Mine roadway on-demand supporting structure optimization method based on artificial intelligence

Abstract

The invention discloses a mine roadway on-demand supporting structure optimization method based on artificial intelligence, and relates to the technical field of mine roadway supporting. Roadway measurement data of roadway measurement points are collected; the roadway measurement data are captured and extracted, fracture geological parameters are obtained, graded induction is conducted through the fracture geological parameters, and the rock mass quality grade is obtained; constructing a roadway section model and performing parameter acquisition to obtain original rock mass mechanical parameters, and performing simulation construction through the roadway section model to obtain a same-level simulation parameter library; performing simulation comparison according to the same-level simulation parameter library to obtain a parameter simulation result, and performing defect enhancement on the roadway measurement point according to the parameter simulation result to obtain an optimal support scheme; the design efficiency and accuracy of a supporting scheme are improved, the stability and safety of a roadway are improved, excessive supporting or insufficient supporting is avoided, and the supporting economical efficiency is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of mine tunnel support, and in particular to an artificial intelligence-based mine tunnel on-demand support structure optimization method. Background Art

[0002] The stability and safety of mine roadways are critical issues in mining engineering, directly impacting mine production efficiency, safe production, and the safety of workers. Repeated mining disturbances increase ground pressure in roadways and near stopes, reduce rock mass strength, and degrade rock quality, leading to frequent ground pressure phenomena such as spalling, roof falls, and collapses.

[0003] The emergence of the concept of on-demand support design for rock tunnels provides a new and effective design approach for intelligent optimization design and rapid drawing production in mines, avoiding the awkwardness of design that varies from person to person. In order to solve the problems existing in existing mine tunnel support methods, an artificial intelligence-based on-demand support structure optimization method for mine tunnels is now provided. The artificial intelligence algorithm is used to process and analyze mine tunnel data, establish a tunnel stability prediction model, and adjust and optimize the support structure in real time based on the prediction model and tunnel usage requirements. This improves the intelligence level of mine tunnel support, reduces the impact of human factors on support effects, improves the safety and stability of tunnels, reduces the risk of tunnel accidents, reduces the cost of mine tunnel support, and improves the economic benefits of mines. Summary of the Invention

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] An artificial intelligence-based mine tunnel support structure optimization method includes the following steps:

[0006] Step S1: collecting tunnel measurement data of tunnel measurement points;

[0007] Step S2: capturing and extracting the tunnel measurement data to obtain fracture geological parameters, and classifying and summarizing the fracture geological parameters to obtain the rock mass quality grade;

[0008] Step S3: Construct a tunnel cross-section model and collect parameters to obtain the original rock mass mechanical parameters, perform simulation construction through the tunnel cross-section model, and obtain a simulation parameter library of the same level;

[0009] Step S4: Perform simulation comparison based on the simulation parameter library of the same level to obtain parameter simulation results, and enhance the defects of the tunnel measurement points based on the parameter simulation results to obtain the best support solution.

[0010] Preferably, the process of collecting tunnel measurement data includes:

[0011] Lay and capture the target mine tunnel and obtain tunnel measurement points;

[0012] The target acquisition terminal is set according to the obtained tunnel measurement points, and information is collected through the target acquisition terminal to obtain tunnel measurement data.

[0013] Preferably, the process of capturing and extracting the roadway measurement data includes:

[0014] Acquire tunnel measurement data, perform close-up capture of tunnel measurement data, obtain close-up data of cracks, conduct sample tests on tunnel measurement data, and obtain uniaxial compressive strength;

[0015] The obtained fracture close-up data is captured to obtain fracture geological parameters.

[0016] Preferably, the process of classifying and summarizing fracture geological parameters includes:

[0017] Based on the fracture geological parameters, the rock mass is summarized to obtain the incompleteness index;

[0018] The rock mass is graded according to the obtained incompleteness index to obtain the rock mass quality index;

[0019] The obtained rock mass quality indicators are graded and scored to obtain the rock mass quality grade.

