While-drilling advanced identification method for rock burst and rockburst dynamic disaster dangerous area
By constructing drilling in the area to be excavated and combining panoramic drilling peeping and machine learning methods, a relationship between drilling multi-parameters and ground stress is established to predict the dynamic disaster hazard zones in underground projects, and resource waste and capacity crises caused by blind design are solved, and advanced identification of dynamic disaster hazard zones is achieved.
Patent Information
- Application Number
- CN202510573620.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-06
- Publication Date
- 2025-07-18
AI Technical Summary
The existing technology cannot effectively identify the hazardous areas of impact ground pressure and rock burst power in unknown areas in underground projects, resulting in blind design and mining, resulting in resource waste and capacity crisis.
By constructing drilling holes into the area to be excavated, drilling parameters are recorded in real time and crack information is obtained using a panoramic drilling peeper. Combining three-axis loading tests and machine learning methods, a correspondence between drilling multi-parameters and ground stress and intensity parameters is established, dynamic disaster risk levels are predicted, and microseismic monitoring verification results are used for correction.
It realizes advanced identification of impact ground pressure and rock burst power disaster hazard areas, avoids blind mining, improves the accuracy and production efficiency of engineering design, and solves the resource waste and capacity crisis caused by blind design.
Smart Images
Figure CN120331754A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of rock burst and rockburst risk assessment, and specifically to a method for drilling-while-drilling advanced identification of dynamic disaster danger zones of rock burst and rockburst. Background Art
[0002] Dynamic disasters in underground engineering, such as rock burst (the medium is mainly coal and it occurs in coal mines) and rockburst (the medium is rock and it occurs in deep tunnels and chambers), have significant characteristics such as strong suddenness, complex disaster-causing mechanisms, and high concealment of hazard sources. Currently, rock burst and rockburst dynamic disasters are facing two major dilemmas:
[0003] I. The problem of blind design and blind excavation
[0004] Existing in-situ stress testing technologies (such as the hollow inclusion method) can only obtain point data in the areas that have been exposed by humans. Numerical simulation is restricted by the spatial heterogeneity of coal and rock parameters and the idealization of boundary conditions, and the prediction reliability is insufficient; technologies such as microseismic and stress online monitoring essentially belong to "post-response" monitoring and cannot achieve forward-looking perception of the stress state in the area to be excavated. This lack of technology has led to the dependence of roadway driving design on empirical inferences, and often the roadway has to be abandoned due to the exposure of high-stress areas or frequent occurrence of dynamic disasters after driving, resulting in huge waste of thousands of meters of roadway engineering and resources.
[0005] II. The production capacity crisis caused by the lack of basis for safety control
[0006] Deep development faces a dual policy dilemma: at the national level, the exploitation of coal resources deeper than one kilometer is strictly restricted based on safety considerations. This control mode of "insufficient data-driven - dominated by empirical decision-making" has led to the imbalance of mine succession and a cliff-like decline in production capacity, seriously restricting the implementation of China's strategy of "marching towards the deep earth". Summary of the Invention
[0007] The purpose of the present invention is to address the problem that the lack of the ability to identify the risk of dynamic disasters in unknown areas in existing underground engineering leads to the difficulty in determining the dynamic disaster danger zone in advance, resulting in blind excavation and production capacity crisis. The invention provides a method for drilling-while-drilling advanced identification of dynamic disaster danger zones of rock burst and rockburst.
[0008] The technical solution of the present invention is as follows:
[0009] A method for drilling-while-drilling advanced identification of dynamic disaster danger zones of rock burst and rockburst, characterized by comprising the following steps:
[0010] (1) Drill a borehole in the area to be excavated, and record the drilling time, displacement, torque, thrust, drilling speed, rotation speed, and vibration speed curve in real time; use a panoramic borehole peephole to conduct high-definition imaging of the hole, and analyze the obtained planar expanded image of the borehole wall to obtain the degree of fracture development and size.
[0011] (2) Combining laboratory tests and numerical simulations, conduct orthogonal triaxial loading drilling tests, and record various drilling parameters for specimens with different strength parameters and different loading confining pressures, including: drilling torque, thrust, rotational speed, drilling speed, drill cuttings, and multi-directional vibration signals;
[0012] (3) Using mathematical statistics or machine learning methods, establish the corresponding relationships between multi-parameter information in drilling and in-situ stress and strength parameters:
[0013] E v = f1(T, n, F, v, S v )
[0014] σ v = f2(T, n, F, v, Sv)
[0015] σ v = f3(σ1, σ2, σ3)
[0016] where: E v is the strength; σ v is the equivalent stress, σ1 is the vertical stress, σ2 and σ2 are the confining pressures in two directions, T is the torque, n is the rotational speed, F is the thrust, v is the drilling speed, and S v is the vibration velocity.
