A tunnel surrounding rock hazard identification model construction method and tunnel surrounding rock support system

Abstract

The present invention discloses a method for constructing a tunnel surrounding rock hazard identification model and a tunnel surrounding rock support system. The hazard identification model is constructed using data from non-hazardous and absolutely hazardous tunnel surrounding rock, data normalization, and a membership function method. The support system can identify hazards within the tunnel based on the hazard identification model and perform support operations. The system can be used with various types of transportation equipment or detection instruments, requires no human intervention, and offers high safety, providing timely protection for personnel and equipment within the tunnel.

Description

Technical Field

[0001] The invention relates to a drilling auxiliary device, in particular to a drilling equipment support device and a related control system. Background Art

[0002] At present, with the increase in coal mining depth and the expansion of mining scope, the geological conditions of coal mine tunnels are becoming more and more complex. When accidents such as roof collapse, rock burst, tunnel floor heave, and collapse occur during underground coal mine production, causing underground workers to be trapped, large-diameter drilling rescue technology is usually used to establish a rescue channel. At this time, specialized mechanical equipment is required for detection, cleaning and rescue, such as geophysical instrument pushing devices, root canal drills, and lifeboats. Since the borehole wall is not stable enough, there is a possibility of secondary collapse, which is very dangerous. For some relatively stable coal and rock formations, remote control operation is generally performed outside the artificial hole using wired or wireless technology. This method is more suitable for shallow holes, but as the borehole depth increases, personnel cannot perform remote operation. At the same time, specialized equipment is prone to collision with the borehole wall during transportation, causing damage. Manual real-time adjustment of the transportation angle and posture is required, which greatly increases labor intensity and causes low efficiency. At the same time, protection cannot be provided when encountering borehole collapse or sudden danger.

[0003] Existing support devices usually require manual control, are unable to identify dangers in real time, and are prone to control lags, resulting in the inability to provide real-time and effective protection for personnel and equipment in the tunnel. Summary of the Invention

[0004] In view of the defects or shortcomings of the existing technology, the present invention provides a method for constructing a tunnel surrounding rock hazard identification model.

[0005] To this end, the method for constructing a tunnel surrounding rock hazard identification model provided by the present invention includes:

[0006] Step 1: Use the method described in Step 1-Step 3 to obtain a comprehensive data set of surrounding rocks of non-hazardous roadways; use the method described in Step 1-Step 3 to obtain a comprehensive data set of surrounding rocks of absolutely hazardous roadways;

[0007] Step 1: collect rock compressive strength, rock shear strength, rock tensile strength, rock wall image and rock wall obstacle information at n locations in the tunnel; and obtain the corresponding rock wall image contrast and saturation from each rock wall image; obtain the rock compressive strength data set P1 = {p 11 ,p 12 ,···p 1i ,···p 1n}、Rock shear strength data set P2={p 21 ,p 22 ,···p 2i ,···p2n}、Rock tensile strength data set P3={p 31 ,p 32 ,···p 3i ,···o 3n}, rock wall image contrast dataset, rock wall image saturation dataset and rock wall obstacle information dataset; i=1,2,3,…,n; n≥50; p 1i is the rock compressive strength at any part of the surrounding rock in the tunnel, p 2i is the rock shear strength at any part of the surrounding rock in the tunnel, p 3i is the rock tensile strength at any part of the surrounding rock in the tunnel;

[0008] Constructing a comprehensive rock strength dataset O = {p 01 ,p 02 ,···p 0i ,···p 0n};

[0009]

[0010] μ1, μ2, and μ3 are adjustment coefficients, ranging from 0 to 10;

[0011] Step 2: Normalize the data in the rock strength comprehensive dataset, rock wall image contrast dataset, rock wall image saturation dataset, and rock wall obstacle information dataset respectively to obtain the rock strength standardized comprehensive dataset D1 = {d 11 ,d 12 ,···d 1i ,···d 1n}、The image contrast normalized dataset D2 of the rock wall is 21 ,d 22 ,···d 2i ,···d 2n}、Saturation normalized dataset of rock wall images D3={d 31 ,d 32 ,···d 3i ,···d 3n} and rock wall obstacle standardized dataset D4 = {d 41 ,d 42 ,···d 4i ,···d 4n};d 1i is the normalized data of the comprehensive rock strength data of any part of the surrounding rock in the tunnel, d 2i is the normalized data of the rock wall image contrast at any part of the surrounding rock in the tunnel, d 3i is the normalized data of the rock wall image saturation at any part of the surrounding rock in the tunnel, d 4iNormalized data of obstacle information at any part of the surrounding rock in the tunnel;

[0012] Step 3: Fuse each data set in step 2 to construct a comprehensive data set Q = {q1,q2,···q i ,···q n}, Where q i is any data in the comprehensive data set; c1, c2, c3, and c4 are weighting coefficients, each independently taking values ​​from 0 to 1, and c1+c2+c3+c4=1, and c2=3=4; γ1, γ2, γ3, γ4, λ1, λ2, λ3, and λ4 are variable adjustment coefficients, each independently taking values ​​from 0 to 3, and γ2=γ3, λ2=λ3.

