An intelligent optimization method and system for smooth blasting holes in tunnels in layered rock masses

Abstract

The present invention provides a method and system for intelligent optimization of smooth-surface blasting holes in tunnels in layered rock masses, belonging to the field of tunnel engineering and blasting technology. The method includes the steps of collecting geological data of layered rock masses, evaluating and extracting features, constructing a multi-objective optimization function, solving with an intelligent optimization algorithm, and providing construction feedback and dynamic optimization. A variety of geological survey equipment is used to acquire data, and after evaluation and processing with a support vector machine model, an optimization function that considers multiple indicators is constructed. The genetic algorithm and particle swarm algorithm are combined to solve the optimal combination of blasthole parameters, and finally the blasting effect is fed back to dynamically adjust the optimization scheme. The present invention adopts the above-mentioned method and system for intelligent optimization of smooth-surface blasting holes in tunnels in layered rock masses, which can improve the blasting effect, reduce over-excavation and under-excavation and concrete consumption, reduce construction costs, enhance construction safety and efficiency, and have significant economic and social benefits for the construction of layered rock tunnel projects.

Description

Technical Field

[0001] The present invention belongs to the field of tunnel engineering and blasting technology, and in particular to an intelligent optimization method and system for smooth blasting holes in a tunnel in layered rock mass. Background Art

[0002] In tunnel construction, smooth blasting technology is widely used to control the excavation profile, reduce over-excavation and under-excavation, and minimize disturbance damage to the surrounding rock. However, layered rock masses have unique geological characteristics, such as variations in the strike and inclination of bedding planes, differences in rock thickness, and differences in rock strength and integrity. These characteristics pose challenges to traditional smooth blasting technology. Existing technologies often rely on experience and analogy to design blasthole parameters, making it difficult to accurately account for the complexity of layered rock masses. This leads to unstable blasting results, severe over-excavation and under-excavation, poor rock wall flatness, and frequent local collapses, significantly increasing construction costs and impacting project quality and safety.

[0003] Current optimization methods focus on a single blasting performance indicator, such as overbreak or underbreak or rock face smoothness, without comprehensively considering multiple factors affecting blasting effectiveness. This single-target optimization approach fails to balance blasting effectiveness with economic benefits, and thus struggles to meet the comprehensive requirements of modern tunnel engineering for high quality, high efficiency, low cost, and high safety.

[0004] Therefore, to address the challenges of smooth blasting in layered rock tunnels, an intelligent optimization method and system are urgently needed. This system must be able to accurately analyze the geological characteristics of the layered rock mass, comprehensively consider multiple factors affecting blasting effectiveness, and intelligently optimize blasthole parameters, thereby improving blasting quality, reducing construction costs, and ensuring the safety and efficiency of tunnel construction. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent optimization method and system for smooth blasting holes in layered rock tunnels, which can improve blasting effects, reduce overbreak and underbreak and concrete consumption, reduce construction costs, enhance construction safety and efficiency, and have significant economic and social benefits for layered rock tunnel engineering construction.

[0006] To achieve the above object, the present invention provides an intelligent optimization method for smooth blasting holes in a tunnel in layered rock mass, comprising the following steps:

[0007] Step S1: using a variety of geological survey equipment to obtain geological data of the tunnel face in layered rock mass;

[0008] Step S2: perform quality assessment and feature extraction on the acquired geological data, determine the reliability and applicability of the data based on a preset data evaluation model, remove abnormal data and noise interference, and extract geological features related to tunnel smooth blasting;

[0009] Step S3: constructing a multi-objective optimization function based on the evaluated geological data and the extracted geological features;

[0010] Step S4: using an intelligent optimization algorithm to solve the constructed multi-objective optimization function and search for the optimal combination of blasthole parameters;

[0011] Step S5: Apply the solved optimal blasthole parameters to the actual tunnel smooth blasting construction. At the same time, monitor and evaluate the actual blasting effect after the blasting. Feedback the evaluation results to the multi-objective optimization function construction link, update the geological data and adjust the weight coefficients to achieve dynamic optimization of subsequent blasthole optimization plans.

[0012] Preferably, in step S1, the geological survey equipment includes geological radar, three-dimensional laser scanner, sonic detector and drilling coring equipment. By using a variety of equipment in combination, detailed geological information of the layered rock mass can be obtained from different angles and depths, thereby improving the comprehensiveness and accuracy of the geological data.

[0013] Preferably, in step S1, the geological data include bedding plane strike, dip, rock mass integrity parameters, joint development degree, and surrounding rock strength parameters.

