Real-time informatization blasting control method based on image recognition

Abstract

The invention relates to the field of tunnel engineering over-excavation control, and discloses a real-time informatization blasting control method based on image recognition, which comprises the following steps: S1, establishing a three-dimensional coordinate system; s2, taking a photo of a tunnel face in the tunnel and determining physical and mechanical information of a rock mass in the tunnel; s3, performing geological survey on the tunnel face picture; s4, carrying out informatization advanced pre-reinforcement on the local broken section according to the fracture occurrence; s5, drilling parameters of the drill jumbo are determined according to the overall and local surrounding rock grading condition of the tunnel face, the geological condition of the tunnel face and the reinforcing effect, and drilling is conducted; s6, obtaining coordinates of the three-dimensional point cloud data, converted into the three-dimensional coordinate system in the S1, of the tunnel contour after blasting through three-dimensional laser scanning; s7, calculating a linear over-excavation value and a cross section over-excavation area value of the position and an over-excavation volume value of a blasting section; s8, the drilling parameters are optimized; and S9, returning to the step S5 in combination with the optimized blasting drilling parameters. The method can reduce the average overbreak value of the comprehensive linear shape.

Description

Technical Field

[0001] The invention relates to the field of over-excavation control in tunnel engineering, and in particular to a real-time information-based blasting control method based on image recognition. Background Art

[0002] With the increasingly fierce competition in the engineering construction market and the continuous enhancement of environmental awareness, the problem of improving the quality and efficiency of tunnel engineering construction will become increasingly prominent, and the demand for enterprises to reduce costs and increase efficiency is increasing. Judging from the statistical data of tunnel projects currently under construction, the problem of over-excavation and over-consumption in tunnel projects is prominent. According to relevant statistics, the over-consumption rate of shotcrete in particular has even exceeded 130%, which has greatly increased the construction cost and carbon emissions, and has had an adverse impact on the efficiency improvement and environmental protection of enterprises. It has become a major problem that needs to be urgently solved in domestic tunnel engineering construction. The over-excavation and over-consumption of tunnel projects are mainly caused by over-excavation of tunnels and over-consumption of shotcrete, and over-excavation of tunnels accounts for 80% of the total over-consumption. Reducing the over-consumption caused by over-excavation can not only improve the tunnel efficiency and reduce the project cost, but also reduce the carbon emissions of tunnel construction, which will bring both economic and social benefits to enterprises.

[0003] The reason why the over-consumption problem has not been well solved for a long time is mainly due to the lack of experienced geological engineers on site, resulting in large geological over-excavation; insufficient performance of drilling equipment causing over-excavation and over-consumption; insufficient blasting technology causing over-excavation and over-consumption; lack of on-site real-time over-excavation and over-consumption detection and process management measures, and the over-excavation is displayed in two-dimensional form during the process, which cannot continuously and accurately feedback the blasting quality. Therefore, how to automatically identify the geology of the tunnel face based on image recognition, how to control over-excavation in real time based on 3D laser scanning technology, how to improve the intelligent drilling system of the rock drilling rig to control the drilling quality, and how to control over-excavation technology through information-based precise pre-reinforcement are urgently needed. Summary of the invention

[0004] The present invention provides a real-time information-based blasting control method based on image recognition. Compared with traditional blasting control technology, the method can automatically identify the geology of the tunnel face, adopt three-dimensional laser scanning, improve the intelligent drilling system of the rock drilling rig, and perform information-based precise pre-reinforcement to perform real-time control of over-excavation, ultimately achieving effective control of blasting over-excavation and reducing the average over-excavation value of the comprehensive linear shape.

[0005] The present invention is achieved through the following technical solutions:

[0006] A real-time information-based blasting control method based on image recognition, comprising:

[0007] S1. Establish a three-dimensional coordinate system to obtain the three-dimensional coordinates of each mileage and each three-dimensional line segment of the tunnel to be excavated;

[0008] S2, taking photos of the tunnel face in the tunnel and determining the physical and mechanical information of the rock mass in the tunnel, extracting the tunnel face information from the tunnel face photos based on image recognition, and calculating the BQ index by combining the tunnel face information with the physical and mechanical information of the rock mass to perform overall and local surrounding rock classification of the tunnel face;

[0009] S3, conducting a geological survey on the tunnel face photo, and judging whether the surrounding rock is broken according to the tunnel face. If not, jump to S5, otherwise jump to S4;

[0010] S4. Taking the over-excavation and under-excavation technical indicators as constraints, carry out information-based advanced pre-reinforcement of local broken sections based on the three-dimensional scanning results and the fracture occurrence;

[0011] S5. Determine the drilling parameters of the drilling rig and drill holes according to the overall and local surrounding rock classification of the face, the geological conditions of the face and the reinforcement effect;

[0012] S6. After the blasting construction is carried out, the coordinates of the three-dimensional point cloud data of the tunnel contour after the blasting converted into the three-dimensional coordinate system of S1 are obtained by three-dimensional laser scanning;

[0013] S7, comparing the coordinates of the same point at the same position before and after the blasting, calculating the linear over-excavation value, the cross-sectional over-excavation area value and the over-excavation volume value of the blasting section at the position, so as to evaluate the over-excavation and under-excavation quality of the excavation surface of the blasting;

[0014] S8. According to the over-excavation of the excavation face during the blasting and the surrounding rock geological conditions of the face after the blasting, the drilling parameters are optimized;

[0015] S9. Return to S5 based on the optimized blasting drilling parameters, conduct the next cycle drilling test on the optimized overbreak situation and evaluate the overbreak and underbreak quality, and continue to further optimize the drilling parameters to reduce the overbreak situation.