[0020] Preferably, the process of constructing a tunnel cross-section model and collecting parameters to obtain original rock mass mechanical parameters includes:

[0021] Expand the model of the target mine tunnel to obtain the tunnel cross-section model;

[0022] Based on the rock mass quality grade of the tunnel measurement points, the tunnel cross-section model is matched with the grade area to obtain the quality grade of the measurement points;

[0023] The tunnel cross-section model is distinguished and marked according to the quality level of the measurement points to obtain the marked tunnel model, and the parameters of the marked tunnel model are collected to obtain the original rock mass mechanical parameters.

[0024] Preferably, the process of constructing a simulation through a tunnel cross-section model and obtaining a simulation parameter library of the same level includes:

[0025] Based on the original rock mass mechanical parameters, the support numerical simulation of the marked tunnel model is carried out to obtain support simulation data;

[0026] Construct an initial parameter level library based on the quality levels of the measured points;

[0027] The initial parameter level library is summarized at the same level according to the support simulation data to obtain the simulation parameter library of the same level.

[0028] Preferably, the process of performing simulation comparison based on the simulation parameter library of the same level includes:

[0029] The marked roadway model is simulated and run according to the obtained simulation parameter library of the same level, and the parameters of the marked roadway model during the simulation are collected to obtain the operating rock mass mechanical parameters;

[0030] Based on the marked tunnel model, the effects of the operating rock mass mechanical parameters are compared with the original rock mass mechanical parameters to obtain the parameter simulation results.

[0031] Preferably, the process of enhancing the defects at the tunnel measurement points according to the parameter simulation results and obtaining the best support scheme includes:

[0032] According to the parameter simulation results, the effect threshold is set, and the parameter simulation results are checked for excellence through the effect threshold to obtain insignificant simulation results. The tunnel measurement points corresponding to the insignificant results are recorded as measurement points to be optimized;

[0033] Based on the simulation parameter library of the same level, support optimization is performed on the optimal measurement points to obtain optimized support data. The marked roadway model is optimized and simulated based on the optimized support data. Parameters of the marked roadway model during the optimization simulation are collected to obtain optimized rock mass mechanical parameters.

[0034] The effects of the optimized rock mechanics parameters are compared based on the original rock mechanics parameters obtained to obtain the optimized parameter simulation results. The optimized parameter simulation results are ranked by effect to obtain the best simulated support data. Based on the best simulated support data, the simulation parameter library of the same level is updated and screened to obtain the best support scheme.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. By collecting comprehensive data from measurement points in mine tunnels, extracting and analyzing the comprehensive data, we can obtain crack image information, combine it with rock mass data information for processing and analysis, and obtain the rock mass quality grade, so as to obtain a reasonable support plan, ensure that the tunnel remains stable during use, and extend the service life of the tunnel;

[0037] 2. Construct a tunnel cross-section model and conduct targeted simulation design under different rock mass quality grades. Based on the simulation results of rock mass quality classification, support schemes that adapt to different geological conditions can be designed to improve the applicability and flexibility of the support scheme;

[0038] 3. Conduct secondary simulation of weak measurement points in the tunnel cross-section model and optimize and update them to avoid over-support or under-support, making the design more reasonable and efficient, reducing unnecessary support materials and construction costs, and improving economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 This is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0041] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0042] like Figure 1 As shown, a mine tunnel support structure optimization method based on artificial intelligence includes the following steps:

[0043] Step S1: collecting tunnel measurement data of tunnel measurement points;

[0044] Step S2: capturing and extracting the tunnel measurement data to obtain fracture geological parameters, and classifying and summarizing the fracture geological parameters to obtain the rock mass quality grade;

[0045] Step S3: Construct a tunnel cross-section model and collect parameters to obtain the original rock mass mechanical parameters, perform simulation construction through the tunnel cross-section model, and obtain a simulation parameter library of the same level;

[0046] Step S4: Perform simulation comparison based on the simulation parameter library of the same level to obtain parameter simulation results, and enhance the defects of the tunnel measurement points based on the parameter simulation results to obtain the best support solution.