[0014] Preferably, in step S2, the data evaluation model is established based on a support vector machine, and the specific process is as follows:

[0015] Data preparation: Collect historical geological data and corresponding blasting effect data, use the geological data as input feature vectors and the blasting effect data as output labels, and standardize the data to make its mean 0 and standard deviation 1;

[0016] Model training: The processed data is divided into a training set and a test set in a ratio of 7:3. The radial basis function is used as the kernel function of the support vector machine. The support vector machine model is trained using the training set, and the performance of the model is optimized by adjusting the penalty parameters and kernel function parameters.

[0017] Model evaluation and optimization: Use the test set to evaluate the trained support vector machine model, calculate the model's accuracy, recall rate, and F1 value, and adjust and optimize the model parameters based on the evaluation results;

[0018] New data evaluation and processing: After the newly acquired geological data is standardized, it is input into the trained support vector machine model. The model outputs the quality assessment results of the data. Based on the assessment results, the new data is preprocessed to extract features related to the bedding plane strike, dip, rock mass integrity parameters, joint development degree, and surrounding rock strength parameters in the geological data.

[0019] Preferably, in step S3, the multi-objective optimization function comprehensively considers the over-excavation and under-excavation after blasting, the residual trace rate of the surrounding holes, the flatness of the rock wall, and the excess consumption of shotcrete, and introduces a cost weight factor to make the optimization scheme take into account the economy. Its mathematical expression is:

[0020] ;

[0021] in, represents the multi-objective optimization function, represents the blasthole parameter vector, including blasthole position, spacing, angle and depth, Indicates over-excavation or under-excavation. Indicates the residual trace rate of peripheral eyes, Indicates the flatness of the rock wall. Indicates the excess consumption of shotcrete. 、 、 、 Indicates the weight coefficient corresponding to each target;

[0022] ;

[0023] ;

[0024] ;

[0025] ;

[0026] in, Indicates the actual blasting The distance from the excavation contour line to the design contour line of the measurement point, Indicates the distance between corresponding points on the design contour line, represents the number of measurement points, Indicates the number of traces remaining around the eye after blasting. Indicates the total number of peripheral eyes, It represents the standard deviation of the distance between the actual excavation contour line of the rock wall after blasting and the designed contour line. Indicates the The actual usage of shotcrete Indicates the Theoretical design quantity of segmented shotcrete, Indicates the number of sections of shotcrete;

[0027] The multi-objective optimization function also incorporates blasting vibration speed and frequency as constraints into the objective function. The constraints are as follows:

[0028] ;

[0029] ;

[0030] in, Indicates the maximum speed of blasting vibration, Indicates the upper limit of the permissible blasting vibration speed, Indicates the lowest frequency of blasting vibration, Indicates the lower limit of the allowable blasting vibration frequency.

[0031] Preferably, the weight coefficient is determined based on the hierarchical analysis method combined with the expert scoring method, and is dynamically adjusted according to the geological characteristics.

[0032] Preferably, in step S4, the intelligent optimization algorithm combines the genetic algorithm and the particle swarm algorithm to explore the optimal solution that meets the constraints in the solution space by simulating the biological evolution process and the group intelligence behavior. The specific implementation steps of the intelligent optimization algorithm are as follows:

[0033] Step S41: Initialize the population, randomly generate a combination of blasthole parameters as the initial solution, and encode each solution. Use real number encoding to convert the blasthole position, spacing, angle and depth into a real number vector;

[0034] Step S42: Calculate the objective function value corresponding to each solution. Substitute the decoded blasthole parameters into the constructed multi-objective optimization function to obtain the function value of each blasting effect index and the comprehensive objective function value as the basis for evaluating the quality of the solution.

[0035] Step S43: selection operation, adopting the tournament selection strategy to select parent individuals from the current population and enter the crossover and mutation operation phase;

[0036] Step S44: crossover operation, combining the simulated binary crossover operator of the genetic algorithm and the position update strategy of the particle swarm algorithm to perform crossover combination on the parent individuals to generate new offspring individuals;

[0037] Step S45: mutation operation, using a polynomial mutation operator to perform local search on offspring individuals, perturb the blasthole parameters, increase the diversity of solutions, prevent the algorithm from falling into local optimality, and control the length of the mutation step by setting the mutation distribution index;

[0038] Step S46: Update the population, merge the parent individuals and the offspring individuals, sort them according to the objective function value, and generate a new population as the basis for the next iteration. At the same time, check the constraints of the solutions in the new population, eliminate solutions that do not meet the engineering requirements, and use a repair strategy to process the out-of-bounds variables to make them meet the constraints.

