Tunnel blasting vibration reduction method and system

Abstract

The invention provides a tunnel blasting vibration reduction method and system, and relates to the technical field of tunnel blasting. A three-dimensional model is constructed according to historical blasting operation, and a propagation path is judged according to the position relation between a vibration pickup and a tunnel face; according to violent change points of surrounding rock data in the propagation path, geological discontinuity points with relatively uneven surrounding rock geology are judged, the distance and the surrounding rock data are combined to jointly serve as a propagation feature set, and the propagation feature set and an explosion feature set formed by explosive data are trained in a full-connection neural network model to obtain a vibration prediction model; the method comprises the following steps: constructing an explosion characteristic set and a propagation characteristic set according to a blasting plan and actual tunnel surrounding rock data, and evaluating the vibration, generated by each plan, of each part of a tunnel so as to select a suitable plan. And possible vibration at each position of the tunnel can be accurately judged.

Description

Technical Field

[0001] This invention relates to the field of tunnel blasting technology, specifically to a method and system for reducing vibration during tunnel blasting. Background Technology

[0002] Blasting at the tunnel face is a common method in tunnel excavation. However, the vibrations generated by blasting can affect tunnel safety. Therefore, strict control of blasting operations is necessary to minimize the harm caused by vibrations. This requires a thorough assessment of the vibrations before blasting. Vibrations propagate through the surrounding rock, and the vibration at a particular location is directly related not only to the distance from the blasting point but also to geological changes during the propagation process. In the prior art, CN117610315A discloses a tunnel intelligent blasting design system based on multi-source geological information, including intelligent identification of multi-source geological information of the tunnel face and refined blasting zoning; based on the blasting zoning of the tunnel face, the optimal cycle advance, the explosive consumption per unit area of ​​each area, and the layout of cut holes are accurately determined, and then the blasting parameters of the peripheral holes, cut holes and other holes are designed in sequence; according to the maximum blasting vibration velocity allowed by the surrounding environment, the maximum single-stage detonation charge, detonation sequence and delay time difference are determined in sequence, and then the detonation network design is completed; through post-blast information feedback, the intelligent optimization of blasting design parameters is realized.

[0003] While the publicly available technical documents have achieved optimization and monitoring of the blasting entity, they have not considered the impact of blasting on the surrounding rock of the tunnel, nor have they considered the changes in blasting vibration caused by the propagation process, and therefore cannot accurately determine the vibration generated by a particular blast at various points in the surrounding rock.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for reducing vibration during tunnel blasting, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method and system for vibration reduction in tunnel blasting, comprising the following steps: Step 1: Obtain the surrounding rock data, explosive data and vibration data for each blasting operation based on historical blasting operations. The surrounding rock data is obtained through sampling points. Establish a three-dimensional model of the tunnel and working face for each blasting operation, and mark the sampling points and corresponding surrounding rock data in the three-dimensional model. Step 2: Organize the explosive data into an explosion feature set. The vibration data is obtained through vibration pickups placed in the tunnel. The vibration data and the corresponding positions of the vibration pickups are marked in a three-dimensional model. Step 3: Based on the position of each vibration pickup relative to the working face, obtain the surrounding rock data within a certain range between the two, and determine the geological discontinuities in the vibration propagation path based on the changes in the surrounding rock data within this range. At the same time, obtain the mean value of the surrounding rock data within this range and the distance from the vibration pickup to the working face to form a propagation feature set. Step 4: Using the vibration data of each vibration pickup as the output, and the explosion feature set and propagation feature set corresponding to each vibration pickup as the input, train in a fully connected neural network model to obtain a vibration prediction model. Then, based on the actual application scenario, establish the propagation feature set and explosion feature set as inputs to the vibration prediction model to obtain vibration data of various locations in the tunnel corresponding to multiple contingency plans.

[0007] Furthermore, based on historical blasting operations, the surrounding rock data, explosive data, and vibration data corresponding to each blast are obtained. The surrounding rock data is obtained by drilling and sampling in the surrounding rock, and the surrounding rock data includes rock hardness and rock content. A three-dimensional model of the tunnel and the working face is established, and an orthogonal coordinate system is established. The working face is set to be located in the YOZ plane, and the positive direction of the X-axis is perpendicular to the working face and towards the completed tunnel. The unit length is the same as the actual unit length. The sampling points for obtaining the surrounding rock data and the surrounding rock data of each sampling point are marked in the three-dimensional model.