[0017] (4) Establish the following discrimination criteria for the impact hazard level:
[0018]
[0019] where: k = f(E v ), is the fracture coefficient. According to the degree of coal body fragmentation in the peep image, different regions of the borehole are divided into a complete zone, a fracture zone, and a broken zone. The region without fractures or with minor fractures is regarded as the complete zone, and the crushed stone region is regarded as the broken zone. For the rock mass in the complete zone, k takes 1, for the rock mass in the broken zone, k takes 0, and for the rock mass in the fracture zone, k is comprehensively taken as 0 - 1 according to the number and size of unit fractures. σ v is the equivalent stress, σ1 is the vertical stress, σ2 and σ2 are the confining pressures in two directions; T is the torque, n is the rotational speed, F is the thrust, v is the drilling speed, and S v is the vibration velocity.
[0020] (5) According to step 4, segment the collected data with the sliding window length, and directly predict the dynamic disaster hazard levels at different positions of the borehole in step 1, and draw a prediction map of the dynamic disaster hazard area.
[0021] (6) During the actual tunneling process, verify the prediction results based on microseismic, borehole cuttings, and on-site observations of dynamic disaster situations. According to the verification results, correct the corresponding relationships between the drilling parameters, fracture characteristics, and dynamic disaster hazard levels in step 6. The judgment criteria are as follows:
[0022] 1) When a dynamic disaster occurs in a certain area, it is directly judged as a red warning area, and the prediction effect is judged according to the coincidence rate between the disaster occurrence area and the predicted dangerous area.
[0023] 2) Use microseismic monitoring to record the energy level and occurrence location of strong mine tremors. When a mine tremor with an energy level greater than the red threshold appears, the area where the mine tremor occurs is directly judged as a red warning area; when a mine tremor with an energy level greater than the yellow threshold appears in a certain area, the area where the mine tremor occurs is directly judged as a yellow warning area; the prediction effect is judged according to the coincidence rate and level between the mine tremor occurrence area and the predicted dangerous area.
[0024] The boreholes include boreholes with a length of 20m to 300m constructed every certain distance during the tunneling process, and also include boreholes with a length greater than 300m directly constructed by using kilometer directional drilling.
[0025] The drilling displacement, torque, thrust, drilling speed, rotation speed, and vibration speed are obtained by sensors installed on the drilling rig and drill pipes.
[0026] The size of the microcracks is less than 50mm.
[0027] The strength parameters are cohesion C and internal friction angle
[0028] The length of the sliding window is 5 to 40m.
[0029] The beneficial effects of the present invention are:
[0030] By constructing boreholes in the position area, the present invention obtains curves such as drilling time, displacement, torque, thrust, drilling speed, rotation speed, and vibration speed during drilling, and obtains the fracture development characteristics by using a panoramic borehole peephole. Through the corresponding relationship between the multi-parameter information during drilling and the in-situ stress and strength parameters, as well as the dynamic disaster discrimination criterion, the dynamic disaster risk level at different positions of the borehole is predicted, and a prediction map of the dynamic disaster risk area is drawn, so as to realize the advanced identification of the impact ground pressure and rock burst dynamic disaster risk area during drilling, and solve the problems of blind excavation and production capacity crisis caused by the difficulty in determining the dynamic disaster risk area in advance. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] Figure 1 It is a flow chart of the method for advanced identification of the impact ground pressure and rock burst dynamic disaster risk area during drilling.
[0032] Figure 2 It is a schematic diagram of constructing a borehole in the area to be mined.
[0033] Figure 3 It is a schematic diagram of constructing a directional long borehole (borehole with a length greater than 300m) in the area to be mined.
[0034] Figure 4It is a schematic diagram of drilling data curves (torque, drilling speed, rotation speed, etc.).
[0035] Figure 5 It is a schematic diagram of the triaxial loading drilling test of the present invention.
[0036] Figure 6 It is a schematic diagram of predicting in-situ stress and strength of rock mass by machine learning method, showing the combination ways of convolutional neural network (CNN), long short-term memory network (LSTM) and BP neural network or random forest.
[0037] Figure 7 It is a schematic diagram of the prediction results of the dynamic disaster risk levels at different positions of the borehole. Detailed implementation manners
[0038] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0039] As Figure 1-7 shown.