[0039] Step S47, repeating steps S42 to S46 until the preset iteration termination condition is reached, and outputting the optimal blasthole parameter combination and the corresponding objective function value and each blasting effect index value.

[0040] Preferably, the termination conditions include: reaching the maximum number of iterations, the convergence accuracy of the objective function value, or the blasting effect meeting the engineering requirements.

[0041] Preferably, in step S4, the constraints include a reasonable range constraint of blasthole spacing, an allowable deviation constraint of blasthole angles, and a blasting safety distance constraint;

[0042] The reasonable range constraints of blasthole spacing are as follows:

[0043] ;

[0044] The allowable deviation constraints of the blasthole angle are as follows:

[0045] ;

[0046] The blasting safety distance constraints are as follows:

[0047] ;

[0048] ;

[0049] in, Indicates the The actual blasthole spacing of the blasthole type, Indicates the The minimum permissible blasthole spacing for similar blastholes, Indicates the The maximum permissible blasthole spacing for similar blastholes, , Indicates the groove eye. Indicates auxiliary eye, Indicates peripheral eyes, Indicates the design angle of the blasthole. Indicates the actual angle, Indicates the allowable deviation, Indicates the distance between the blasting point and the protected target. Indicates safe distance. represents the coefficient related to geological conditions and blasting type, Indicates the maximum dosage of a single segment.

[0050] The present invention also provides an intelligent optimization system for smooth blasting holes in tunnels in layered rock mass, comprising:

[0051] The data acquisition module is used to obtain geological data of the tunnel face in layered rock mass and the surrounding rock in front of it using a variety of geological survey equipment;

[0052] The data processing module is used to perform quality assessment and feature extraction on the acquired geological data. It determines the reliability and applicability of the data based on a preset data evaluation model, removes abnormal data and noise interference, and extracts geological features related to tunnel smooth blasting.

[0053] The optimization module is used to construct a multi-objective optimization function based on the evaluated geological data and extracted geological features, and use an intelligent optimization algorithm to solve the constructed multi-objective optimization function to search for the optimal combination of blasthole parameters;

[0054] The construction feedback module is used to apply the solved optimal blasthole parameters to the actual tunnel smooth blasting construction. At the same time, the actual blasting effect is monitored and evaluated after the blasting. The evaluation results are fed back to the multi-objective optimization function construction link, the geological data is updated, and the weight coefficients are adjusted to achieve dynamic optimization of subsequent blasthole optimization plans.

[0055] Therefore, the present invention adopts the above-mentioned intelligent optimization method and system for smooth blasting holes in tunnels in layered rock masses, and the beneficial technical effects are as follows:

[0056] (1) By using a variety of geological survey equipment to obtain detailed geological data of the tunnel face in layered rock, including bedding plane strike, inclination, rock integrity parameters, joint development degree, and surrounding rock strength parameters, and combining the support vector machine model to perform data quality assessment and feature extraction, it is possible to accurately remove abnormal data and noise interference, extract geological features closely related to smooth blasting, and provide a reliable data basis for subsequent optimization calculations, thereby effectively improving the prediction accuracy and reliability of blasting effects.

[0057] (2) The constructed multi-objective optimization function comprehensively considers multiple factors such as the amount of over-excavation and under-excavation after blasting, the residual trace rate of the surrounding holes, the flatness of the rock wall, and the excess consumption of shotcrete. It also introduces a cost weight factor and incorporates the blasting vibration speed and frequency as constraints into the optimization process, achieving a comprehensive quantification and integrated optimization of the blasting effect. Compared with the traditional single-objective optimization method, this method can ensure the blasting effect while taking into account the control of construction costs, making the optimization scheme more economical and practical.

[0058] (3) An intelligent optimization algorithm combining genetic algorithms and particle swarm optimization effectively explores the optimal solution that satisfies the constraints in the solution space by simulating the biological evolution process and swarm intelligence behavior. This algorithm can dynamically adjust the optimization direction of blasting parameters according to geological characteristics, avoiding falling into local optimal solutions, improving optimization efficiency and global search capabilities. At the same time, the construction feedback link feeds the actual blasting effect evaluation results back to the optimization model for updating geological data and adjusting weight coefficients, realizing dynamic adjustment of subsequent blast hole optimization plans, further enhancing the adaptability and intelligence of the optimization method to complex layered rock geological conditions.

