute a time series set At this time, the environmental timing detection data includes the temperature timing detection data And humidity timing detection data .

[0057] In a further optional method, the environmental change rate is analyzed by referring to the following process: the environmental parameter data at each detection moment in the environmental time series detection data of the risk collection point is subtracted from the previous detection moment, and the absolute value of the difference result is taken and divided by the environmental parameter data at the previous detection moment to obtain the interval environmental change rate of each adjacent detection moment.

[0058] In the embodiment of the above scheme, since there is no previous detection time at t1, the adjacent detection times are t2 and t1, t3 and t2, t4 and t3, and t5 and t4.

[0059] The fluctuation of the interval environment change rate of each adjacent detection moment is statistically analyzed, where the fluctuation can be used to extract the maximum interval environment change rate and the minimum interval environment change rate from the interval environment change rate, and then the difference between the maximum interval environment change rate and the minimum interval environment change rate is divided by the time length of the time series set, and compared with the pre-configured allowable fluctuation. Exemplarily, the allowable fluctuation is 0.4. If it is less than or equal to the allowable fluctuation, it means that the environmental parameters change relatively smoothly. Taking the mean as the environment change rate can better reflect the real change trend of the environmental parameters. At this time, the mean of the interval environment change rate of each adjacent detection moment is taken as the environment change rate. Otherwise, it means that the environmental parameters change greatly. Taking the maximum interval environment change rate as the environment change rate can ensure that the maximum change amplitude is captured. At this time, the maximum interval environment change rate among the interval environment change rates of each adjacent detection moment is taken as the environment change rate.

[0060] It should be added that, since the environmental parameters include temperature and humidity, the statistical environmental change rate includes the temperature change rate and the humidity change rate.

[0061] S6. Obtain the current environmental parameters of the risk collection point, and thereby combine the current environmental parameters with the environmental change rate to form several test groups, each test group corresponding to independent variable environmental parameter data and fixed environmental parameter data.

[0062] It should be pointed out that the current environmental parameters of the risk collection point refer to the environmental parameter data detected by the risk collection point at the time of risk detection.

[0063] The operations for forming several test groups in the above are as follows: combining the current environmental parameters of the risk collection point with the change rate of its corresponding environmental parameters, and obtaining the predicted values ​​of various environmental parameters through multiple calculations.

[0064] In the example of the above scheme, it is assumed that the current environmental parameters of the risk collection point are temperature: 30°C and humidity: 20%, the temperature change rate is 0.05, and the humidity change rate is 0.10. Since the calculation of the environmental change rate is to subtract the environmental parameter data at each detection moment from the previous detection moment, and take the absolute value of the difference result and divide it by the environmental parameter data at the previous detection moment, at this time, temperature: 30°C and humidity: 20% are used as the environmental parameter data at the previous detection moment. In the present invention, the increase in temperature and humidity will increase the pressure, and the risk collection point is screened based on the environmental increase rate. Therefore, the temperature prediction value calculated for the first time is 30+0.05×30=31.5°C, the humidity prediction value calculated for the first time is 20%+0.10×20%=22%, the temperature prediction value calculated for the second time is 31.5+0.05×31.5=33°C (rounded off), and the humidity prediction value calculated for the second time is 22%+0.10×22%=24.2%. The predicted values ​​of each environmental parameter in each calculation are obtained by analogy.

[0065] It should be pointed out that the humidity mentioned in the present invention is relative humidity.

[0066] The independent variable environmental parameter is determined, and thus the independent variable environmental parameter prediction value calculated at each risk collection point and the current data of another environmental parameter constitute the independent variable environmental parameter data and fixed environmental parameter data corresponding to each test group.