[0047] It should be further explained that, in the specific implementation process, the mining of the ore body breaks the original stress balance, and the original stress is transferred to the surrounding pillars and other core safety guarantee positions of the mine, generating new redistributed stress, resulting in the continuous increase of stress concentrated on the pillars, causing the continuous appearance of ground pressure phenomena such as roadway spalling, roof fall, and collapse. This is one of the root causes of the widespread instability of the surrounding rock of mines. The mining disturbance causes the expansion of rock cracks and the continuous reduction of strength, which is the second root cause of the widespread instability of the surrounding rock of mine rock tunnels. Therefore, it is of great significance to strengthen the timely and effective support of the rock mass around the passages where personnel and mechanical equipment frequently enter and exit the mine, such as the mining road, rock drilling road, and transportation road, and eliminate the safety hazards of roadways and chambers due to spalling, roof fall, etc. The process of collecting roadway measurement data includes:

[0048] Lay and capture the target mine tunnel and obtain tunnel measurement points;

[0049] The target mine tunnel represents a mine tunnel that needs to be supported, and the laying capture represents setting measurement points in the target mine tunnel and collecting information at the measurement points;

[0050] According to the obtained tunnel measurement points, a target acquisition terminal is set, and information is collected through the target acquisition terminal to obtain tunnel measurement data;

[0051] The tunnel measurement data includes digital close-up photography, field survey records, and rock sample data. Among them, the digital close-up photography refers to photos taken of various parts of the tunnel rock mass, including the roof, walls and floor, and the photos can clearly reflect the structural characteristics of the rock mass, including cracks, joints, rock layers, etc. The field survey records represent engineers' inspection of the rock mass of the mine tunnel and the records of macroscopic characteristics, such as crack development and rock mass structure. The rock sample data refers to rock samples of tunnel measurement points selected in the mine tunnel.

[0052] The tunnel measurement data is captured and extracted to obtain fracture geological parameters, which are then classified and summarized to obtain the rock mass quality grade. The specific process includes:

[0053] Acquire tunnel measurement data, perform close-up capture on the acquired tunnel measurement data, and obtain close-up data of fractures;

[0054] The close-up capture refers to the process of capturing crack images in a mine tunnel using an image extraction method from digital close-up photographs in tunnel measurement data and on-site survey records to obtain close-up crack data, representing image information of cracks in the mine tunnel. The image extraction method refers to the process of obtaining three-dimensional coordinates or two-dimensional coordinates of the crack image using a three-dimensional image acquisition method or a two-dimensional image acquisition method, i.e., the corresponding position coordinates can be marked while extracting the crack image.

[0055] The obtained roadway measurement data were subjected to sample tests to obtain the uniaxial compressive strength;

[0056] The sample test refers to performing a point load test and making corrections based on the rock sample data in the tunnel measurement data to obtain the uniaxial compressive strength of the rock. The point load test means clamping the rock sample between two spherical loading cones and applying a load until the rock sample is fractured. The obtained uniaxial compressive strength is marked as ,in, represents the uncorrected uniaxial compressive strength of rock, , P is the peak load when pressurized, D is the distance between the two cone end points, is the size effect correction factor, and , m is the correction index, which can be 0.40-0.45, and is 0.42 in this embodiment. In this embodiment, it represents the uniaxial compressive strength of a standard rock sample with a diameter of 50 mm;

[0057] Capture information on the obtained fracture close-up data to obtain fracture geological parameters;

[0058] It should be further explained that, in a specific implementation process, the information capture means calculating the crack information in the crack close-up data, wherein the crack information includes but is not limited to the coordinates of the crack centroid, the crack length, and the crack inclination. The specific process includes:

[0059] The ratio of photo pixels to length is obtained by selecting the flake-like reference object in each crack image. Then, the coordinate values ​​of the two endpoints of the crack are read from the current coordinate system by clicking the two endpoints of the crack with the mouse. The coordinates of the centroid, crack length, crack inclination and other parameters are obtained based on the coordinates and ratio of the two endpoints. The centroid represents the center point where the cracks intersect.