[0059] (4) The intelligent optimization system of the present invention organically integrates data collection, processing, optimization, and construction feedback to form a complete solution for optimizing blastholes for smooth blasting in layered rock tunnels. The system can effectively reduce over-excavation and under-excavation, improve the residual trace rate of peripheral holes and the smoothness of rock walls, and reduce concrete consumption, thereby significantly improving the construction quality and efficiency of smooth blasting in tunnels, reducing construction costs, and enhancing construction safety. It has significant economic and social benefits for the construction of layered rock tunnels. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 This is a flow chart of an intelligent optimization method for smooth blasting holes in a tunnel in layered rock mass according to the present invention;

[0061] Figure 2 Flowchart constructed for multi-objective optimization function;

[0062] Figure 3 A flow chart for determining weight coefficients;

[0063] Figure 4 Flowchart for solving the constructed multi-objective optimization function by the intelligent optimization algorithm. DETAILED DESCRIPTION

[0064] The technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.

[0065] Unless otherwise defined, technical or scientific terms used in the present invention shall have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

[0066] Example 1

[0067] like Figure 1 As shown, an intelligent optimization method for smooth blasting holes in a tunnel in layered rock mass includes the following steps:

[0068] Step S1: using a variety of geological survey equipment to obtain geological data of the tunnel face in layered rock mass;

[0069] Geological survey equipment includes geological radar, 3D laser scanner, sonic detector and drilling coring equipment.

[0070] Geological radar: IDS SIR-4000, scanning distance 10 cm × 10 cm, frequency 1 GHz;

[0071] 3D laser scanner: RIEGL VZ-4000, point cloud density 50 pts / m 2 ;

[0072] Acoustic wave detector: PUNDIT, test interval 50cm, wave velocity range 1000~4500m / s;

[0073] Drilling and coring: 75mm coring drill, drilling depth 5m, RQD (core integrity) test.

[0074] The geological data include bedding plane strike, dip, rock mass integrity parameters, joint development degree and surrounding rock strength parameters.

[0075] Step S2: perform quality assessment and feature extraction on the acquired geological data, determine the reliability and applicability of the data based on a preset data evaluation model, remove abnormal data and noise interference, and extract geological features related to tunnel smooth blasting.

[0076] The data evaluation model is established based on the support vector machine, and its specific process is as follows:

[0077] Data preparation: Collect historical geological data and corresponding blasting effect data, use the geological data as input feature vectors and the blasting effect data as output labels, and standardize the data to make its mean 0 and standard deviation 1;

[0078] Model training: The processed data is divided into a training set and a test set in a ratio of 7:3. The radial basis function is used as the kernel function of the support vector machine. The support vector machine model is trained using the training set, and the performance of the model is optimized by adjusting the penalty parameters and kernel function parameters.

[0079] Model evaluation and optimization: Use the test set to evaluate the trained support vector machine model, calculate the model's accuracy, recall rate, and F1 value, and adjust and optimize the model parameters based on the evaluation results;

[0080] New data evaluation and processing: After the newly acquired geological data is standardized, it is input into the trained support vector machine model. The model outputs the quality assessment results of the data. Based on the assessment results, the new data is preprocessed to extract features related to the bedding plane strike, dip, rock mass integrity parameters, joint development degree, and surrounding rock strength parameters in the geological data.

[0081] Step S3: construct a multi-objective optimization function based on the evaluated geological data and the extracted geological features.

[0082] like Figure 2 As shown in Figure 2, the multi-objective optimization function comprehensively considers the over-excavation and under-excavation after blasting, the residual trace rate of the surrounding holes, the flatness of the rock wall, and the excess consumption of shotcrete, and introduces the cost weight factor. Its mathematical expression is:

[0083] ;

[0084] in, represents the multi-objective optimization function, represents the blasthole parameter vector, including blasthole position, spacing, angle and depth, Indicates over-excavation or under-excavation. Indicates the residual trace rate of peripheral eyes, Indicates the flatness of the rock wall. Indicates the excess consumption of shotcrete. 、 、 、 Indicates the weight coefficient corresponding to each target;

[0085] ;

[0086] ;

[0087] ;

[0088] ;

[0089] in, Indicates the actual blasting The distance from the excavation contour line to the design contour line of the measurement point, Indicates the distance between corresponding points on the design contour line, represents the number of measurement points, Indicates the number of traces remaining around the eye after blasting. Indicates the total number of peripheral eyes, It represents the standard deviation of the distance between the actual excavation contour line of the rock wall after blasting and the designed contour line. Indicates the The actual usage of shotcrete Indicates the Theoretical design quantity of segmented shotcrete, Indicates the number of sections of shotcrete;

[0090] The multi-objective optimization function also incorporates the blasting vibration velocity as a constraint into the objective function. The constraints are as follows:

[0091] ;

[0092] in, Indicates the maximum speed of blasting vibration, Indicates the upper limit of the permissible blasting vibration speed.