[0008] Furthermore, the explosive data includes explosive equivalent per unit area, explosion area, and average explosion depth. The explosion area is the area of ​​the region where explosives are buried in the working face, and the average explosion depth is the average depth of all the pits where explosives are buried. The explosive equivalent, explosion area, and average explosion depth are summarized to form an explosion feature set.

[0009] Furthermore, the vibration data is obtained by vibration pickups buried in the surrounding rock of the tunnel, and the position of each vibration pickup is marked in a three-dimensional model. The vibration data represents the maximum amplitude detected at the position of the vibration pickup during blasting.

[0010] Furthermore, a perpendicular line is drawn from each vibration pickup point to the working face, and the length of the perpendicular line is recorded to represent the distance from the vibration pickup to the working face. A radius is defined with this perpendicular line as the axis. A cylindrical space was constructed, and the surrounding rock data of all sampling points within this space were identified. Starting from the working face and moving towards the positive X-axis, the sampling points within the cylindrical space were numbered from smallest to largest. The rock hardness change rate and rock content change rate of adjacent sampling points were calculated using the following formulas: in, Indicates the first The rate of change in rock hardness at each sampling point Indicates the first The rate of change of rock content at each sampling point and They represent the first The sampling point and the first Rock hardness at each sampling point and They represent the first The sampling point and the first Rock content at each sampling point Represents the retrieval variables for the sampling points. , , This represents the total number of sampling points within the cylindrical space.

[0011] Furthermore, a change rate threshold is set. When the change rate of rock hardness or rock content at a certain sampling point exceeds the change rate threshold, this point is marked as a geological discontinuity. Simultaneously, the average rock hardness and average rock content of all sampling points within the cylindrical space corresponding to each pickup are obtained. Combined with the number of geological discontinuities and the distance from the pickup to the working face, a propagation feature set is formed, based on the following formula: in, Represents the propagation feature set, This represents the average rock hardness. This represents the average rock content. Indicates the number of discontinuities. Indicates distance; Obtain the propagation feature set corresponding to each vibration pickup.

[0012] Furthermore, the explosion feature set generated by each blasting operation is mapped to the propagation feature set of each vibration pickup. The amplitude of each vibration pickup in the blasting operation is set as a label. The explosion feature set and the propagation feature set are input into a fully connected neural network model for training to obtain a vibration prediction model, and the output is the amplitude.

[0013] Furthermore, before the actual tunnel blasting operation, the three-dimensional model of the tunnel and the surrounding rock data of each sampling point are obtained in accordance with step 1. At the same time, a propagation feature set is formed for each location based on the distance from each location in the tunnel to the tunnel face. At this time, there is more than one propagation feature set. The distance from the target location of the vibration data to be obtained to the tunnel face is the distance in the propagation feature set. Multiple explosion feature sets corresponding to multiple blasting plans are obtained. The propagation feature set and the explosion feature set are input into the vibration prediction model to obtain the amplitude of each location in the tunnel corresponding to each blasting plan.

[0014] Furthermore, the construction workers selected the corresponding blasting scheme based on the amplitude at various locations in the tunnel.