[0013] Step 2: Construct the tunnel surrounding rock hazard identification model F:

[0014]

[0015] Where a and b are the maximum and minimum values ​​of the comprehensive data constructed in step 3, respectively; k is the parabola coefficient, which is the ratio of a to b and ranges from 0 to 1; a is the minimum value in the comprehensive data set of surrounding rocks of non-hazardous roadways, and b is the maximum value in the comprehensive data set of surrounding rocks of absolutely hazardous roadways; f is the average value of all data in the comprehensive data set of surrounding rocks of the roadways to be identified, and the comprehensive data set of surrounding rocks of the roadways to be identified is obtained using the method described in steps 1-3;

[0016] When F<0.6, it indicates that the surrounding rock of the roadway to be identified is safe. When the hazard function F≥0.6, it indicates that the surrounding rock of the roadway to be identified is dangerous. When the hazard function F≥0.9, it indicates that the surrounding rock of the roadway to be identified is absolutely dangerous.

[0017] Optionally, μ2=1, μ1=6~10μ2, μ3=5~8μ2.

[0018] The present invention also provides a method for identifying the danger of tunnel surrounding rock. The provided method for identifying the danger of tunnel surrounding rock includes:

[0019] Step 1: Obtain a comprehensive data set of the surrounding rock of the roadway to be identified by using the above steps 1 to 3, and calculate the average value of all data in the comprehensive data set;

[0020] Step 2: Substitute the average value into the model constructed by the above method to identify the danger of the surrounding rock tunnel to be identified.

[0021] Furthermore, after sorting the data in the comprehensive data set from large to small or from small to large in step 1, the first 10% of the data and the last 10% of the data are removed, and then the average value of the remaining data is calculated.

[0022] The present invention also provides a tunnel surrounding rock support system. The tunnel surrounding rock support system provided by the present invention is used to support the tunnel surrounding rock and includes a hydraulic support cylinder. The end of the hydraulic support cylinder supporting the tunnel surrounding rock is equipped with a stress sensor, which is used to detect the force between the support cylinder and the tunnel surrounding rock wall.

[0023] It also includes a tunnel surrounding rock data acquisition system and a surrounding rock hazard identification system; the tunnel surrounding rock data acquisition system is used to collect rock compressive strength, rock shear strength, rock tensile strength, rock wall images and rock wall obstacle information around the support cylinder;

[0024] The surrounding rock hazard identification system identifies the surrounding rock hazard around the support cylinder according to the method described in claim 2, and when the surrounding rock hazard is identified, the support cylinder is controlled to perform support work; when the surrounding rock is identified as absolutely dangerous and the data collected by the stress sensor is greater than the threshold, the personnel and equipment in the tunnel need to be evacuated.

[0025] Furthermore, the surrounding area of ​​the support cylinder is within 1m of the support cylinder.

[0026] The present invention provides a method for constructing a tunnel surrounding rock hazard identification model and a tunnel surrounding rock support system. The hazard identification model is constructed using data from surrounding rock of non-hazardous and absolutely hazardous tunnels, data normalization, and a membership function method. The support system can identify hazards within the tunnel based on the hazard identification model and perform support operations. It can be used with various types of transportation equipment or detection instruments, requires no human intervention, and offers high safety, providing timely protection for personnel and equipment within the tunnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is an example of a safe (non-hazardous) roadway surrounding rock. Its characteristics are that the roadway floor is basically flat and free of breakage; the rock structure is complete and free of obstacles, and the roadway is in good stress condition.

[0028] Figure 2 This is an example of dangerous roadway surrounding rock. Its characteristics are that the roadway floor is partially bulging, the rock structure is damaged, some rocks are broken, some obstacles appear, and the stress state of the roadway changes. This type of roadway surrounding rock poses certain safety risks.

[0029] Figure 3 This is an example of an absolutely dangerous roadway surrounding rock. Its characteristics are that the roadway roof bulges, forming a partial net bag, a large number of obstacles bulge, rock fragmentation, rock structure damage, and changes in the roadway stress state.