[0060] Based on the fracture geological parameters, the rock mass is summarized to obtain the incompleteness index;

[0061] The index summarizes that according to the presence of a large number of various structural planes and weak interlayers in the roadway rock mass, there is a significant difference between the rock mass strength and the rock strength, so it is necessary to classify the rock mass quality and then select parameters such as bearing capacity according to the quality grade. The incompleteness index includes the volume joint number, rock quality index and rock mass integrity index;

[0062] Label the volume joint number as ,in, , K represents the conversion factor, n is the number of joints in a single crack image; A is the crack image size;

[0063] The rock quality index is labeled as RQD, where In this embodiment, the volume joint cardinality estimation method is adopted. <4.5, RQD=100, when >35, RQD=0;

[0064] The rock mass integrity index is marked as ,in, , is the rock mass longitudinal wave velocity, is the longitudinal wave velocity of the rock (block) in the room;

[0065] The rock mass is graded according to the obtained incompleteness index to obtain the rock mass quality index;

[0066] It should be further explained that, in the specific implementation process, the quality classification of the rock mass is a comprehensive reflection of the quality of the engineering rock mass. In actual engineering, the rock mass is generally graded according to various specifications, and then the bearing capacity and other parameters are selected accordingly according to the quality grade. In this embodiment, the RMR method is used to grade the rock mass around the target mine tunnel to obtain the rock mass quality index, which includes uniaxial compressive strength, rock quality index, structural plane spacing, structural plane conditions, groundwater conditions, and the relationship between the structural plane occurrence and the engineering direction. The obtained rock mass quality index is marked as RMR, where RMR = +RQD+R3+R4+R5+R6, R3 is the distance between structural planes, R4 is the structural plane condition, R5 is the groundwater condition, and R6 is the relationship between the structural plane occurrence and the project direction;

[0067] The obtained rock mass quality indicators are graded and scored to obtain rock mass quality grades, wherein the rock mass quality grades include first quality grade, second quality grade, third quality grade, fourth quality grade, and fifth quality grade;

[0068] The grading score indicates that the rock quality index is graded according to the calculation results of the rock quality index to obtain the rock quality grade. In this embodiment, the rock quality index is divided into five grade intervals. When 100 ≥ RMR ≥ 81, it is recorded as the first quality grade, indicating very good rock mass; when 80 ≥ RMR ≥ 61, it is recorded as the second quality grade, indicating good rock mass; when 60 ≥ RMR ≥ 41, it is recorded as the third quality grade, indicating average rock mass; when 40 ≥ RMR ≥ 21, it is recorded as the fourth quality grade, indicating poor rock mass; when 21 > RMR, it is recorded as the fifth quality grade, indicating very poor rock mass.

[0069] Construct a tunnel cross-section model and collect parameters to obtain the original rock mass mechanical parameters. Use the tunnel cross-section model to perform simulation construction and obtain a simulation parameter library of the same level. The specific steps include:

[0070] Obtaining a target mine tunnel, expanding the model of the obtained target mine tunnel, and obtaining a tunnel cross-section model;

[0071] The model expansion means converting the actual target mine roadway into a three-dimensional model, namely a roadway cross-section model, which is used to provide a simulation space and perform simulations at different quality levels to obtain the best support solution. The function and structure of the roadway cross-section model are exactly the same as those of the actual target mine roadway.