[0015] Meanwhile, this application discloses a tunnel blasting vibration reduction system for implementing the aforementioned tunnel blasting vibration reduction method, comprising: 3D model building module: Based on historical blasting operations, obtain surrounding rock data, explosive data and vibration data for each blasting operation. The surrounding rock data is obtained through sampling points. Build a 3D model of the tunnel and working face for each blasting operation, and mark the sampling points and corresponding surrounding rock data in the 3D model. Data correspondence module: Organizes explosive data into an explosion feature set. The vibration data is acquired by vibration pickups arranged in the tunnel. The vibration data and the corresponding positions of the vibration pickups are marked in a three-dimensional model. Vibration propagation analysis module: Based on the position of each vibration pickup relative to the working face, it acquires surrounding rock data within a certain range between the two, and determines the geological discontinuities in the vibration propagation path based on the changes in the surrounding rock data within this range. At the same time, it acquires the mean value of the surrounding rock data within this range and the distance from the vibration pickup to the working face to form a propagation feature set. Prediction model building module: The vibration data of each vibration pickup is used as the output, and the explosion feature set and the corresponding propagation feature set of each vibration pickup are used as the input. The model is trained in a fully connected neural network model to obtain a vibration prediction model. The propagation feature set and explosion feature set are established according to the actual application scenario and input into the vibration prediction model to obtain vibration data of various locations in the tunnel corresponding to various contingency plans.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention constructs a three-dimensional model based on historical blasting operations and marks the sampling points for acquiring surrounding rock data and the positions of vibration pickups for acquiring vibration data in the three-dimensional model. The propagation path is determined by the positional relationship between the vibration pickups and the tunnel face. Geological discontinuities with relatively uneven geological uniformity are identified based on the points of drastic change in surrounding rock data along the propagation path. Geological discontinuities, distances, and surrounding rock data are used together as a propagation feature set. Combined with the explosion feature set formed by explosive data, a vibration prediction model is trained in a fully connected neural network model. An explosion feature set and a propagation feature set for actual tunnel applications are constructed using blasting plans and actual tunnel surrounding rock data. The vibrations generated at various points in the tunnel by each plan are evaluated to select the appropriate plan. This invention considers the vibration propagation process and the influence of surrounding rock changes on vibration during the vibration evaluation process, accurately predicting the possible vibrations at various points in the tunnel. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall method flow of the present invention; Figure 2 This is a schematic diagram of the overall system structure of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0019] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0020] Example: Please see Figure 1 The present invention provides a technical solution: A method and system for vibration reduction in tunnel blasting, comprising the following steps: Step 1: Obtain the surrounding rock data, explosive data and vibration data for each blasting operation based on historical blasting operations. The surrounding rock data is obtained through sampling points. Establish a three-dimensional model of the tunnel and working face for each blasting operation, and mark the sampling points and corresponding surrounding rock data in the three-dimensional model. Step 1 includes the following: Based on historical blasting operations, obtain the surrounding rock data, explosive data and vibration data corresponding to each blast. The surrounding rock data is obtained by drilling and sampling in the surrounding rock, and the surrounding rock data includes rock hardness and rock content. This step provides geological foundational information on historical blasting, ensuring that surrounding rock conditions are accurately recorded and incorporated into subsequent analyses in a traceable manner. Surrounding rock data obtained through borehole sampling, particularly rock hardness and content, provides crucial input for building 3D models of the tunnel and face, enabling the models to accurately reflect the distribution and variability of the surrounding rock's mechanical properties. This not only improves the spatial accuracy of the 3D model but also provides a reliable baseline for future comparative analysis and parameter sensitivity studies of different construction sections, enhancing the predictability and comparability of vibration propagation behavior under different blasting conditions. Furthermore, this sub-step provides traceability of data sources for subsequent steps in data annotation, feature set construction, and model training, ensuring that the reasoning throughout the entire process chain is based on authentic historical observations, thus improving the stability and reliability of the entire system in practical applications.

[0021] The propagation of vibration is affected by the medium. Therefore, when considering the specific situation of vibration at a certain point in the tunnel, it is necessary to consider not only the vibration generated by the blasting operation itself, but also the changes or decay of the vibration during the propagation process.

[0022] A three-dimensional model of the tunnel and the working face is established, and an orthogonal coordinate system is established. The working face is set to be located in the YOZ plane, and the positive direction of the X-axis is perpendicular to the working face and towards the completed tunnel. The unit length is the same as the actual unit length. The sampling points for obtaining the surrounding rock data and the surrounding rock data of each sampling point are marked in the three-dimensional model.

[0023] As a preferred embodiment, the three-dimensional model only needs to construct the theoretical outline of the tunnel surrounding rock surface and the tunnel face surface, without needing to establish detailed surface features and solid material features of the surrounding rock and the tunnel face surface.

[0024] This step provides a unified spatial reference system consistent with actual construction geometry, ensuring physical meaning and consistency in the labeling, coordinate positioning, and distance calculation of all sampling points and surrounding rock data in the 3D model. By defining the tunnel face as being in the YOZ plane, with the positive X-axis perpendicular to the face and pointing towards the completed tunnel, subsequent analyses of the blasting area, vibration propagation path, and geological variations can be performed directly based on realistic geometric relationships, avoiding errors caused by geometric mismatches. Consistency between unit length and actual length ensures consistency in distance conversion and triggering conditions across different steps, avoiding prediction deviations due to scale mismatches. This spatial framework provides a crucial structural foundation for the propagation path analysis in step 3, the vibration prediction model input feature design in step 4, and the integration with the actual blasting scheme, ensuring smooth data transitions and repeatability across different stages.