[0067] It should be noted that the purpose of determining the independent variable environmental parameters and the fixed environmental parameters is to exclude the influence of other environmental parameter changes on the experimental results during the test simulation, so as to more accurately evaluate the influence of the independent variable environmental parameters on the pressure change. For example, when the independent variable environmental parameter is temperature, the predicted temperature data corresponding to the test group is used as the independent variable environmental parameter data of the test group, and the current humidity of 20% is used as the fixed environmental parameter data; when the independent variable environmental parameter is humidity, the predicted humidity data corresponding to the test group is used as the independent variable environmental parameter data of the test group, and the current temperature of 30°C is used as the fixed environmental parameter data.

[0068] S7. Input the independent variable environmental parameter data and the fixed environmental parameter data corresponding to each test group into the test model for simulation to obtain the simulated pressure of the risk collection point in each test group.

[0069] S8. Classify the test groups belonging to the same independent variable environmental parameters to form a test group set, and classify and analyze the simulated pressure of each test group in the set, so as to determine the risk tendency environmental parameters of the risk collection point. The specific operations are as follows: If the risk collection point is the entity risk collection point, then classify the simulated pressure of each test group in the test group set with the same independent variable environmental parameters for pressure trend analysis, and then take the environmental parameters corresponding to the maximum pressure trend as the risk tendency environmental parameters.

[0070] The above-mentioned pressure trend analysis can construct a simulated pressure curve with the simulated pressure of each test group in each environmental parameter as the horizontal coordinate and the simulated pressure as the vertical coordinate, and then obtain the overall change rate of the curve as the pressure trend.

[0071] If the risk collection point is a trend risk collection point, the test groups with the same independent variable environmental parameters will be classified and numbered according to the test order, and the simulated pressure of each test group in the test group set will be compared with the warning pressure to screen out the test group number that reaches the warning pressure. At the same time, the risk proximity index corresponding to each environmental parameter is calculated. The risk proximity index can be obtained by subtracting the test group number that reaches the warning pressure from the number 1, and using the natural constant as the base and the difference result as the exponent to construct an exponential function to obtain the risk proximity index, and then screen out the environmental parameter corresponding to the maximum risk proximity index as the risk tendency environmental parameter.

[0072] It needs to be explained that when the risk collection point is an entity risk collection point or a trend risk collection point, the identification of risk tendency environmental parameters has different focuses. Specifically, when the risk collection point is an entity risk collection point, it means that the pressure at this point is already in a warning state. Then the identification of the risk tendency environmental parameters of this collection point is to compare the pressure change trend under different environmental parameters, determine which environmental parameter change has led to a significant increase in pressure, and select the environmental parameter with the largest pressure change trend as the risk tendency environmental parameter; when the risk collection point is a trend risk collection point, it means that the pressure at this point is not yet in a warning state, but the environmental increase is large, and there may be potential risks. Then the identification of the risk tendency environmental parameters of this collection point is to simulate the pressure change under different environmental parameters, calculate the time required to reach the warning pressure, and select the environmental parameter that causes the pressure to reach the warning pressure the fastest as the risk tendency environmental parameter. In this way, by distinguishing different stages of the pressure state, different analysis methods are used to identify the risk tendency environmental parameters, which is more targeted and reasonable.

[0073] It should be additionally added that it is not ruled out that a certain risk collection point is both an ontological risk collection point and a trend risk collection point. In this case, the risk tendency environmental parameters obtained by using the risk collection point as the ontological risk collection point are compared with the risk tendency environmental parameters obtained as the trend risk collection point. If they are consistent, the consistent risk tendency environmental parameters are used as the final risk tendency environmental parameters of the risk collection point. If they are inconsistent, the risk tendency environmental parameters obtained by using the risk collection point as the ontological risk collection point are used as the final risk tendency environmental parameters of the risk collection point. This can ensure that risks that are currently in an alert state are dealt with first to prevent further deterioration.

[0074] S9. The location of the risk collection point and the risk tendency environmental parameters are communicated to the control center, which performs targeted processing based on the risk tendency environmental parameters. For example, when the risk tendency environmental parameter is temperature, a cooling process is performed, and when the risk tendency environmental parameter is humidity, a dehumidification process is performed, so as to more effectively control the pressure risk.