[0072] Based on the tunnel measurement points, the tunnel cross-section model is graded and matched according to the rock mass quality grade to obtain the measurement point quality grade, which includes the first-level measurement point grade, the second-level measurement point grade, the third-level measurement point grade, the fourth-level measurement point grade, and the fifth-level measurement point grade;

[0073] The grade area matching means that in the roadway cross-section model, the rock mass quality grade at the roadway measurement point is matched, and the rock mass quality grade at the roadway measurement point is obtained, which is recorded as the measurement point quality grade. It means that each roadway measurement point obtains the corresponding rock mass quality grade, which is used to set the corresponding support scheme parameters according to the rock mass quality grade. According to the requirements of the roadway measurement point, the most suitable support parameters are obtained, which can effectively ensure the stability of the roadway support structure and set the support scheme according to local conditions.

[0074] The roadway cross-section model is distinguishably marked according to the obtained measurement point quality grades to obtain a marked roadway model, wherein the distinguishing marks indicate that the measurement point quality grades corresponding to the roadway measurement points are distinguishably marked in the roadway cross-section model, for example, by using different colors to distinguish them, so that the user can quickly distinguish the rock mass quality grades of the roadway measurement points in the marked roadway model; for example, roadway measurement points of the first quality grade are marked in gray, roadway measurement points of the second quality grade are marked in green, roadway measurement points of the third quality grade are marked in blue, roadway measurement points of the fourth quality grade are marked in orange, and roadway measurement points of the fifth quality grade are marked in red;

[0075] Parameters of the obtained marked roadway model are collected to obtain original rock mass mechanical parameters, and the obtained original rock mass mechanical parameters are associated with the corresponding study area;

[0076] The parameter collection means collecting the mechanical parameters of the rock mass in the marked tunnel model, wherein the original rock mass mechanical parameters include but are not limited to rock strength, rock elastic modulus and Poisson's ratio, and stress data;

[0077] Based on the original rock mass mechanical parameters, the support numerical simulation of the obtained marked tunnel model is carried out to obtain support simulation data;

[0078] It should be further explained that, in the specific implementation process, the numerical simulation represents the implementation of different support mode values ​​under different rock mass quality levels in the marked tunnel model, which is the support simulation data. According to the changes in the original rock mass mechanical parameters corresponding to the different support mode data, the effectiveness of the support mode values ​​is analyzed, and the most suitable support scheme is finally generated. In this embodiment, plain shotcrete support and anchor mesh support are used for tunnel support. The advantages of plain shotcrete support and anchor mesh support are used to find the most suitable support application method in the marked tunnel model, that is, different thicknesses of shotcrete, different specifications of anchor rods and anchor meshes are designed, and the thickness of shotcrete, anchor rod specifications, and anchor mesh specifications are collected to obtain all possible support simulation data. Then, the support simulation data are screened to obtain the optimal result, which is the optimal support method.

[0079] The numerical simulation process includes:

[0080] According to the obtained study area level corresponding to the study area, different shotcrete data and different anchor mesh specifications are set according to the study area level to represent all possible support values ​​applied to the marked roadway model, and the set data are recorded, which is the support simulation data; for example, the rock mass quality grade at the roadway measurement point c1 of the marked roadway model is the secondary measurement point level, then based on the secondary measurement point level, taking the simulation of plain shotcrete support as an example, different support simulation data are set, such as concrete strength, concrete thickness and different combinations of shotcrete parameters, and different combinations are recorded to obtain support simulation data. When the concrete strength is q1 and the concrete thickness is q2, the support simulation data is obtained. The concrete strength is h1 and the injection parameter is p1, which is a set of support simulation data. It can be changed to concrete strength q2, concrete thickness h2 and injection parameter p2, or concrete strength q1, concrete thickness h2 and injection parameter p3, which is a set of support simulation data. That is, under the condition of meeting the secondary measurement point level, all possible support simulation data within the range are set. Among them, "under the condition of meeting the secondary measurement point level" means that the support simulation data conforms to the secondary rock mass quality of the measurement point, that is, the maximum strength of the plain injection that the secondary rock mass quality level can withstand will not destroy the rock mass quality, but only protect the rock mass quality;

[0081] Constructing an initial parameter level library based on the obtained measurement point quality grades, wherein the constructed initial parameter level library is used to store original rock mass mechanical parameters, measurement point quality grades, and support simulation data, and the initial parameter level library includes a first original level library, a second original level library, a third original level library, a fourth original level library, and a fifth original level library;