[0025] Step 2: Organize the explosive data into an explosion feature set. The vibration data is obtained through vibration pickups placed in the tunnel. The vibration data and the corresponding positions of the vibration pickups are marked in a three-dimensional model. Step 2 includes the following: The explosive data includes explosive equivalent per unit area, explosion area, and average explosion depth. The explosion area is the area of ​​the region where explosives are buried in the working face, and the average explosion depth is the average depth of all the pits where explosives are buried. The explosive equivalent, explosion area, and average explosion depth are summarized to form an explosion feature set.

[0026] By structuring and standardizing the key characteristics of the blast source, a set of blast features that are easy to compare and analyze is formed. By transforming elements such as explosive equivalent per unit area, blast area, and average blast depth into unified features that can be input into the model, fair comparisons can be made between different blasting schemes, revealing the relative contributions of various blasting parameters to the vibration response. This provides a clear input for subsequent integration with vibration observation data, helps establish a quantifiable description of the relationship between the blast source and the geological propagation path, and provides data support for optimizing blasting designs through multi-scheme comparison in actual construction, making the decision-making process more scientific, traceable, and repeatable.

[0027] The vibration data is obtained by vibration pickups buried in the surrounding rock of the tunnel. The position of each vibration pickup is marked in a three-dimensional model. The vibration data represents the maximum amplitude detected at the position of the vibration pickup during blasting.

[0028] By combining vibration observation data with specific spatial location information, a spatially located vibration dataset is formed. By deploying vibration pickups in the tunnel surrounding rock and recording the maximum amplitude during blasting, and simultaneously marking the pickup locations in a three-dimensional model, precise spatial mapping of the observation results can be achieved, facilitating the analysis of vibration propagation characteristics under tunnel geometry and geological conditions. This approach enhances data traceability and improves the model's capabilities across different locations.

[0029] Step 3: Based on the position of each vibration pickup relative to the working face, obtain the surrounding rock data within a certain range between the two, and determine the geological discontinuities in the vibration propagation path based on the changes in the surrounding rock data within this range. At the same time, obtain the mean value of the surrounding rock data within this range and the distance from the vibration pickup to the working face to form a propagation feature set. Step 3 includes the following: Step 301: Draw a perpendicular line from each vibration pickup to the working face, and record the length of the perpendicular line to represent the distance from the vibration pickup to the working face. Define a radius using this perpendicular line as the axis. A cylindrical space was constructed, and the surrounding rock data of all sampling points within this space were identified. Starting from the working face and moving towards the positive X-axis, the sampling points within the cylindrical space were numbered from smallest to largest. The rock hardness change rate and rock content change rate of adjacent sampling points were calculated using the following formulas: in, Indicates the first The rate of change in rock hardness at each sampling point Indicates the first The rate of change of rock content at each sampling point and They represent the first The sampling point and the first Rock hardness at each sampling point and They represent the first The sampling point and the first Rock content at each sampling point Represents the retrieval variables for the sampling points. , , This represents the total number of sampling points within the cylindrical space.

[0030] The dependent variable is the rate of change of rock hardness. and rock content change rate This directly reflects the local gradient or relative variation of the surrounding rock in the spatial direction from the actual construction environment. In other words, if the hardness or composition of a certain segment of the surrounding rock changes significantly from the previous sampling point to the current sampling point, it will... or The large absolute values ​​indicate potential geological discontinuities or heterogeneity, which has direct practical significance for subsequently determining the vibration propagation path and characteristics. The independent variables include the original observations of two adjacent points, where... and Indicates sampling point Compared to the rock hardness at the previous sampling point, and Indicates sampling point Rock content compared to the previous sampling point; dependent variable and By substituting these two sets of original observations into the formula and The calculations show that their magnitude and sign directly reflect the relative spatial variation of the hardness and content of the surrounding rock. The greater the variation, the stronger the gradient and the more likely there is a significant discontinuity in the geological conditions, which in turn has a more significant impact on the propagation of vibration in the tunnel.