[0040] A method for real-time ahead identification of dynamic disaster risk areas of rock bursts and rock bursts while drilling, the process of which is as Figure 1 shown, and it includes the following steps:
[0041] (1) Figure 2 Construct ordinary boreholes (hole depth 20m - 300m) in the area to be mined or excavated. In the figure, 1 is the drilling rig and 2 is the borehole. Figure 3 Construct directional long boreholes (boreholes greater than 300m) in the area to be mined or excavated. Record the drilling time, displacement, and curves of torque, thrust, drilling speed, rotation speed, and vibration speed in real time ( Figure 4 ); use a panoramic borehole peephole instrument to conduct high-definition imaging of the hole, and analyze the obtained planar unfolded image of the borehole wall to obtain the degree and size of crack development.
[0042] (2) Figure 5 In, in combination with laboratory tests and numerical simulations, carry out triaxial loading drilling orthogonal tests, and record various drilling parameters for specimens with different strength parameters (cohesion C, internal friction angle ) and different loading confining pressures, including: drilling torque, thrust, rotation speed, drilling speed, drill cuttings, multi-directional vibration signals.
[0043] For example: configure specimens with 4 strength parameters (different cohesion C, internal friction angle ). The triaxial loading drilling orthogonal test design considers 3 factors such as initial stresses σ1, σ2, σ3, and takes 4 levels. Therefore, the orthogonal test is designed with 4 factors and 4 levels (represented by A, B, C, and D respectively), and a total of 16 groups of tests need to be arranged, as shown in Table 1.
[0044] Table 1 Drilling orthogonal test design scheme
[0045]
[0046] In the table: A1, A2, A3, A4 represent four levels of σ3 respectively; B1, B2, B3, B4 represent four levels of σ2 respectively; C1, C2, C3, C4 represent four levels of σ1 respectively; D1, D2, D3, D4 represent four levels of intensity parameters respectively.
[0047] (3) Using mathematical statistics or machine learning methods, establish the corresponding relationship between drilling multi-parameter information and ground stress and strength parameters. The machine learning model includes but is not limited to convolutional neural network (CNN), long short-term memory network (LSTM), BP neural network, etc.;
[0048] For example: Figure 6 The machine learning process includes: 1) data preprocessing: denoising and normalizing the drilling multi-parameter data to eliminate dimensional differences; 2) feature extraction: using convolutional neural network (CNN) to extract local features of logging curves; 3) time series modeling: using long short-term memory network (LSTM) to capture the time series features of logging curves; 4) ground stress and strength prediction: using BP neural network or random forest to predict ground stress and strength.
[0049] (4) The criteria for determining the impact hazard level are as follows:
[0050]
[0051] Where: k = f(E v ), is the fracture coefficient. According to the degree of coal body fragmentation in the peek image, different areas of the borehole are divided into intact zone, fracture zone and broken zone. The area without fractures or small fractures is regarded as the intact zone, and the gravel area is regarded as the broken zone. The intact zone rock mass k is 1, the broken zone rock mass k is 0, and the fracture zone rock mass is 0 to 1 according to the number and size of unit fractures. v is the equivalent stress, σ1 is the vertical stress, σ2 and σ2 are the confining pressures in two directions; T is the torque, n is the rotation speed, F is the thrust, v is the drilling speed, S v is the vibration speed.
[0052] (5) The collected data is processed in segments using a sliding window (length 5 to 40 m), the dynamic disaster hazard level at different locations of the borehole in step 1 is directly predicted, and a dynamic disaster hazard area prediction map is drawn.
[0053] like: Figure 7 The diagram is a schematic diagram of the prediction results of the dynamic disaster hazard level at different locations of the borehole. In the figure, 3 is a non-hazardous area, 4 is a yellow warning area, 5 is a red warning area, and 6 is a broken zone. The broken zone is not easy to accumulate energy and can be regarded as a non-hazardous area.
[0054] (6) During the actual tunneling process, the prediction results are verified based on microseismic monitoring, drill cuttings, and on-site observation of dynamic disaster conditions. According to the verification results, the corresponding relationships between the drilling parameters, fracture characteristics, and dynamic disaster risk levels in step 6 are corrected. The judgment criteria are as follows:
[0055] 1) If a dynamic disaster occurs in a certain area, it is directly judged as a red warning area, and the prediction effect is judged based on the coincidence rate between the disaster occurrence area and the predicted danger area.
[0056] 2) Use microseismic monitoring to record the energy level and occurrence location of strong mine tremors. For example, if a mine tremor with an energy level of E MAX ≥10 5 J (red threshold) appears, the area where the mine tremor occurs is directly judged as a red warning area; for example, if a mine tremor with an energy level of E MAX ≥10 4 J (yellow threshold) appears in a certain area, the area where the mine tremor occurs is directly judged as a yellow warning area; the prediction effect is judged based on the coincidence rate and level between the mine tremor occurrence area and the predicted danger area.