[0075] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent detection and control method for cavern pressure based on an artificial cavern test model, characterized in that: The steps include: S1. Arrange information collection points in the cavern and set up detection terminals at the arranged information collection points, wherein the detection terminals are composed of a pressure gauge and environmental detection equipment; The arrangement of information collection points in the cave is implemented as follows: S11, performing a three-dimensional scan on the cavern to obtain a three-dimensional image of the cavern, and identifying a central area of ​​the cavern from the image; S12, using an edge detection algorithm to identify the edge contour of the image from the three-dimensional image of the cavern; S13, pre-generating a template of a known support structure, and then using a template matching algorithm to search for an area matching the template in the three-dimensional image, and marking the searched area to obtain a support structure area; S14. Perform uniform redundant layout of information collection points in the central area and supporting structure area of ​​the cavern, and perform single layout of information collection points in other areas except the central area and supporting structure area; S2, start the detection terminal of each information collection point to detect pressure and environmental parameters to obtain detection data, and extract valid detection data from it, the environmental parameters are temperature and humidity; S3, identify risk collection points based on the extracted valid detection data; S4, extracting construction attribute information from the construction file of the cavern to construct an artificial cavern test model of the cavern, thereby marking the position in the constructed test model based on the position of the risk collection point; S5. Collect environmental time series detection data from risk collection points and analyze the environmental change rate; S6. Obtain the current environmental parameters of the risk collection point, and thereby combine the current environmental parameters with the environmental change rate to form several test groups, each test group corresponding to independent variable environmental parameter data and fixed environmental parameter data; S7, inputting the independent variable environmental parameter data and the fixed environmental parameter data corresponding to each test group into the test model for simulation to obtain the simulated pressure of the risk collection point in each test group; S8, classifying the test groups belonging to the same independent variable environmental parameters to form a test group set, and classifying and analyzing the simulated pressure of each test group in the group set, thereby determining the risk tendency environmental parameters of the risk collection point; S9. Communicate the location of risk collection points and risk tendency environmental parameters to the control center, which will perform pressure control.

2. The intelligent detection and control method for cavern pressure based on an artificial cavern test model according to claim 1 is characterized in that: The process of extracting effective detection data is as follows: S21, dividing the cavern space area into a single layout area and a redundant layout area according to whether the information collection points are redundantly laid out; S22, taking the detection data of the detection terminals in a single deployment area as valid detection data; S23, comparing the detection timestamps of multiple detection terminals in the redundant deployment area, if the detection timestamps are consistent, executing S24-S26, otherwise executing S27; S24, calculating the standard deviation of the detection data of multiple detection terminals, and comparing the standard deviation calculation result with the specified threshold value. If the standard deviation calculation result does not reach the specified threshold value, execute S25, otherwise execute S26; S25, taking average data from the detection data of multiple detection terminals as valid detection data; S26, numbering each detection terminal, and extracting each detection terminal in sequence as a target detection terminal in the order of the numbers, calculating the standard deviation of the detection data of other detection terminals and the detection data of the target detection terminal to obtain the standard deviation calculation result with each detection terminal as the target detection terminal, and then taking the detection data of the detection terminal corresponding to the minimum standard deviation calculation result as the valid detection data; S27. Compare the detection timestamps of multiple detection terminals to identify the benchmark detection timestamp, and then use linear interpolation to align the detection data corresponding to the detection terminal whose detection timestamp is not the benchmark detection timestamp to the benchmark detection timestamp to obtain interpolation alignment data. At the same time, different weights are assigned to different detection terminal data according to the degree of proximity between the detection timestamp and the benchmark detection timestamp, thereby performing weighted averaging calculation on the detection data or interpolation alignment data of each detection terminal in combination with the assigned weights to obtain effective detection data.