[0016] As an optimization, the specific steps of S1 are:

[0017] S1.1. Establish a three-dimensional coordinate system, obtain the cross section of the two-dimensional tunnel design, and determine the coordinate origin according to the actual situation, where the X coordinate is the tunnel span direction, the Y coordinate is the tunnel height direction, and the Z coordinate is the tunnel line longitudinal direction;

[0018] S1.2, converting the circular arc of the cross section of the tunnel into a plurality of two-dimensional line segments to obtain the two-dimensional coordinates of the plurality of two-dimensional line segments;

[0019] S1.3. Query the route map of the tunnel, and extract the horizontal curve parameters, vertical curve parameters and widening offset value of the tunnel;

[0020] S1.4. According to the horizontal curve parameters, vertical curve parameters and widening offset value of the tunnel, multiple two-dimensional line segments constituting the cross section of the tunnel are converted into multiple three-dimensional line segments to obtain the three-dimensional coordinates of each three-dimensional line segment at each mileage of the tunnel.

[0021] As an optimization, the horizontal curve parameters include circular curve radius, circular curve length, transition curve length, straight-slow point, transition-circular point, circular-slow point, and transition-straight point; the vertical curve parameters include slope, slope length, and vertical curve radius.

[0022] As an optimization, the rock mass physical and mechanical information includes uniaxial compressive strength Rc, rock mass integrity index Kv, groundwater coefficient K1, structural surface attitude state coefficient K2 and initial ground stress coefficient K3, and the face information includes face attitude, trace, spacing and roughness.

[0023] As an optimization, the specific steps of S4 are:

[0024] The grouting pipe is inserted into the broken rock cracks at a certain length according to a set angle for grouting. The length of the grouting pipe inserted into the rock cracks is specifically:

[0025]

[0026] The angle set by the grouting pipe includes a plane angle and an elevation angle, and the plane angle is:

[0027] θ = arctg(X2-X1) / (Z2-Z1);

[0028] The elevation angle is:

[0029] ψ=arctg(Y2-Y1) / L';

[0030] L′ is the horizontal projection length:

[0031]

[0032] Among them, (X1, Y1, Z1) are the coordinates of the starting hole on the rock fracture, and (X2, Y2, Z2) are the coordinates of the final hole in the rock mass.

[0033] As an optimization, the specific method of judging whether the surrounding rock is broken according to the tunnel face is:

[0034] When the rock integrity index Kv>0.15 and the groundwater coefficient K1<0.7, the surrounding rock is judged to be unbroken;

[0035] When the rock integrity index Kv≤0.15 and the groundwater coefficient K1≥0.7, the surrounding rock is judged to be broken.

[0036] As an optimization, the drilling parameters include the slot hole drilling angle, the slot hole drilling spacing, the number of slot holes and the drilling length.

[0037] As an optimization, S5 also includes real-time determination of the current surrounding rock grade during drilling, and the integration of geological information from multiple constructions into the coordinates of the drilling rig to establish a three-dimensional geological model, ultimately forming a complete address database for the entire tunnel.

[0038] As an optimization, the specific steps for real-time determination of the current surrounding rock grade during drilling are:

[0039] S5.1. Detect drilling pressure, drilling flow rate and drilling speed during drilling;

[0040] S5.2. Convert drilling pressure, drilling flow rate and drilling speed into surrounding rock grade, fracture conditions and rock mass strength;

[0041] S5.3. Install cameras on the drill pipe to collect information about surrounding rock crushing;

[0042] S5.4. The surrounding rock grade, fracture condition and rock mass strength obtained through conversion from S5.2 are used to verify the surrounding rock grade, rock mass integrity index and uniaxial compressive strength obtained in S2.

[0043] As an optimization, the specific steps of S7 are:

[0044] S7.1. Draw a perpendicular line from any point in the 3D point cloud to the designed excavation contour line, with a vertical distance d z(i),c(j) :

[0045]

[0046] Among them, d z(i),c(j) is the linear over-excavation value at that location, i.e., the vertical distance; z(i) is the i-th mileage in the longitudinal direction of the tunnel; c(j) is the j-th three-dimensional line segment at that mileage; X z(i),a(j) is the horizontal coordinate of the jth laser point at the ith mileage; X z(i),b(j) Y is the horizontal coordinate of the design contour corresponding to the jth laser point at the i-th mileage; z(i),a(j) Y is the ordinate of the jth laser point at the i-th mileage; z(i),b(j) is the ordinate of the design contour corresponding to the jth laser point at the i-th mileage;

[0047] S7.2. Obtain the cross-sectional over-excavation area and over-excavation volume of the blasting section by integral calculation:

[0048]

[0049]

[0050] Among them, Az(i),c(j) is the over-excavation area at the i-th mileage and j-th three-dimensional line segment in the longitudinal direction of the tunnel, V z(i),c(j) It is the over-excavation volume within the longitudinal length L of the tunnel.

[0051] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0052] 1. The present invention relates to an automatic identification technology of tunnel face geology based on image recognition, which is used to solve the problem of lack of experienced engineers at the construction site, the problem of inaccurate and untimely geology at the construction site, and the problem of real-time detection and evaluation of over-excavation at the site;

[0053] 2. By establishing a coordinate system, the 2D design cross section of the tunnel is obtained, and the cross section is converted into a 3D model according to the route map to obtain the 3D design coordinates and other attribute information (such as the thickness and strength of shotcrete in different tunnel sections, the thickness and strength of the secondary lining, the spacing and length of anchor bolts and other design support parameters), and compared with the 3D scanning model, the linear over-excavation value, over-excavation area value, and over-excavation volume value of the location are calculated, so as to realize real-time evaluation of the over-excavation and under-excavation quality;

[0054] 3. Use mobile phones to take photos of the tunnel face, conduct data analysis on joint occurrence and other information, and combine tests to determine the rock strength, and then calculate the BQ value to determine the surrounding rock classification in real time, providing a basis for the next step of excavation;

[0055] 4. According to the geological picture of the face, timely judgment and pre-reinforcement are carried out to realize real-time information-based pre-reinforcement, making on-site construction safer and faster; at the same time, the position relationship between the length l and the angle α of the grouting pipe is proposed. After the length and angle are determined, and based on the distance a between the grouting pipe and the excavation line, the thickness of the grouting reinforcement ring h=1*α-a can be calculated, which can ensure the thickness of the grouting reinforcement ring and provide a basis for the acceptance of concealed projects.