[0093] like Figure 3 As shown in the figure, the weight coefficient is determined based on the hierarchical analysis method combined with the expert scoring method, and is dynamically adjusted according to the geological characteristics:

[0094] 1. Establish a hierarchical model: Take the blasting effect index and cost factors as the target layer, and the over-excavation and under-excavation, the residual trace rate of the surrounding holes, the rock wall flatness, the excess consumption of shotcrete, and the blasting vibration speed and frequency as the criterion layer to construct a hierarchical model.

[0095] 2. Construct a judgment matrix: Invite experts in the field of tunnel blasting engineering to compare the factors in the criterion layer, assign corresponding values ​​according to their relative importance, and construct a judgment matrix. Indicates the The factor relative to The importance of each factor ranges from 1 to 9. =1 means that the two factors are equally important. =3 means the Factors This factor is slightly more important. =5 means the Factors This factor is obviously important. =7 means the Factors This factor is highly important. =9 means the Factors This factor is extremely important. .

[0096] 3. Calculate the weight vector: Use the eigenvalue method to solve the judgment matrix, obtain the weight vector corresponding to each factor, and perform a consistency test. The consistency test determines whether the judgment matrix has satisfactory consistency by calculating the consistency index (CI) and the random consistency index (RI). If CI / RI ≤ 0.1, the judgment matrix is ​​considered to have satisfactory consistency; otherwise, the judgment matrix needs to be adjusted.

[0097] 4. Combined with the expert scoring method: Collect the scoring results of multiple experts on the importance of each blasting effect indicator and cost factor, normalize the scoring results, and obtain the expert scoring weight of each factor.

[0098] 5. Dynamically adjust each weight based on the weight vector obtained by the hierarchical analysis method, the weight vector obtained by the expert scoring method, the extracted bedding plane strike, dip, rock mass integrity parameters, joint development degree, and surrounding rock strength parameters:

[0099] Over-excavation and under-excavation weight coefficient :

[0100] ;

[0101] Peripheral eye residual trace rate weight coefficient :

[0102] ;

[0103] Rock wall flatness weight coefficient :

[0104] ;

[0105] Excess consumption weight coefficient of shotcrete :

[0106] ;

[0107] in, 、 They represent the weight coefficients obtained by the analytic hierarchy process and the expert scoring method, respectively. 、 、 、 Both represent the basic weights of rock wall flatness obtained by the hierarchical analysis method. 、 、 、 Both represent the basic weight of rock wall flatness obtained by expert scoring method. It represents the adjustment coefficient of the influence of bedding surface on the weight of over-excavation and under-excavation. It represents the adjustment coefficient of the rock mass integrity parameters and joint development degree on the weight of the residual trace rate of the surrounding eye, Indicates the adjustment coefficient of surrounding rock strength parameters to rock wall flatness weight, Indicates the adjustment coefficient of the rock mass structure surface spacing on the excess consumption weight of shotcrete. represents the bedding plane dip, represents the maximum bedding plane dip, represents the rock mass integrity parameter, represents the maximum rock mass integrity parameter, Indicates the degree of joint development, Indicates the maximum degree of joint development, represents the surrounding rock strength parameter, represents the maximum surrounding rock strength parameter, Indicates the spacing between rock mass structural planes, Indicates the maximum spacing between rock mass structural planes.

[0108] Step S4: Using an intelligent optimization algorithm to solve the constructed multi-objective optimization function and search for the optimal combination of blasthole parameters.

[0109] like Figure 4 As shown in the figure, the intelligent optimization algorithm combines the genetic algorithm and the particle swarm algorithm. By simulating the biological evolution process and swarm intelligence behavior, it explores the optimal solution that meets the constraints in the solution space. The specific implementation steps of the intelligent optimization algorithm are as follows:

[0110] Step S41: Initialize the population, randomly generate a certain number of blasthole parameter combinations as initial solutions, and encode each solution. Use real number encoding to convert the blasthole position, spacing, angle, and depth into real number vectors;

[0111] Step S42: Calculate the objective function value corresponding to each solution. Substitute the decoded blasthole parameters into the constructed multi-objective optimization function to obtain the function value of each blasting effect index and the comprehensive objective function value as the basis for evaluating the quality of the solution.