[0057] 3) The prediction effect is judged based on the coincidence rate between the predicted danger area and the actual danger area. For example: if the coincidence rate between the prediction result and the actual danger area and level exceeds 70%, the prediction method is judged to have medium effect; if it exceeds 80%, the prediction method is judged to have good effect; if it exceeds 90%, the prediction method is judged to have excellent effect; if it is less than 70%, the prediction method is judged to have poor effect.
[0058] The parts not involved in the present invention are the same as the prior art or can be implemented by the prior art.
Claims
1. A method for drilling-while-advancing and ahead-of-time identification of dynamic disaster danger areas of rock bursts and rock bumps, characterized in that: It includes the following steps: (1) Drill holes in the area to be mined, and record the drilling time, displacement, torque, thrust, drilling speed, rotation speed, and vibration speed curve in real time; use a panoramic borehole peephole to conduct high-definition imaging of the hole, analyze the obtained planar expansion image of the borehole wall, and obtain the degree of fissure development and size; (2) Combine laboratory tests and numerical simulations to carry out a three-axis loading drilling orthogonal test, and record various drilling parameters for specimens with different strength parameters and different loading confining pressures; (3) Use mathematical statistics or machine learning methods to establish the corresponding relationship between multi-parameter information of drilling and in-situ stress and strength parameters; E v = f1(T, n, F, v, S v ) σ v = f2(T, n, F, v, Sv) σ v = f3(σ1, σ2, σ3) Where: E v is the strength; σ v is the effective stress, σ1 is the vertical stress, σ2 and σ2 are the confining pressures in two directions, T is the torque, n is the rotational speed, F is the thrust, v is the drilling speed, S v is the vibration velocity; (4) Establish the following discrimination criteria for the impact hazard level: where: k = f(E v ), is the fracture coefficient; according to the degree of coal body fragmentation in the peep image, different regions of the borehole are divided into the intact zone, the fracture zone, and the broken zone; the region without fractures or with minor fractures is regarded as the intact zone, and the crushed stone region is regarded as the broken zone; for the intact zone rock mass, k is taken as 1, for the broken zone rock mass, k is taken as 0, and for the fracture zone rock mass, k is comprehensively taken as 0-1 according to the number and size of unit fractures; σ v is the effective stress, σ1 is the vertical stress, σ2, σ2 are the confining pressures in two directions; T is the torque, n is the rotational speed, F is the thrust, v is the drilling speed, S v is the vibration velocity; (5) According to step 4, segment the collected data with the sliding window length, directly predict the dynamic disaster hazard levels at different positions of the boreholes in step 1, and draw a prediction map of the dynamic disaster hazard area; (6) During the actual tunneling process, verify the prediction results based on microseismic, borehole cuttings, and on-site dynamic disaster observations. According to the verification results, correct the corresponding relationship between the drilling parameters, fissure characteristics, and dynamic disaster hazard levels in step 6. The judgment criteria are as follows: 1) If a dynamic disaster occurs in a certain area, it is directly judged as a red warning area, and the prediction effect is judged according to the coincidence rate between the disaster occurrence area and the predicted dangerous area; 2) Use microseismic monitoring to record the energy level and occurrence location of strong mine tremors. If a mine tremor with an energy level greater than the red threshold appears, the area where the mine tremor occurs is directly judged as a red warning area; if a mine tremor with an energy level greater than the yellow threshold appears in a certain area, the area where the mine tremor occurs is directly judged as a yellow warning area; the prediction effect is judged according to the coincidence rate and level between the mine tremor occurrence area and the predicted dangerous area.
2. The recognition method according to claim 1, wherein: The boreholes include boreholes with a length of 20m to 300m drilled every certain distance during the tunneling process, and also include boreholes with a length greater than 300m directly drilled using a kilometer directional drill.
3. The recognition method according to claim 1, characterized in that: The drilling displacement, torque, thrust, drilling speed, rotation speed, and vibration speed are obtained by sensors installed on the drill rig and drill pipe.
4. The recognition method according to claim 1, wherein: The size of the micro-fissures is less than 50mm.
5. The recognition method according to claim 1, characterized in that: The strength parameters described are cohesion C and angle of internal friction 6. The recognition method according to claim 1, characterized in that: The length of the sliding window is 5 to 40m.
7. The recognition method according to claim 1, wherein: The drilling parameters include: drilling torque, thrust, rotation speed, drilling speed, drill cuttings, and multi-directional vibration signals.