[0082] Based on the quality level of the measurement point, the initial parameter level library is summarized at the same level according to the obtained support simulation data to obtain a simulation parameter library of the same level, wherein the simulation parameter library of the same level includes a first parameter level library, a second parameter level library, a third parameter level library, a fourth parameter level library, and a fifth parameter level library;

[0083] It should be further explained that, in the specific implementation process, the same-level summarization means collecting support simulation data of the same rock quality level according to the quality level of the measurement point, and then the corresponding level of tunnel measurement points, original rock mechanical parameters, and support simulation data are transferred to the corresponding initial parameter level library to obtain a same-level simulation parameter library, which is used to intuitively obtain different support simulation data that can be set in the marked tunnel model under different rock quality levels; for example, the tunnel measurement points corresponding to the secondary quality level, the original rock mechanical parameters of the tunnel measurement points, and all support simulation data that can be obtained by support data simulation at the tunnel measurement points are uploaded to the second original level library to obtain the second parameter level library.

[0084] Perform simulation comparison based on the same-level simulation parameter library to obtain parameter simulation results. Based on the parameter simulation results, perform defect enhancement on the tunnel measurement points to obtain the optimal support solution. The specific steps include:

[0085] The marked roadway model is simulated and run according to the obtained simulation parameter library of the same level, and the parameters of the marked roadway model during the simulation are collected to obtain the operating rock mass mechanical parameters;

[0086] It should be further explained that, in a specific implementation process, the simulation operation means applying corresponding support simulation data to each tunnel measurement point in the marked tunnel model according to the obtained simulation parameter library of the same level. Each time the support simulation data is applied, the rock mechanical parameters are recorded and recorded as the running rock mechanical parameters. For example, when the tunnel measurement point is supported by anchor mesh support, each time the anchor mesh support parameters are applied, the rock mechanical parameters of the tunnel measurement point are recorded, such as the anchor rod parameters and the steel mesh parameters. When the anchor rod parameters are m1 and the steel mesh parameters are j1, the rock mechanical parameters at that time are recorded. When the anchor rod parameters are m1 and the steel mesh parameters are j2, the rock mechanical parameters at that time are recorded. When the anchor rod parameters are m1 and the steel mesh parameters are j3, the rock mechanical parameters at that time are recorded. . .

[0087] Based on the marked tunnel model, the effects of the original rock mass mechanical parameters obtained are compared with the operating rock mass mechanical parameters to obtain parameter simulation results. The parameter simulation results represent the improvement ratio of the original rock mass mechanical parameters compared to the operating rock mass mechanical parameters after the support simulation data is applied. For example, if the stress data in the operating rock mass mechanical parameters is improved by 5% compared to the stress data in the original rock mass mechanical parameters, then +5% represents the parameter simulation result under this support simulation data.

[0088] Furthermore, the effect comparison indicates that the stability and safety of the tunnel measurement point after the support parameters are applied are compared with the effect before the application, and the parameter simulation result is obtained. The parameter simulation result indicates whether the support simulation data applied at the tunnel measurement point is effective, that is, whether the stability and support of the measurement point can be improved to protect the safety of the tunnel. Then, it is necessary to compare the rock mechanics parameters before and after the support simulation data are applied, and divide the effect of the applied support simulation data according to the comparison result to obtain the parameter simulation result. The divided effect indicates that the operating rock mechanics parameters after the support simulation data are applied are increased compared with the original rock mechanics parameters. The percentage of addition is the degree of support effect. For example, by analyzing the changes in stress data before and after the application of support simulation data at the measurement point of the roadway, it can be concluded whether the applied support simulation data is effective. That is, according to the stress after applying the support simulation data, compared with the original stress data, the stress is increased by e%. This increased e% is the simulation effect, that is, the parameter simulation result, which means that the bearing capacity of the roadway measurement point is effectively improved on the original basis to improve the overall stability. If after applying a support simulation data, compared with the original stress data, it is increased by f%, e<f, then it means that the effect of the second support simulation data is better than the first support simulation data;