[0030] Figure 4This is another example of an absolutely dangerous roadway surrounding rock. It is characterized by bulging roadway side panels, severe bulging of obstacles, and extensive crushing. The rock structure is damaged, and there are changes in the roadway's stress state. DETAILED DESCRIPTION

[0031] Unless otherwise specified, the scientific and technical terms used herein are understood according to the knowledge of ordinary technicians in the relevant fields.

[0032] The safe (non-dangerous) tunnel surrounding rock mentioned in this article generally refers to: the tunnel roof, side walls and floor are flat, without cracks, etc., the surrounding rock structure is complete, the stress state is good, and various types of transportation equipment can pass safely. For example, a photo of the safe tunnel surrounding rock collected by the bottom plate binocular camera at a certain moment is as follows: Figure 1 shown.

[0033] The dangerous tunnel surrounding rock generally refers to: the tunnel roof, side walls and floor are partially broken, cracked, etc., the structure of the surrounding rock layer is partially damaged, and there are some obstacles that are raised or sunken. There are certain safety risks for various types of transportation equipment to pass through. For example, a photo of the dangerous tunnel surrounding rock captured by the bottom plate binocular camera at a certain moment is as follows Figure 2 shown.

[0034] The absolutely dangerous tunnel surrounding rock refers to the tunnel roof, side plate or floor with serious bulging, a large number of cracks and fissures, the tunnel contact surface is subject to large rock forces, the rock inside the surrounding rock layer is broken or has obvious gaps, there is stress concentration, the obstacles are obviously raised or sunken, and there is the possibility of impact ground pressure. There is a high risk when various types of transportation equipment pass through. Common types of binocular cameras collect photos such as Figure 3-4 shown.

[0035] The surrounding rock obstacles include heaves, ribs, net pockets, and / or loose tops. The obstacle information is the protruding or recessed dimensions of the obstacles, as collected by sensors. In a specific embodiment, lidar measurements can be used. In a specific embodiment, the compressive and tensile strengths of the surrounding rock can be measured using an intrusion meter; the shear strength can be measured using a shear meter; and the surrounding rock wall image can be measured using a binocular camera.

[0036] In the scheme of the present invention, it is necessary to perform data normalization on each data set to ensure that there is no mutual interference between the data and that each data has no dimension effect. The rock strength comprehensive data set P = {p 01 ,p 02 ,···p 0i ,···p 0n} and rock strength standardized comprehensive data set D1 = {d 11 ,d 12 ,···d 1i ,···d 1n} as an example, the standardized comprehensive data value of each rock strength can be obtained by the following formula:

[0037]

[0038] In the scheme of the present invention, μ1, μ2, and μ3 are adjustment coefficients, ranging from 0 to 10, and are set independently according to the importance of each data value; it is recommended that μ2 = 1, μ1 = 6 to 10μ2, and μ3 = 5 to 8μ2. c1, c2, c3, and c4 are weighting coefficients, ranging from 0 to 1, and summing to 1, c2 = c3 = c4. In the specific scheme, the values ​​can be determined based on the importance of each data and can be set independently according to the specific tunnel construction conditions. γ1, γ2, γ3, γ4 and λ1, λ2, λ3, and λ4 are variable adjustment coefficients, ranging from 0 to 3, with γ2 = γ3 and λ2 = λ3. In the specific scheme, the variable adjustment coefficients can be determined based on the importance of each data and can be set independently according to the specific tunnel construction conditions.

[0039] Example 1:

[0040] This specific example uses the method of the present invention to build an identification model by collecting data on surrounding rock of non-dangerous lanes and absolutely dangerous lanes in a lane of the Huainan Guqiao Coal Mine:

[0041] Specifically, 10 shear meters, 10 indenters, 10 binocular cameras and 10 lidars were set up in the tunnel surrounding rock; 3 were set up on the tunnel roof and floor, and 2 were set up on each of the two side walls; data were collected at intervals of 0.1m, and 100 sets of rock strength data were obtained.

[0042] It can be concluded that the rock shear strength data set P1 of the surrounding rock of the non-dangerous tunnel and the absolutely dangerous tunnel varies in the range of 0.10MPa~0.20MPa and 0.38MPa~0.50MPa respectively, the rock compressive strength data set P2 varies in the range of 5.52MPa~8.49MPa and 18.44MPa~23.34MPa respectively, and the rock tensile strength P3 varies in the range of 0.05MPa~0.10MPa and 0.20MPa~0.35MPa respectively.

[0043] Taking μ1=8, μ2=1, and μ3=7, the variation ranges of the comprehensive data sets P of rock strength without danger and absolute danger are obtained by formula 1, which are 5.36MPa~8.45MPa and 15.45MPa~18.85MPa respectively; after standardizing the comprehensive data set P, the standardized comprehensive data sets D1 of rock strength without danger and absolute danger are obtained, and the variation ranges are both 0~1.