[0056] 5. Based on drilling data or images, the detection information is converted into geological information such as surrounding rock grade and crack conditions; geological conditions and over-excavation prediction are judged in real time.

[0057] 6. Effectively control blasting over-excavation, and reduce the comprehensive linear average over-excavation value from the current 28cm to 15cm, thereby reducing costs, increasing efficiency, and reducing carbon emissions.

[0058] 7. The entire process from on-site collection, classification and determination to design scheme matching only takes 8-12 minutes to complete. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0060] Figure 1 It is a flow chart of a real-time information-based blasting control method based on image recognition of the present invention;

[0061] Figure 2 This is a schematic diagram for calculating the linear over-excavation value;

[0062] Figure 3 It is the position relationship diagram of the advance grouting pipe length and angle calculation;

[0063] Figure 4 It is a process diagram for calculating BQ value;

[0064] Figure 5 To predict the location map of the over-digging value in the next cycle;

[0065] Figure 6 This is a schematic diagram of the grouting longitudinal section arrangement;

[0066] Figure 7 for Figure 6 Cross-section of the middle AA;

[0067] Figure 8 for Figure 6 Cross-section of the middle BB;

[0068] Fig. 9 Schematic diagram of the arrangement of the advance small catheter. DETAILED DESCRIPTION

[0069] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0070] A real-time information-based blasting control method based on image recognition in Example 1, such as Figure 1 As shown, including:

[0071] S1. Establish a three-dimensional coordinate system to obtain the three-dimensional coordinates of each mileage and each three-dimensional line segment of the tunnel to be excavated.

[0072] Specifically, S1.1, establish a three-dimensional coordinate system, obtain the cross section of the two-dimensional tunnel design, and determine the coordinate origin according to the actual situation, wherein the X coordinate is the tunnel span direction, the Y coordinate is the tunnel height direction, and the Z coordinate is the tunnel line longitudinal direction;

[0073] S1.2. Convert the circular arc of the cross section of the tunnel into a plurality of two-dimensional line segments to obtain the two-dimensional coordinates of the plurality of two-dimensional line segments; the length of the two-dimensional line segments is 5 cm;

[0074] The specific coordinate origin is formulated as follows:

[0075]

[0076] S1.3. Query the route map of the tunnel and extract the horizontal curve parameters, vertical curve parameters and widening offset value of the tunnel; the horizontal curve parameters include circular curve radius, circular curve length, transition curve length, straight-slow point, transition-circular point, circular-slow point, and transition-straight point; the vertical curve parameters include slope, slope length, and vertical curve radius.

[0077] S1.4. According to the horizontal curve parameters, vertical curve parameters and widening offset value of the tunnel, the multiple two-dimensional line segments constituting the cross section of the tunnel are converted into multiple three-dimensional line segments, and the three-dimensional coordinates and other attribute information of each mileage and each three-dimensional line segment of the tunnel are obtained. The coordinates of the top surface of the track or the top surface of the surface layer at each mileage are calculated using the line parameters, and then a three-dimensional model is generated using BIM software such as Revit.

[0078] S2, taking photos of the tunnel face in the tunnel and determining the physical and mechanical information of the rock mass in the tunnel, extracting the tunnel face information from the tunnel face photos based on image recognition, and calculating the BQ index through the tunnel face information combined with the physical and mechanical information of the rock mass to perform overall and local surrounding rock classification of the tunnel face.

[0079] Place searchlights on both sides of the drilling rig and adjust the lighting direction to reduce the shadows on the face. Take multiple photos of the face from left to right. Try to use the default magnification of the mobile phone camera when taking photos (to make each group of several different photos more accurately compared, to ensure the accurate extraction of point cloud normal vectors and joint occurrence), and make the core area of ​​the face as full of photos as possible. Determine the compressive strength of the rock mass through on-site uniaxial point load tests, on-site excavation face drilling measurement, and on-site rock mass rebound tests (Schmidt hammer rebound tester). Set up a 5G wireless transmission network in the cave to connect the photos taken by the mobile phone with the physical and mechanical information of the rock mass. The data is uploaded to the cloud platform through the WeChat applet; on the cloud platform, the face is processed by point cloud and three-dimensional reconstruction, and the face information is extracted in the analysis module to generate the normal vector of each point and calculate the joint attitude; the analysis module automatically extracts information (face attitude, trace, spacing, roughness, etc.), and combines the physical and mechanical information of the rock mass (a total of 5 parameters including uniaxial compressive strength Rc, rock integrity index Kv, groundwater coefficient K1, structural surface attitude state coefficient K2 and initial ground stress coefficient K3) to calculate the BQ index for overall and local surrounding rock classification of the face.

[0080] The rock mass integrity index Kv should adopt the elastic wave detection value, Kv = (v pm / v pr ) 2 ; If there is no detection value, the structural plane spacing can be obtained through image recognition technology, and the rock volume joint number (Jv) can be calculated through the structural plane spacing, and then the corresponding Kv value can be obtained by looking up the table below.