[0089] An effect threshold is set according to the obtained parameter simulation results, wherein the effect threshold represents a threshold set according to the improvement effect of each rock mass mechanical parameter and is used to determine whether the support simulation data is effective;

[0090] The parameter simulation results are checked for excellence through the effect threshold, and insignificant simulation results are obtained. The tunnel measurement points corresponding to the insignificant results are recorded as measurement points to be optimized;

[0091] It should be further explained that, in the specific implementation process, the excellent screening means screening out the support simulation data with unqualified effects in the parameter simulation results according to the effect threshold, and optimizing the support simulation data of the point. It is no longer a single plain shotcrete or anchor mesh support, but a double support method combining plain shotcrete and anchor mesh support. Among them, "unqualified support simulation data" means the support simulation data corresponding to the parameter simulation result being less than the effect threshold, and is recorded as an insignificant simulation result, indicating that the effect formed by the support parameters in this case is not significant, or not good enough, then it is necessary to optimize and select better support parameters to improve the stability of the roadway, then the roadway measurement point corresponding to this support simulation data is the measurement point that needs to be optimized, that is, the measurement point to be optimized;

[0092] Based on the simulation parameter library of the same level, support optimization is performed on the optimal measurement points to obtain optimized support data. The marked roadway model is optimized and simulated based on the optimized support data. Parameters of the marked roadway model during the optimization simulation are collected to obtain optimized rock mass mechanical parameters.

[0093] Compare the effects of the optimized rock mass mechanics parameters based on the original rock mass mechanics parameters to obtain the optimized parameter simulation results, sort the optimized parameter simulation results to obtain the best simulated support data, and update and screen the simulation parameter library of the same level based on the best simulated support data to obtain the best support scheme;

[0094] It should be further explained that, in the specific implementation process, the support optimization means that all possible support simulation data are included in the simulation parameter library of the same level, and there are corresponding support simulation data at the measurement point to be optimized. After removing the insignificant simulation results, the combination of shotcrete support and anchor mesh support is carried out, that is, shotcrete support is carried out first, and then anchor mesh support is carried out, then all possible support simulation data are generated and applied to the marked tunnel model for secondary simulation operation, and at the same time, the rock mechanics parameters are collected and recorded as optimized rock mechanics parameters, and the effect comparison process of the rock mechanics parameters before and after the support simulation data is applied is repeated to obtain the optimized parameter simulation results, and the group with the best effect is screened out from all the optimized parameter simulation results. The support simulation data is recorded as the best simulated support data. For the remaining tunnel measurement points in the marked tunnel model, the support simulation data with the best parameter simulation results are also selected in the simulation parameter library of the same level. Together with the best simulated support data of the measurement points to be optimized, they constitute the best support scheme of the marked tunnel model, that is, the best support scheme of the target mine tunnel. On the premise that a single support scheme can meet the stability requirements, a double support structure is implemented for weak measurement points to avoid unnecessary excessive support, save materials and construction costs, and improve economic benefits. It can also be adapted to local conditions and adjusted according to the actual situation during the construction process. The best support scheme is generated according to the needs of the measurement points to improve the flexibility and adaptability of construction.

[0095] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A mine tunnel support structure optimization method based on artificial intelligence, characterized in that: The following steps are involved: Step S1: collecting tunnel measurement data of tunnel measurement points; Step S2: capturing and extracting the tunnel measurement data to obtain fracture geological parameters, and classifying and summarizing the fracture geological parameters to obtain the rock mass quality grade; Step S3: Construct a tunnel cross-section model and collect parameters to obtain the original rock mass mechanical parameters, perform simulation construction through the tunnel cross-section model, and obtain a simulation parameter library of the same level; Step S4: Perform simulation comparison based on the simulation parameter library of the same level to obtain parameter simulation results, and enhance the defects of the tunnel measurement points based on the parameter simulation results to obtain the best support solution.