[0112] Step S43: Selection operation, using the tournament selection strategy, selects the solution with higher fitness from the current population as the parent individual, and enters the crossover and mutation operation phase. The fitness is calculated based on the objective function value. The higher the fitness, the better the solution, and the greater the probability of being selected.

[0113] Step S44: crossover operation, combining the simulated binary crossover operator of the genetic algorithm and the position update strategy of the particle swarm algorithm to perform crossover combination on the parent individuals to generate new offspring individuals;

[0114] Step S45, mutation operation, using a polynomial mutation operator to perform local search on offspring individuals, perturbing the blasthole parameters with a certain mutation probability to increase the diversity of solutions and prevent the algorithm from falling into local optimality. By setting the mutation distribution index, the length of the mutation step is controlled;

[0115] Step S46: Update the population, merge the parent individuals and the offspring individuals, sort them according to the objective function value, select the solutions with higher fitness to form a new population, and use it as the basis for the next iteration. At the same time, check the constraints of the solutions in the new population, eliminate the solutions that do not meet the engineering requirements, and use the repair strategy to process the out-of-bounds variables so that they meet the constraints.

[0116] Step S47, repeating steps S42 to S46 until the preset iteration termination condition is reached, and outputting the optimal blasthole parameter combination and the corresponding objective function value and each blasting effect index value.

[0117] In step S4, the constraints include the reasonable range constraint of the blasthole spacing, the allowable deviation constraint of the blasthole angle, and the blasting safety distance constraint.

[0118] The reasonable range constraints of blasthole spacing are as follows:

[0119] ;

[0120] The allowable deviation constraints of the blasthole angle are as follows:

[0121] ;

[0122] The blasting safety distance constraints are as follows:

[0123] ;

[0124] ;

[0125] in, Indicates the The actual blasthole spacing of the blasthole type, Indicates the The minimum permissible blasthole spacing for similar blastholes, Indicates the The maximum permissible blasthole spacing for similar blastholes, , Indicates the groove eye. Indicates auxiliary eye, Indicates peripheral eyes, Indicates the design angle of the blasthole. Indicates the actual angle, Indicates the allowable deviation, Indicates the distance between the blasting point and the protected target. Indicates safe distance. represents the coefficient related to geological conditions and blasting type, Indicates the maximum dosage of a single segment.

[0126] Step S5: Apply the solved optimal blasthole parameters to the actual tunnel smooth blasting construction. At the same time, monitor and evaluate the actual blasting effect after the blasting. Feedback the evaluation results to the multi-objective optimization function construction link, update the geological data and adjust the weight coefficients to achieve dynamic optimization of subsequent blasthole optimization plans.

[0127] Example 2

[0128] An intelligent optimization system for smooth blasting holes in tunnels in layered rock mass, comprising:

[0129] The data acquisition module is used to obtain geological data of the tunnel face in layered rock mass and the surrounding rock in front of it using a variety of geological survey equipment;

[0130] The data processing module is used to perform quality assessment and feature extraction on the acquired geological data. It determines the reliability and applicability of the data based on a preset data evaluation model, removes abnormal data and noise interference, and extracts geological features related to tunnel smooth blasting.

[0131] The optimization module is used to construct a multi-objective optimization function based on the evaluated geological data and extracted geological features, and use an intelligent optimization algorithm to solve the constructed multi-objective optimization function to search for the optimal combination of blasthole parameters;

[0132] The construction feedback module is used to apply the solved optimal blasthole parameters to the actual tunnel smooth blasting construction. At the same time, the actual blasting effect is monitored and evaluated after the blasting. The evaluation results are fed back to the multi-objective optimization function construction link, the geological data is updated, and the weight coefficients are adjusted to achieve dynamic optimization of subsequent blasthole optimization plans.

[0133] It is worth noting that the contents not elaborated in detail in the present invention are all prior art and are well known to those skilled in the art.