[0081] Structural surface spacing S(mm) d≥300 100<d<300 50<d<100 30<d<50 d≤30 Jv <3 3～10 10～20 20～35 ≥35 Kv >0.75 0.75～0.55 0.55～0.35 0.35～0.15 ≤0.15

[0082] Through the image recognition of the tunnel face, there are n groups of structural face spacings, namely S1, S2...Sn, then S=min(S1, S2...Sn).

[0083] The groundwater influence correction factor K1 refers to Table A.0.3-1 in Appendix A of the Highway Tunnel Design Code.

[0084]

[0085] Where p is the water pressure in the surrounding rock fissures of the underground project (MPa);

[0086] Q is the amount of water seepage in the tunnel per 10 meters per minute, and its unit is L / (minx10m).

[0087] The structural surface attitude coefficient K2 refers to Table A.0.3-2 in Appendix A of the Highway Tunnel Design Code

[0088]

[0089]

[0090] The inclination, direction and tendency of the structural surface can be obtained through the face image recognition technology. At the same time, the azimuth of the tunnel axis can be obtained by querying the tunnel line map. The angle between the structural surface direction and the tunnel axis can be obtained through calculation and analysis.

[0091] The initial geostress coefficient K3 refers to Table A.0.3-3 in Appendix A of the Highway Tunnel Design Code.

[0092]

[0093] Calculate the BQ index to classify the overall and local surrounding rock of the face; the matching results of the face information, surrounding rock classification and tunnel design support parameters (including shotcrete strength and thickness, secondary lining strength and thickness, anchor length and spacing, etc.) extracted by the cloud platform can be viewed through the on-site mobile phone WeChat applet or the web cloud platform.

[0094] like Figure 4 As shown in the figure, after obtaining the tunnel face photos, the cloud platform uses CloudCompare software (point cloud filtering processing) to perform point cloud analysis and three-dimensional reconstruction on the tunnel face photos, and obtains the normal vector, so as to infer the inclination and dip angle of the tunnel face joints, etc. Through computer Matlab software code programming, testfig software is used to run the tunnel face information, and the structural surface grouping attitude, independent structural surface attitude, roughness, traces, trace spacing and structural surface spacing of the tunnel face are extracted. The rock volume joint number Jv is calculated through the structural surface spacing, and the rock integrity index Kv is calculated through the rock volume joint number Jv.

[0095] The uniaxial compressive strength Rc is obtained through the rebound test, and the basic rock quality index BQ value is calculated together with the rock integrity index Kv: BQ=100+3Rc+250Kv.

[0096] At the same time, the following restrictions apply:

[0097] When Rc>90Kv+30, Rc=90Kv+30 and Kv should be substituted into the calculation of BQ value.

[0098] When Kv>0.04Rc+0.4, Kv=0.04Rc+0.1 and Rc should be substituted into the calculation of BQ value.

[0099] When the surrounding rock mass is graded in detail, the basic rock mass quality index BQ should be corrected according to the influence of groundwater, main weak structural surface and initial stress state:

[0100] [BQ] = BQ - 100 (K1 + K2 + K3)

[0101] Where: [BQ]——rock mass correction quality index;

[0102] K1——groundwater impact correction factor;

[0103] K2——Correction coefficient of the influence of the main weak structural surface attitude;

[0104] K3——Correction coefficient for the influence of initial stress state.

[0105] The values ​​of K1, K2 and K3 can be determined according to Tables A.0.3-1, A.0.3-2 and A.0.3-3 in Appendix A of the "Highway Tunnel Design Code" respectively.

[0106] S3, conducting a geological survey on the tunnel face photo, and judging whether the surrounding rock is broken according to the tunnel face. If not, jump to S5, otherwise jump to S4;

[0107] Based on the above-mentioned mobile phone photos of the tunnel face and the geological survey of the tunnel face photos, the corresponding surrounding rock is judged whether it is broken according to the tunnel face; if the surrounding rock is not broken (when Kv>0.15 and K1<0.7), the next step of drilling with the drilling rig is directly carried out; if the surrounding rock is broken (when Kv≤0.15 and K1≥0.7), the reinforcement parameters (including grouting pressure, grouting materials, drilling spacing, and drilling depth) are proposed with technical indicators such as over-break and under-break as constraints. Generally, the smaller the surrounding rock integrity coefficient Kv, the more broken the surrounding rock is. Without considering other factors such as water level, the tunnel seepage and water pressure will be greater when the surrounding rock is broken, so the larger K1 is.

[0108] The reinforcement parameters are set as follows according to the rock mass integrity index Kv:

[0109]

[0110] S4. According to the three-dimensional scanning results, the local broken section is subjected to information-based advance reinforcement based on the fracture occurrence to reduce over-excavation.

[0111] The specific steps for pre-reinforcement are:

[0112] The grouting pipe is inserted into the broken rock cracks at a set angle and a certain length for grouting. Figure 3 As shown, the length of the grouting pipe inserted into the rock mass crack is specifically:

[0113] O1O2 Length:

[0114] The angle set by the grouting pipe includes a plane angle and an elevation angle, and the plane angle is:

[0115] θ = arctg(X2-X1) / (Z2-Z1);

[0116] The elevation angle is:

[0117] ψ=arctg(Y2-Y1) / L';

[0118] L′ is the horizontal projection length:

[0119]

[0120] Among them, (X1, Y1, Z1) are the coordinates of the starting hole on the rock fracture, and (X2, Y2, Z2) are the coordinates of the final hole in the rock mass.