2. The method for optimizing mine tunnel support structure on demand based on artificial intelligence according to claim 1, characterized in that: The process of collecting roadway measurement data includes: Lay and capture the target mine tunnel and obtain tunnel measurement points; The target acquisition terminal is set according to the obtained tunnel measurement points, and information is collected through the target acquisition terminal to obtain tunnel measurement data.

3. The method for optimizing mine tunnel support structure on demand based on artificial intelligence according to claim 1, characterized in that: The process of capturing and extracting tunnel measurement data includes: Acquire tunnel measurement data, perform close-up capture of tunnel measurement data, obtain close-up data of cracks, conduct sample tests on tunnel measurement data, and obtain uniaxial compressive strength; The obtained fracture close-up data is captured to obtain fracture geological parameters.

4. The method for optimizing mine tunnel support structure on demand based on artificial intelligence according to claim 1, characterized in that: The process of classifying fracture geological parameters includes: Based on the fracture geological parameters, the rock mass is summarized to obtain the incompleteness index; The rock mass is graded according to the obtained incompleteness index to obtain the rock mass quality index; The obtained rock mass quality indicators are graded and scored to obtain the rock mass quality grade.

5. The method for optimizing mine tunnel support structure on demand based on artificial intelligence according to claim 2, characterized in that: The process of constructing a tunnel cross-section model and collecting parameters to obtain the original rock mass mechanical parameters includes: Expand the model of the target mine tunnel to obtain the tunnel cross-section model; Based on the rock mass quality grade of the tunnel measurement points, the tunnel cross-section model is matched with the grade area to obtain the quality grade of the measurement points; The tunnel cross-section model is distinguished and marked according to the quality level of the measurement points to obtain the marked tunnel model, and the parameters of the marked tunnel model are collected to obtain the original rock mass mechanical parameters.

6. The method for optimizing mine tunnel support structure on demand based on artificial intelligence according to claim 5, characterized in that: The process of constructing a simulation model through a tunnel cross-section model and obtaining a simulation parameter library of the same level includes: Based on the original rock mass mechanical parameters, the support numerical simulation of the marked tunnel model is carried out to obtain support simulation data; Construct an initial parameter level library based on the quality levels of the measured points; The initial parameter level library is summarized at the same level according to the support simulation data to obtain the simulation parameter library of the same level.

7. The method for optimizing mine tunnel support structure on demand based on artificial intelligence according to claim 6, characterized in that: The process of comparing simulations based on the same level simulation parameter library includes: The marked roadway model is simulated and run according to the obtained simulation parameter library of the same level, and the parameters of the marked roadway model during the simulation are collected to obtain the operating rock mass mechanical parameters; Based on the marked tunnel model, the effects of the operating rock mass mechanical parameters are compared with the original rock mass mechanical parameters to obtain the parameter simulation results.

8. The method for optimizing mine tunnel support structure on demand based on artificial intelligence according to claim 7, characterized in that: The process of enhancing the defects at the tunnel measurement points based on the parameter simulation results and obtaining the optimal support solution includes: According to the parameter simulation results, the effect threshold is set, and the parameter simulation results are checked for excellence through the effect threshold to obtain insignificant simulation results. The tunnel measurement points corresponding to the insignificant results are recorded as measurement points to be optimized; Based on the simulation parameter library of the same level, support optimization is performed on the optimal measurement points to obtain optimized support data. The marked roadway model is optimized and simulated based on the optimized support data. Parameters of the marked roadway model during the optimization simulation are collected to obtain optimized rock mass mechanical parameters. The effects of the optimized rock mechanics parameters are compared based on the original rock mechanics parameters obtained to obtain the optimized parameter simulation results. The optimized parameter simulation results are ranked by effect to obtain the best simulated support data. Based on the best simulated support data, the simulation parameter library of the same level is updated and screened to obtain the best support scheme.

Images