[0134] Therefore, the present invention adopts the above-mentioned intelligent optimization method and system for smooth blasting holes in layered rock tunnels, which can improve blasting effects, reduce over-excavation and under-excavation and concrete consumption, reduce construction costs, enhance construction safety and efficiency, and have significant economic and social benefits for layered rock tunnel engineering construction.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent optimization method for smooth blasting holes in tunnels in layered rock mass, characterized by: The following steps are involved: Step S1: using a variety of geological survey equipment to obtain geological data of the tunnel face in layered rock mass; Step S2: perform quality assessment and feature extraction on the acquired geological data, determine the reliability and applicability of the data based on a preset data evaluation model, remove abnormal data and noise interference, and extract geological features related to tunnel smooth blasting; Step S3: constructing a multi-objective optimization function based on the evaluated geological data and the extracted geological features; Step S4: using an intelligent optimization algorithm to solve the constructed multi-objective optimization function and search for the optimal combination of blasthole parameters; Step S5: Apply the obtained optimal blasthole parameters to actual tunnel smooth blasting construction. Simultaneously, monitor and evaluate the actual blasting effect after blasting. Feedback the evaluation results to the multi-objective optimization function construction link, update the geological data and adjust the weight coefficients to achieve dynamic optimization of subsequent blasthole optimization schemes. In step S3, the multi-objective optimization function comprehensively considers the over-excavation and under-excavation after blasting, the residual trace rate of the surrounding holes, the flatness of the rock wall, and the excess consumption of shotcrete, and introduces a cost weight factor. Its mathematical expression is: ； in, represents the multi-objective optimization function, represents the blasthole parameter vector, including blasthole position, spacing, angle and depth, Indicates over-excavation or under-excavation. Indicates the residual trace rate of peripheral eyes, Indicates the flatness of the rock wall. Indicates the excess consumption of shotcrete. 、 、 、 Indicates the weight coefficient corresponding to each target; ； ； ； ； in, Indicates the actual blasting The distance from the excavation contour line to the design contour line of the measurement point, Indicates the distance between corresponding points on the design contour line, represents the number of measurement points, Indicates the number of traces remaining around the eye after blasting. Indicates the total number of peripheral eyes, It represents the standard deviation of the distance between the actual excavation contour line of the rock wall after blasting and the designed contour line. Indicates the The actual usage of shotcrete Indicates the Theoretical design quantity of segmented shotcrete, Indicates the number of sections of shotcrete; The multi-objective optimization function also incorporates the blasting vibration velocity as a constraint into the objective function. The constraints are as follows: ； in, Indicates the maximum speed of blasting vibration, Indicates the upper limit of the permissible blasting vibration speed; The weight coefficient is determined based on the analytic hierarchy process combined with the expert scoring method, and is dynamically adjusted according to geological characteristics; Among them, the weight coefficient of over-excavation and under-excavation : ； Peripheral eye residual trace rate weight coefficient : ； Rock wall flatness weight coefficient : ； Excess consumption weight coefficient of shotcrete : ； in, 、 They represent the weight coefficients obtained by the analytic hierarchy process and the expert scoring method, respectively. 、 、 、 Both represent the basic weights of rock wall flatness obtained by the hierarchical analysis method. 、 、 、 Both represent the basic weight of rock wall flatness obtained by expert scoring method. It represents the adjustment coefficient of the influence of bedding surface on the weight of over-excavation and under-excavation. It represents the adjustment coefficient of the rock mass integrity parameters and joint development degree on the weight of the residual trace rate of the surrounding eye, Indicates the adjustment coefficient of surrounding rock strength parameters to rock wall flatness weight, Indicates the adjustment coefficient of the rock mass structure surface spacing on the excess consumption weight of shotcrete. represents the bedding plane dip, represents the maximum bedding plane dip, represents the rock mass integrity parameter, represents the maximum rock mass integrity parameter, Indicates the degree of joint development, Indicates the maximum degree of joint development, represents the surrounding rock strength parameter, represents the maximum surrounding rock strength parameter, Indicates the spacing between rock mass structural planes, Indicates the maximum spacing between rock mass structural planes.

2. The intelligent optimization method for smooth blasting holes in layered rock mass according to claim 1 is characterized in that: In step S1, the geological survey equipment includes geological radar, three-dimensional laser scanner, sonic detector and drilling coring equipment.

3. The intelligent optimization method for smooth blasting holes in layered rock mass tunnels according to claim 1 is characterized in that: In step S1, the geological data includes bedding plane strike, dip, rock mass integrity parameters, joint development degree, and surrounding rock strength parameters.

4. The intelligent optimization method for smooth blasting holes in layered rock mass according to claim 1 is characterized in that: In step S2, the data evaluation model is established based on the support vector machine, and the specific process is as follows: Collect historical geological data and corresponding blasting effect data, use geological data as input feature vectors and blasting effect data as output labels, and standardize the data to make its mean 0 and standard deviation 1; The processed data is divided into a training set and a test set in a ratio of 7:

3. The radial basis function is used as the kernel function of the support vector machine. The support vector machine model is trained using the training set, and the performance of the model is optimized by adjusting the penalty parameters and kernel function parameters. Use the test set to evaluate the trained support vector machine model, calculate the model's accuracy, recall rate, and F1 value, and adjust and optimize the model parameters based on the evaluation results; After the newly acquired geological data are standardized, they are input into the trained support vector machine model. The model outputs the quality assessment results of the data. Based on the assessment results, the new data are preprocessed to extract features related to the bedding plane strike, dip, rock mass integrity parameters, joint development degree and surrounding rock strength parameters in the geological data.