[0121] Within the excavation contour, for section B, if the first circle (the outermost circle) is arranged with i holes (for inserting grouting pipes), the angle between every two holes is 360 / i degrees, and the hole arrangement radius is r1B,

[0122] 1-1 The drilling coordinates are (cos(0×360 / i+90°)×r1B, sin(0×360 / i+90°)×r1B, 0);

[0123] The coordinates of 1-k drilling holes are (cos((k-1)×360 / i+90°)×r1B, sin((k-1)×360 / i+90°)×r1B, 0);

[0124] The range of k is 1≤k≤i;

[0125] Taking the fifth circle with 8 holes and the sixth circle with 3 holes as an example, the calculated coordinates are as follows:

[0126] There are 8 holes arranged in the 5th circle, the angle between every two holes is 360 / 8=45 degrees, and the hole arrangement radius is r5B.

[0127] 5-1 The drilling coordinates are (cos(0×45°+90°)×r5B, sin(0×45°+90°)×r5B, 0);

[0128] 5-2 The drilling coordinates are (cos(1×45°+90°)×r5B, sin(1×45°+90°)×r5B, 0);

[0129] 5-3 The drilling coordinates are (cos(2×45°+90°)×r5B, sin(2×45°+90°)×r5B, 0);

[0130] 5-4 The drilling coordinates are (cos(3×45°+90°)×r5B, sin(3×45°+90°)×r5B, 0);

[0131] 5-5 The drilling coordinates are (cos(4×45°+90°)×r5B, sin(4×45°+90°)×r5B, 0);

[0132] 5-6 The drilling coordinates are (cos(5×45°+90°)×r5B, sin(5×45°+90°)×r5B, 0);

[0133] The coordinates of the drilling hole 5-7 are (cos(6×45°+90°)×r5B, sin(6×45°+90°)×r5B, 0);

[0134] The drilling coordinates of 5-8 are (cos(7×45°+90°)×r5B, sin(7×45°+90°)×r5B, 0).

[0135] The sixth circle has three holes, the angle between every two holes is 360 / 3=120 degrees, and the hole arrangement radius is r6B.

[0136] 6-1 The drilling coordinates are (cos(0×120°+90°)×r6B, sin(0×120°+90°)×r6B, 0);

[0137] 6-2 The drilling coordinates are (cos(1×120°+90°)×r6B, sin(1×120°+90°)×r6B, 0);

[0138] 6-3 The drilling coordinates are (cos(2×120°+90°)×r6B, sin(2×120°+90°)×r6B, 0).

[0139] For section C, if i holes are arranged in the first circle, the angle between every two holes is 360 / i degrees, the hole arrangement radius is r1C, the distance between section C and section B is L1, and the number of holes set in section B must be the same as the number of holes set in section C, section D has one less hole than the first circle, and section E has one less hole than the first and second circles;

[0140] 1-1 The drilling coordinates are (cos(0×360 / i+90°)×r1C, sin(0×360 / i+90°)×r1C, L1);

[0141] The coordinates of 1-k drilling holes are (cos((k-1)×360 / i+90°)×r1C, sin((k-1)×360 / i+90°)×r1C, L1);

[0142] The range of k is 1≤k≤i;

[0143] Assuming that there are 8 holes in the 5th circle and 3 holes in the 6th circle, the calculated coordinates are as follows:

[0144] There are 8 holes arranged in the 5th circle, the angle between every two holes is 360 / 8=45 degrees, and the hole arrangement radius is r5C.

[0145] 5-1 The drilling coordinates are (cos(0×45°+90°)×r5C, sin(0×45°+90°)×r5C, L1);

[0146] 5-2 The drilling coordinates are (cos(1×45°+90°)×r5C, sin(1×45°+90°)×r5C, L1);

[0147] 5-3 The drilling coordinates are (cos(2×45°+90°)×r5C, sin(2×45°+90°)×r5C, L1);

[0148] 5-4 The drilling coordinates are (cos(3×45°+90°)×r5C, sin(3×45°+90°)×r5C, L1);

[0149] 5-5 The drilling coordinates are (cos(4×45°+90°)×r5C, sin(4×45°+90°)×r5C, L1);

[0150] 5-6 The drilling coordinates are (cos(5×45°+90°)×r5C, sin(5×45°+90°)×r5C, L1);

[0151] The coordinates of the 5-7 drilling hole are (cos(6×45°+90°)×r5C, sin(6×45°+90°)×r5C, L1);

[0152] The drilling coordinates of 5-8 are (cos(7×45°+90°)×r5C, sin(7×45°+90°)×r5C, L1).

[0153] The sixth circle has three holes, the angle between every two holes is 360 / 3=120 degrees, and the hole arrangement radius is r6C.

[0154] 6-1 The drilling coordinates are (cos(0×120°+90°)×r6C, sin(0×120°+90°)×r6C, L1);

[0155] 6-2 The drilling coordinates are (cos(1×120°+90°)×r6C, sin(1×120°+90°)×r6C, L1);

[0156] 6-3 The drilling coordinates are (cos(2×120°+90°)×r6C, sin(2×120°+90°)×r6C, L1).

[0157] Section B is the first section where grouting starts, Section C is the second section after 10 meters, Section D is the third section after 18 meters, and Section E is the fourth section after 30 meters. Figure 6-8 shown.

[0158] The number of holes in each circle is determined by the hole arrangement radius and the spacing between the holes in each circle. The circumferential spacing between the holes in each circle is about 80 to 100 cm. For example, the first circle has 22 holes, the second circle has 22 holes, the third circle has 22 holes, the fourth circle has 14 holes, the fifth circle has 6 holes, and the sixth circle has 3 holes.

[0159] Section B: The radius of the holes in the first circle is 50cm+5×80cm, the radius of the holes in the second circle is 50cm+4×80cm, the radius of the holes in the third circle is 50cm+3×80cm, the radius of the holes in the fourth circle is 50cm+2×80cm, the radius of the holes in the fifth circle is 50cm+1×80cm, and the radius of the holes in the sixth circle is 50cm+0×80cm.

[0160] In the above radii, 50 represents the radius of the sixth circle, which is also the radius of the smallest circle, and the radii of the remaining circles increase by 80 cm respectively.