[0044] The image contrast and image saturation of 100 non-hazardous and absolutely hazardous tunnel rock walls were measured using a binocular camera. The image contrast ranges were 85.5-90.5 and 95.8-99.8, respectively, and the image saturation ranges were 70.8%-75.5% and 90.7%-96.6%, respectively. The data were standardized using the normalization method to obtain a standardized image contrast dataset D2 for non-hazardous and absolutely hazardous tunnel rock walls, with a range of 0-1. The image saturation standardized dataset D3 also had a range of 0-1.

[0045] The sizes of 100 non-hazardous and absolutely hazardous roadway rock wall obstacles were measured using lidar. The sizes of the roadway rock wall obstacles ranged from -20 mm to 20 mm and from -400 mm to 300 mm, respectively. The data were standardized using the normalization method, and the standardized data set D4 of the roadway wall obstacle sizes ranged from 0 to 1.

[0046] Taking the weighting coefficients c1=0.4, c2=c3=c4=0.2, γ1=2, γ2=γ3=1, γ4=1.5, λ1=2, λ2=λ3=1, λ4=1, the comprehensive data set Q of the surrounding rock of the non-dangerous tunnel and the absolutely dangerous tunnel is obtained, with the variation ranges of 0~0.95 and 0~1.68 respectively. The critical values ​​of the comprehensive data sets of non-dangerous and absolutely dangerous are selected, namely a=0.88 and b=1.42, and k=0.62 is calculated at the same time.

[0047] Example 2:

[0048] This embodiment uses the model constructed in Example 1 and the method of the present invention to identify the danger of a hydraulic cylinder support device at a certain location in a certain tunnel:

[0049] Data collection was performed within a fixed distance of 1m from the front end of the hydraulic cylinder support device. A total of 10 shear gauges, indenters, binocular cameras, and lidar sensors were installed across the entire cross-section. Three were installed on the tunnel roof and floor, and two on each of the two sidewalls. Data was collected at 0.1m intervals, yielding 100 sets of rock compressive strength data. Stress sensors were also installed on the support cylinders, with the stress sensor's threshold set at 90% of the cylinder's maximum load capacity. In this example, a 140 / 80 cylinder with an operating pressure of 16MPa was selected, resulting in a maximum load capacity of 246.3kN and a force threshold of 221.67kN.

[0050] The rock shear strength dataset P1 varies in the range of 0.22MPa to 0.32MPa, the rock compressive strength dataset P2 varies in the range of 10.52MPa to 15.44MPa, and the rock tensile strength P3 varies in the range of 0.12MPa to 0.17MPa. Taking μ1=8, μ2=1, and μ3=7, the rock strength comprehensive dataset P varies in the range of 9.37MPa to 11.39MPa according to formula 1. After standardizing the comprehensive dataset P, the rock strength standardized comprehensive dataset D1 is obtained, which varies in the range of 0 to 1.

[0051] The contrast and saturation of 100 tunnel rock wall images were measured using a binocular camera. The image contrast ranged from 90.9 to 97.2, and the image saturation ranged from 78.5% to 95.5%. The data were standardized using the normalization method to produce a standardized dataset D2 for tunnel rock wall image contrast, with a range of 0 to 1, and a standardized dataset D3 for image saturation, with a range of 0 to 1. The sizes of 100 tunnel rock wall obstacles were measured using a lidar radar. The sizes of the obstacles ranged from -140 mm to 230 mm. The data were standardized using the normalization method to produce a standardized dataset D4 for tunnel rock wall obstacle sizes, with a range of 0 to 1.

[0052] Taking weighting coefficients c1=0.4, c2=c3=c4=0.2, γ1=2, γ2=γ3=1, γ4=1.5, λ1=2, λ2=λ3=1, λ4=1, we get the comprehensive data set Q, which varies from 0 to 1.34. After removing the first 10% and the last 10% of the data, the average of the remaining data sets is solved to get the judgment value f of the comprehensive data set, which is 1.12. The judgment value f can be used to make a hazard judgment.

[0053] The hazard function F is calculated by formula (2) and is 0.54, indicating that there is no danger at this time and the support device can work normally.

[0054] Similarly, the danger can be judged in real time to guide the support device to work. The entire transportation process is smooth, and the danger function F of the entire transportation process is less than 0.6. There is basically no tunnel danger, which can effectively guide the work of the protection device.