3. The intelligent detection and control method for cavern pressure based on an artificial cavern test model according to claim 2 is characterized in that: The risk collection point identification is as follows: Extract pressure data from the effective detection data of the single deployment area and the redundant deployment area, and compare it with the warning pressure. If the pressure data of a certain area reaches the warning pressure, the information collection point deployed in the area is used as the main risk collection point, and the detection time of the effective detection data is used as the risk detection time; Environmental parameter data is extracted from the effective detection data of the single deployment area and the redundant deployment area, and the environmental parameter data at the earliest detection time is obtained from several previous detection times as the previous environmental parameter data. At this time, the extracted environmental parameter data is subtracted from the previous environmental parameter data and divided by the previous environmental parameter data to obtain the analysis environment increase degree, which is then compared with the set critical value. If the environmental increase degree of a certain area reaches the critical value, the information collection points deployed in the area are used as trend risk collection points, and the detection time of the effective detection data is used as the risk detection time.

4. The intelligent detection and control method for cavern pressure based on an artificial cavern test model as claimed in claim 3 is characterized by: The process of collecting environmental time series detection data of risk collection points is as follows: The risk detection moment of the risk collection point and several previous detection moments constitute a time series set, and then the environmental parameter data are extracted from the effective detection data of the risk collection point at each detection moment in the time series set to constitute the environmental time series detection data of the risk collection point.

5. The intelligent detection and control method for cavern pressure based on an artificial cavern test model according to claim 4 is characterized in that: The analysis of the environmental change rate refers to the following process: Subtract the environmental parameter data at each detection time from the environmental time series detection data of the risk collection point from the previous detection time, and take the absolute value of the difference result and divide it by the environmental parameter data at the previous detection time to obtain the interval environmental change rate of each adjacent detection time; The fluctuation of the interval environment change rate of each adjacent detection moment is statistically analyzed and compared with the pre-configured allowable fluctuation. If it is less than or equal to the allowable fluctuation, the average of the interval environment change rates of each adjacent detection moment is taken as the environment change rate. Otherwise, the maximum interval environment change rate among the interval environment change rates of each adjacent detection moment is taken as the environment change rate.

6. The intelligent detection and control method for cavern pressure based on an artificial cavern test model as claimed in claim 3 is characterized by: The current environmental parameter of the risk collection point refers to the environmental parameter data detected by the risk collection point at the time of risk detection.

7. The intelligent detection and control method for cavern pressure based on an artificial cavern test model according to claim 1 is characterized in that: The operation of forming several test groups is as follows: Combine the current environmental parameters of the risk collection point with the change rate of its corresponding environmental parameters, and obtain the predicted values ​​of each environmental parameter through multiple calculations; The independent variable environmental parameter is determined, and thus the independent variable environmental parameter prediction value calculated at each risk collection point and the current data of another environmental parameter constitute the independent variable environmental parameter data and fixed environmental parameter data corresponding to each test group.

8. The intelligent detection and control method for cavern pressure based on an artificial cavern test model as claimed in claim 3 is characterized by: The process of determining the risk tendency environment parameters of the risk collection point is as follows: If the risk collection point is the entity risk collection point, the simulated pressure of each test group in the test group with the same independent variable environmental parameter classification is analyzed for pressure trend, and then the environmental parameter corresponding to the maximum pressure trend is taken as the risk tendency environmental parameter.

9. The intelligent detection and control method for cavern pressure based on an artificial cavern test model as claimed in claim 3 is characterized by: Determining the risk tendency environment parameters of the risk collection point also includes the following process: If the risk collection point is a trend risk collection point, the test groups with the same independent variable environmental parameters will be classified and numbered according to the test order, and the simulated pressure of each test group in the test group set will be compared with the warning pressure, and the test group number that reaches the warning pressure will be screened out. At the same time, the risk proximity index corresponding to each environmental parameter will be calculated, and then the environmental parameter corresponding to the maximum risk proximity index will be screened out as the risk tendency environmental parameter.

Images