[0161] The final hole of the first circle of section B is just at the outermost edge of section C, the final hole of the second circle of section B is just at the outermost edge of section D, the final hole of the third circle of section B is just at the outermost edge of section E, the final hole of the fourth circle of section B corresponds to the final hole of the second circle of section E (section E has 4 circles), the final hole of the fifth circle of section B corresponds to the final hole of the third circle of section E, and the final hole of the sixth circle of section B corresponds to the final hole of the fourth circle of section E. Therefore, it can be determined that the angle α1 of the first circle, the angle α2 of the second circle, the angle α3 of the third circle, the angle α4 of the fourth circle, the angle α5 of the fifth circle, and the angle α6 of the sixth circle are reduced accordingly. The entire tunnel section is used for all drilling and grouting, that is, full-section grouting. It can also be used for half-section grouting.

[0162] α is the angle between the line connecting the two drill holes and the horizontal axis.

[0163] It can be seen from the above figure that the radius of each circle of holes in sections C, D, and E increases successively, and the degree of increase can be calculated by angle. For example, the radius of the fourth circle of holes in section D is (50cm+2×80cm)+18m×sinα4, where 18m is the distance from section B to section D.

[0164] The number of holes in each circle is determined by the hole arrangement radius and the spacing between the holes in each circle. The circumferential spacing between the holes in each circle is about 80 to 100 cm. For example, the first circle has 22 holes, the second circle has 22 holes, the third circle has 22 holes, the fourth circle has 14 holes, the fifth circle has 6 holes, and the sixth circle has 3 holes.

[0165] For example, (X1, Y1, Z1) are the coordinates of the hole in section B, and (X2, Y2, Z2) can be the coordinates of the hole in section C; or (X1, Y1, Z1) are the coordinates of the hole in section B, and (X2, Y2, Z2) can be the coordinates of the hole in section D; or (X1, Y1, Z1) are the coordinates of the hole in section B, and (X2, Y2, Z2) can be the coordinates of the hole in section E.

[0166] (X1, Y1, Z1), (X2, Y2, Z2), combined with the S1.1 coordinate system and the three-dimensional coordinates of S1.4, considering the circumferential spacing, radial spacing and longitudinal distance of the boreholes, can be calculated.

[0167] In addition, the thickness of the grouting reinforcement ring is h = l*α-a, l is the length of the grouting pipe, α is the angle between the grouting pipe and the horizontal axis, and a is the distance from the grouting pipe to the excavation contour line on section B (the distance from the point where the grouting pipe intersects with section B to the corresponding excavation contour line, for example, the center of the circle surrounded by section B is taken as the starting point, and the starting point and the point where the grouting pipe intersects with section B are extended to the excavation contour line, and the length of the extended line is the distance from the grouting pipe to the excavation contour line). This can ensure the thickness of the grouting reinforcement ring and provide a basis for the acceptance of concealed projects.

[0168] S5. Determine the drilling parameters of the drilling rig and drill holes according to the overall and local surrounding rock classification of the face, the geological conditions of the face and the reinforcement effect. Determine the current surrounding rock grade in real time during drilling, and combine the geological information of multiple constructions with the coordinates of the drilling rig to establish a three-dimensional geological model, ultimately forming a complete address database for the entire tunnel.

[0169] According to the surrounding rock grade (whole and local surrounding rock grade of the tunnel face), the geological conditions of the tunnel face and the pre-reinforcement effect, the drilling parameters (groove hole drilling angle, groove hole drilling spacing, number of groove holes and drilling length) are determined, and the drilling rig's fully automatic operation software is used for drilling. Refer to the following table for specific settings:

[0170]

[0171] The software is used to analyze the pressure, flow rate, speed and other data detected during the drilling process. Based on the drilling data, the detection information (detected drilling pressure, drilling flow rate and drilling speed) is converted into geological information such as surrounding rock grade and rock integrity index. A camera is placed on the drill pipe to collect the surrounding rock crushing situation to further verify the correctness and rationality of kv. The detection information can verify the compressive strength detected by the Schmidt rebound hammer, the rock integrity index Kv calculated by the cloud platform and the surrounding rock grade determined by the BQ value.

[0172] If the structure obtained by S5 is significantly different from the result obtained by S2, and the difference exceeds 25%, the owner, geological survey, design, construction and supervision need to jointly judge. If there are no other reasons and the results of S5 and S2 are both true, the smaller value of the two is taken. The specific relationship between drilling data and surrounding rock grade is shown in the following table:

[0173] Rock mass strength Drilling pressure Drilling flow Drilling speed Rock mass integrity index kv 45~75MPa 140~150bar 60~70L / min 2~2.5m / min 0.5～0.9 Class III surrounding rock 15~45MPa 130~140bar 70~80L / min 2.5~3.5m / min 0.3～0.5 IV grade surrounding rock 5~15MPa 120~130barn 80~95L / min 0.5~2.5m / min 0.1～0.3 V-level surrounding rock

[0174] After grouting and before excavation, set up a small advance conduit. The drilling position of the small advance conduit is set 20cm outside the excavation contour line, with an angle of 10-15° and a length of 2-3.5m. Fig. 9 shown.

[0175] The over-excavation value of the next cycle of the advance small guide tube is predicted by the drilling position, angle and length of the advance small guide tube (the advance small guide tube is a form of advance support. In most cases, it is outside the arch of the tunnel section, while the grouting is basically in the inner area of ​​the tunnel section). The over-excavation value of the next cycle of the advance small guide tube is l0×sin∝, where l0 represents the length of the stop grouting section of the advance small guide tube, and β represents the angle of the advance small guide tube. For details on the relative relationship, see Figure 5 For example, if the stop grouting section of the leading small conduit is 50 cm and the angle of the leading small conduit is 15°, the over-excavation value of the next cycle is 50 cm × sin15° = 12.9 cm.