Claims

1. A method for constructing a tunnel surrounding rock hazard identification model, characterized in that: Methods include: Step 1: Use the method described in Step 1-Step 3 to obtain a comprehensive data set of surrounding rocks of non-hazardous roadways; use the method described in Step 1-Step 3 to obtain a comprehensive data set of surrounding rocks of absolutely hazardous roadways; Step 1: Collect surrounding rock in the tunnel n The rock compressive strength, rock shear strength, rock tensile strength, rock wall image and rock wall obstacle information of each part are obtained; and the corresponding rock wall image contrast and saturation are obtained from each rock wall image; the rock compressive strength data set is obtained , rock shear strength dataset , rock tensile strength dataset , rock wall image contrast dataset, rock wall image saturation dataset and rock wall obstacle information dataset; i= 1,2,3,…, n ; n ≥50; is the rock compressive strength at any part of the surrounding rock in the tunnel, is the rock shear strength at any part of the surrounding rock in the tunnel, is the rock tensile strength at any part of the surrounding rock in the tunnel; Constructing a comprehensive rock strength dataset ; （1） 、 、 is the adjustment coefficient, ranging from 0 to 10; Step 2: Normalize the data in the rock strength comprehensive dataset, rock wall image contrast dataset, rock wall image saturation dataset, and rock wall obstacle information dataset to obtain the rock strength standardized comprehensive dataset. , rock wall image contrast normalization dataset , rock wall image saturation standardization dataset and rock wall obstacle standardized dataset ; It is the normalized data of the comprehensive rock strength data of any part of the surrounding rock in the tunnel. is the normalized data of the rock wall image contrast at any part of the surrounding rock in the tunnel, is the normalized data of the rock wall image saturation at any part of the surrounding rock in the tunnel, Normalized data of obstacle information at any part of the surrounding rock in the tunnel; Step 3: Fusion of each dataset in step 2 to construct a comprehensive dataset , Where, is any data in the comprehensive dataset; 、 、 、 are weighted coefficients, each independently ranging from 0 to 1, and ,at the same time ; 、 、 、 、 、 are variable adjustment coefficients, each with an independent value of 0 to 3, and = , = ; Step 2: Construct a tunnel surrounding rock hazard identification model : （2） Where a and b are the maximum and minimum values ​​of the comprehensive data constructed in step 3, respectively; k is the parabola coefficient, which is the ratio of a to b and ranges from 0 to 1; a takes the minimum value in the comprehensive data set of surrounding rock of non-dangerous roadways, and b takes the maximum value in the comprehensive data set of surrounding rock of absolutely dangerous roadways; is the average value of all data in the comprehensive data set of the surrounding rock of the roadway to be identified, and the comprehensive data set of the surrounding rock of the roadway to be identified is obtained using the method described in Step 1-3; when When the hazard function When the hazard function When , it indicates that the surrounding rock of the roadway to be identified is absolutely dangerous.

2. The method for constructing a tunnel surrounding rock hazard identification model according to claim 1, wherein: 。 3. A method for identifying the danger of surrounding rock in a roadway, characterized in that: Methods include: Step 1, using Step 1 to Step 3 described in claim 1 to obtain a comprehensive data set of the surrounding rock of the roadway to be identified, and calculating the average value of all data in the comprehensive data set; Step 2: bringing the average value into the model constructed in claim 1 to identify the danger of the surrounding rock tunnel to be identified.

4. The method for identifying the danger of surrounding rock in a tunnel according to claim 3, wherein: In step 1, after sorting the data in the comprehensive data set from large to small or small to large, the first 10% of the data and the last 10% of the data are removed, and then the average value of the remaining data is calculated.

5. A tunnel surrounding rock support system for supporting tunnel surrounding rock, comprising a hydraulic support cylinder, characterized in that: A stress sensor is installed at the end of the hydraulic support cylinder supporting the tunnel surrounding rock, and the stress sensor is used to collect the magnitude of the force between the support cylinder and the tunnel surrounding rock wall; It also includes a tunnel surrounding rock data acquisition system and a surrounding rock hazard identification system; the tunnel surrounding rock data acquisition system is used to collect rock compressive strength, rock shear strength, rock tensile strength, rock wall images and rock wall obstacle information around the support cylinder; The surrounding rock hazard identification system identifies the surrounding rock hazard around the support cylinder according to the method described in claim 2, and when the surrounding rock hazard is identified, the support cylinder is controlled to perform support work; when the surrounding rock is identified as absolutely dangerous and the data collected by the stress sensor is greater than the threshold, the personnel and equipment in the tunnel need to be evacuated.

6. The tunnel surrounding rock support system according to claim 5, characterized in that: The area around the support cylinder is within 1m of the support cylinder.

Images