5. The intelligent optimization method for smooth blasting holes in layered rock mass tunnels according to claim 1 is characterized in that: In step S4, the intelligent optimization algorithm combines the genetic algorithm and the particle swarm algorithm to explore the optimal solution that meets the constraints in the solution space by simulating the biological evolution process and swarm intelligence behavior. The specific implementation steps of the intelligent optimization algorithm are as follows: Step S41: Initialize the population, randomly generate a combination of blasthole parameters as the initial solution, and encode each solution. Use real number encoding to convert the blasthole position, spacing, angle and depth into a real number vector; Step S42: Calculate the objective function value corresponding to each solution. Substitute the decoded blasthole parameters into the constructed multi-objective optimization function to obtain the function value of each blasting effect index and the comprehensive objective function value as the basis for evaluating the quality of the solution. Step S43: selection operation, adopting the tournament selection strategy to select parent individuals from the current population and enter the crossover and mutation operation phase; Step S44: crossover operation, combining the simulated binary crossover operator of the genetic algorithm and the position update strategy of the particle swarm algorithm to perform crossover combination on the parent individuals to generate new offspring individuals; Step S45: mutation operation, using a polynomial mutation operator to perform local search on offspring individuals, perturb the blasthole parameters, increase the diversity of solutions, prevent the algorithm from falling into local optimality, and control the length of the mutation step by setting the mutation distribution index; Step S46: Update the population, merge the parent individuals and the offspring individuals, sort them according to the objective function value, and generate a new population as the basis for the next iteration. At the same time, check the constraints of the solutions in the new population, eliminate solutions that do not meet the engineering requirements, and use a repair strategy to process the out-of-bounds variables to make them meet the constraints. Step S47, repeating steps S42 to S46 until the preset iteration termination condition is reached, and outputting the optimal blasthole parameter combination and the corresponding objective function value and each blasting effect index value.

6. The intelligent optimization method for smooth blasting holes in layered rock mass tunnels according to claim 5, characterized in that: Termination conditions include: reaching the maximum number of iterations, the convergence accuracy of the objective function value, or the blasting effect meeting the engineering requirements.

7. The intelligent optimization method for smooth blasting holes in layered rock mass tunnels according to claim 5, characterized in that: In step S4, the constraints include the reasonable range constraint of the blasthole spacing, the allowable deviation constraint of the blasthole angle, and the blasting safety distance constraint; The reasonable range constraints of blasthole spacing are as follows: ； The allowable deviation constraints of the blasthole angle are as follows: ； The blasting safety distance constraints are as follows: ； ； in, Indicates the The actual blasthole spacing of the blasthole type, Indicates the The minimum permissible blasthole spacing for similar blastholes, Indicates the The maximum permissible blasthole spacing for similar blastholes, , Indicates the groove eye. Indicates auxiliary eye, Indicates peripheral eyes, Indicates the design angle of the blasthole. Indicates the actual angle, Indicates the allowable deviation, Indicates the distance between the blasting point and the protected target. Indicates safe distance. represents the coefficient related to geological conditions and blasting type, Indicates the maximum dosage of a single segment.

8. An intelligent optimization system for smooth blasting holes in tunnels in layered rock mass, characterized by: The method for executing the intelligent optimization method for smooth blasting holes in a tunnel in a layered rock mass according to any one of claims 1 to 7 comprises: The data acquisition module is used to obtain geological data of the tunnel face in layered rock mass and the surrounding rock in front of it using a variety of geological survey equipment; The data processing module is used to perform quality assessment and feature extraction on the acquired geological data. It determines the reliability and applicability of the data based on a preset data evaluation model, removes abnormal data and noise interference, and extracts geological features related to tunnel smooth blasting. The optimization module is used to construct a multi-objective optimization function based on the evaluated geological data and extracted geological features, and use an intelligent optimization algorithm to solve the constructed multi-objective optimization function to search for the optimal combination of blasthole parameters; The construction feedback module is used to apply the solved optimal blasthole parameters to the actual tunnel smooth blasting construction. At the same time, the actual blasting effect is monitored and evaluated after the blasting. The evaluation results are fed back to the multi-objective optimization function construction link, the geological data is updated, and the weight coefficients are adjusted to achieve dynamic optimization of subsequent blasthole optimization plans.

Images