[0176] Displayed in three-dimensional form, the geological information of multiple construction projects can be combined and spliced ​​with the trolley positioning coordinates to build a three-dimensional geological model; a complete geological database for the entire tunnel is established, laying the foundation for the improvement and optimization of subsequent construction.

[0177] S6. After the blasting construction is carried out, the coordinates of the three-dimensional point cloud data of the tunnel contour after the blasting converted into the three-dimensional coordinate system of S1 are obtained by three-dimensional laser scanning;

[0178] Carry out charging and blasting construction; obtain three-dimensional laser point cloud data of the tunnel outline after blasting through three-dimensional laser scanning; convert the three-dimensional laser point cloud data to the aforementioned three-dimensional coordinate system, and calculate the coordinates of each laser point cloud data, which is the coordinates of the three-dimensional point cloud data.

[0179] S7. Compare the coordinates of the same point at the same position before and after the blasting, calculate the linear over-excavation value, cross-sectional over-excavation area value and over-excavation volume value of the blasting section at that position, so as to evaluate the over-excavation and under-excavation quality of the blasting excavation surface; by comparing the coordinates of the same point at the same position, calculate the linear over-excavation value at that location, draw a perpendicular line to the designed excavation contour line from any point of the laser point cloud data, obtain the cross-sectional over-excavation area value and the over-excavation volume value of the blasting section through integral calculation, and evaluate the over-excavation and under-excavation quality.

[0180] The specific steps are as follows:

[0181] S7.1. Draw a perpendicular line from any point in the 3D point cloud to the designed excavation contour line, with a vertical distance d z(i),c(j) ,like Figure 2 As shown:

[0182]

[0183] Among them, d z(i),c(j) is the linear over-excavation value at that location, i.e., the vertical distance; z(i) is the i-th mileage in the longitudinal direction of the tunnel; c(j) is the j-th three-dimensional line segment at that mileage; X z(i),a(j) is the horizontal coordinate of the jth laser point at the ith mileage; X z(i),b(j) Y is the horizontal coordinate of the design contour corresponding to the jth laser point at the i-th mileage; z(i),a(j) Y is the ordinate of the jth laser point at the i-th mileage; z(i),b(j) is the ordinate of the design contour corresponding to the jth laser point at the i-th mileage;

[0184] S7.2. Obtain the cross-sectional over-excavation area and over-excavation volume of the blasting section by integral calculation:

[0185]

[0186]

[0187] Among them, A z(i),c(j) is the over-excavation area at the i-th mileage and j-th three-dimensional line segment in the longitudinal direction of the tunnel, V z(i),c(j) It is the over-excavation volume within the longitudinal length L of the tunnel.

[0188] S8. According to the over-excavation of the excavation face during the blasting and the surrounding rock geological conditions of the face after the blasting, the drilling parameters are optimized;

[0189] In actual construction, due to the influence of drilling machinery, a certain external insertion angle is inevitably formed. In order to reduce over-excavation, the external insertion angle can be reduced by reducing the hole depth or by moving the drilling position slightly inward to improve the drilling accuracy. For example, the drill rod length is 3.5m, the drilling angle is 5°, and the over-excavation value of the drill rod in the next cycle is 350cm×sin5°=30.5cm. If the hole depth is reduced, the length is reduced to 3.0m, and the over-excavation value of the next cycle is 300cm×sin5°=26.1cm. If the drilling position is moved inward a little h=5cm, the length remains unchanged, 350cm-5cm / tan5°=293cm, and the over-excavation value of the next cycle is 293cm×sin15°=25.5cm.

[0190] Closely following the excavation, the surrounding rock conditions are observed and described, and the state of the joints and fissures of the surrounding rock is predicted. Based on this, auxiliary measures such as locally moving the blastholes inward, leaving some holes empty without charging, increasing the density of blastholes, and locally adjusting the detonation sequence are taken.

[0191]

[0192] S9, return to S5 in combination with the optimized drilling parameters, conduct the next drilling test on the optimized overbreak and evaluate the overbreak and underbreak quality, and continue to further optimize the drilling parameters to reduce the overbreak. The test and evaluation adopt S6 and S7 methods to determine the overbreak and underbreak quality.

[0193] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A real-time information-based blasting control method based on image recognition, It is characterized in that include: S1. Establish a three-dimensional coordinate system to obtain the three-dimensional coordinates of each mileage and each three-dimensional line segment of the tunnel to be excavated; S2, taking photos of the tunnel face in the tunnel and determining the physical and mechanical information of the rock mass in the tunnel, extracting the tunnel face information from the tunnel face photos based on image recognition, and calculating the BQ index by combining the tunnel face information with the physical and mechanical information of the rock mass to perform overall and local surrounding rock classification of the tunnel face; S3, conducting a geological survey on the tunnel face photo, and judging whether the surrounding rock is broken according to the tunnel face. If not, jump to S5, otherwise jump to S4; S4. Taking the over-excavation and under-excavation technical indicators as constraints, carry out information-based advanced pre-reinforcement of local broken sections based on the three-dimensional scanning results and the fracture occurrence; S5. Determine the drilling parameters of the drilling rig and drill holes according to the overall and local surrounding rock classification of the face, the geological conditions of the face and the reinforcement effect; S6. After the blasting construction is carried out, the coordinates of the three-dimensional point cloud data of the tunnel contour after the blasting converted into the three-dimensional coordinate system of S1 are obtained by three-dimensional laser scanning; S7, comparing the coordinates of the same point at the same position before and after the blasting, calculating the linear over-excavation value, the cross-sectional over-excavation area value and the over-excavation volume value of the blasting section at the position, so as to evaluate the over-excavation and under-excavation quality of the excavation surface of the blasting; S8. According to the over-excavation of the excavation face during the blasting and the surrounding rock geological conditions of the face after the blasting, the drilling parameters are optimized; S9. Return to S5 based on the optimized blasting drilling parameters, conduct the next cycle drilling test on the optimized overbreak situation and evaluate the overbreak and underbreak quality, and continue to further optimize the drilling parameters to reduce the overbreak situation.

2. A real-time information-based blasting control method based on image recognition according to claim 1, It is characterized in that The specific steps of S1 are: S1.

1. Establish a three-dimensional coordinate system, obtain the cross section of the two-dimensional tunnel design, and determine the coordinate origin according to the actual situation, where the X coordinate is the tunnel span direction, the Y coordinate is the tunnel height direction, and the Z coordinate is the tunnel line longitudinal direction; S1.2, converting the circular arc of the cross section of the tunnel into a plurality of two-dimensional line segments to obtain the two-dimensional coordinates of the plurality of two-dimensional line segments; S1.

3. Query the route map of the tunnel, and extract the horizontal curve parameters, vertical curve parameters and widening offset value of the tunnel; S1.

4. According to the horizontal curve parameters, vertical curve parameters and widening offset value of the tunnel, multiple two-dimensional line segments constituting the cross section of the tunnel are converted into multiple three-dimensional line segments to obtain the three-dimensional coordinates of each three-dimensional line segment at each mileage of the tunnel.

3. A real-time information-based blasting control method based on image recognition according to claim 2, It is characterized in that The horizontal curve parameters include circular curve radius, circular curve length, transition curve length, straight-to-slow point, transition-to-circular point, circular-to-slow point, and transition-to-straight point. The vertical curve parameters include slope, slope length, and vertical curve radius.

4. A real-time information-based blasting control method based on image recognition according to claim 1, It is characterized in that The rock mass physical and mechanical information includes uniaxial compressive strength Rc, rock mass integrity index Kv, groundwater coefficient K1, structural surface attitude state coefficient K2 and initial ground stress coefficient K3, and the tunnel face information includes tunnel face attitude, trace, spacing and roughness.

5. A real-time information-based blasting control method based on image recognition according to claim 1, It is characterized in that The specific steps of S4 are: The grouting pipe is inserted into the broken rock cracks at a certain length according to a set angle for grouting. The length of the grouting pipe inserted into the rock cracks is specifically: The angle set by the grouting pipe includes a plane angle and an elevation angle, and the plane angle is: θ=arctg(X 2 -X 1 ) / (Z 2 -Z 1 ): The elevation angle is: ψ=arctg(Y 2 -Y 1 ) / L'; L′ is the horizontal projection length: Among them, (X 1 , Y 1 , Z 1 ) is the coordinate of the starting hole on the rock fracture, (X 2 , Y 2 , Z 2 ) are the coordinates of the final hole in the rock mass.

6. A real-time information-based blasting control method based on image recognition according to claim 1, It is characterized in that The specific method of judging whether the surrounding rock is broken according to the tunnel face is: When the rock integrity index Kv>0.15 and the groundwater coefficient K1<0.7, the surrounding rock is judged to be unbroken; When the rock integrity index Kv≤0.15 and the groundwater coefficient K1≥0.7, the surrounding rock is judged to be broken.

7. A real-time information-based blasting control method based on image recognition according to claim 1, It is characterized in that The drilling parameters include the slot hole drilling angle, the slot hole drilling spacing, the number of slot holes and the drilling length.

8. A real-time information-based blasting control method based on image recognition according to claim 1, It is characterized in that S5 also includes real-time determination of the current surrounding rock grade during drilling, and the integration of geological information from multiple constructions into the coordinates of the drilling rig to build a three-dimensional geological model, ultimately forming a complete address database for the entire tunnel.

9. A real-time information-based blasting control method based on image recognition according to claim 8, It is characterized in that The specific steps for real-time determination of the current surrounding rock grade during drilling are: S5.

1. Detect drilling pressure, drilling flow rate and drilling speed during drilling; S5.

2. Convert drilling pressure, drilling flow rate and drilling speed into surrounding rock grade, fracture conditions and rock mass strength; S5.

3. Install cameras on the drill pipe to collect information about surrounding rock crushing; S5.

4. The surrounding rock grade, crack condition and rock mass strength obtained by conversion from S5.2 are used to verify the surrounding rock grade, rock mass integrity index and uniaxial compressive strength obtained from S2.

10. A real-time information-based blasting control method based on image recognition according to claim 1, It is characterized in that The specific steps of S7 are: S7.

1. Draw a perpendicular line from any point in the 3D point cloud to the designed excavation contour line, with a vertical distance d z(i),c(j) : Among them, d z(i),c(j) is the linear over-excavation value at that location, i.e., the vertical distance; z(i) is the i-th mileage in the longitudinal direction of the tunnel; c(j) is the j-th three-dimensional line segment at that mileage; X z(i),a(j) is the horizontal coordinate of the jth laser point at the ith mileage; X z(i),b(j) Y is the horizontal coordinate of the design contour corresponding to the jth laser point at the i-th mileage; z(i),a(j) Y is the ordinate of the jth laser point at the i-th mileage; z(i),b(j) is the ordinate of the design contour corresponding to the jth laser point at the i-th mileage; S7.

2. Obtain the cross-sectional over-excavation area and over-excavation volume of the blasting section by integral calculation: Among them, A z(i),c(j) is the over-excavation area at the i-th mileage and j-th three-dimensional line segment in the longitudinal direction of the tunnel, V z(i),c(j) It is the over-excavation volume within the longitudinal length L of the tunnel.

Images