[0031] By establishing a cylindrical space centered on the vibratory pickup and starting from the tunnel face, the surrounding rock sampling points within the cylinder are analyzed in zones, revealing the variation trends of geological conditions in distance and direction. By sorting the sampling points within the cylinder according to their origin from the tunnel face along the positive X-axis, the rate of change in rock hardness and rock content of adjacent sampling points is calculated, further quantifying the spatial gradient information of the surrounding rock. This information not only provides direct numerical indicators for identifying geological discontinuities but also lays the data foundation for marking geological discontinuities in step 3. Ultimately, this step-by-step output provides a richer geological background description for the propagation path analysis, improves the accuracy and interpretability of subsequent propagation feature sets, and provides clear geological-geometric evidence for the correlation with the pickup location, enhancing the stability and repeatability of subsequent steps.

[0032] In past blasting operations, multiple vibration pickups were often installed, and each pickup would measure different values. This was not only because the pickups were located at different blasting positions, but also because the vibrations detected by different pickups underwent different propagation processes. The inhomogeneity of the medium during the propagation process is the main reason for the changes in vibration. Therefore, it is necessary to take into account the inhomogeneity of the medium during the propagation process of the vibration detected by each pickup.

[0033] Step 302: Set a change rate threshold. When the change rate of rock hardness or rock content at a certain sampling point exceeds the change rate threshold, mark this point as a geological discontinuity. Simultaneously, the average rock hardness and average rock content of all sampling points within the cylindrical space corresponding to each pickup are obtained. Combined with the number of geological discontinuities and the distance from the pickup to the working face, a propagation feature set is formed, based on the following formula: in, Represents the propagation feature set, This represents the average rock hardness. This represents the average rock content. Indicates the number of discontinuities. Indicates distance; Obtain the propagation feature set corresponding to each vibration pickup.

[0034] By setting a threshold for the rate of change, outliers in the rate of change of rock hardness or rock content are explicitly marked as geological discontinuities, thereby quickly and stably identifying geological discontinuities, reducing the risk of misjudgment, and improving the reliability of propagation path analysis. Simultaneously, for each vibration pickup's sampling points within the cylindrical space, the mean rock hardness and mean rock content are calculated, and combined with the number of geological discontinuities and the distance from the pickup to the working face, a propagation feature set is formed. This approach, which integrates statistical features (mean, number of discontinuities, distance) with geological variation features, provides comprehensive input for modeling the geological background of vibration propagation, enhances the robustness and interpretability of the prediction model for vibration behavior under different locations and geological conditions, and provides richer, more structured input data for subsequent vibration prediction model training in step 4.

[0035] When vibrations propagate in a non-uniform medium, points of more drastic change in the medium often mean that the vibration is more likely to undergo drastic changes at those points. Therefore, it is necessary to consider the number of drastic changes that the vibrations experience during propagation, i.e., the number of geological discontinuities, so that this feature can be identified in subsequent model training and taken into account when predicting vibrations.

[0036] Step 4: Using the vibration data of each vibration pickup as the output, and the explosion feature set and propagation feature set corresponding to each vibration pickup as the input, train in a fully connected neural network model to obtain a vibration prediction model. Then, based on the actual application scenario, establish the propagation feature set and explosion feature set as inputs to the vibration prediction model to obtain vibration data of various locations in the tunnel corresponding to multiple contingency plans.

[0037] Step 4 includes the following: Step 401: Match the explosion feature set generated by each blasting operation with the propagation feature set of each vibration pickup, set the amplitude of each vibration pickup in the blasting operation as a label, input the explosion feature set and propagation feature set into the fully connected neural network model for training, obtain the vibration prediction model, and output the amplitude.

[0038] By mapping explosion feature sets to propagation feature sets and labeling the amplitude of each vibration pickup during a blasting operation, a fully connected neural network is used to learn the relationship between explosion source features and propagation path features, constructing a vibration prediction model and outputting the amplitude. This modeling approach achieves a direct mapping between blast source information and geological propagation information, enabling rapid prediction of vibration responses at different locations under various blasting schemes. This provides quantitative and comparable vibration prediction results for engineering decisions. This step integrates historical data, geological features, and observed vibrations into a unified learning framework, promoting data-driven design optimization, improving the ability to assess vibration risks for different blasting plans, and providing repeatable and verifiable prediction tools for on-site construction.

[0039] In each blasting operation, there are multiple vibration pickups, each corresponding to a propagation feature set and a vibration data label. Therefore, each explosion feature set will generate multiple propagation feature sets and corresponding labels. Combining multiple blasting operations will generate a large number of explosion feature sets, propagation feature sets and labels, which will facilitate the training of vibration prediction models.

[0040] In a preferred embodiment, the explosion feature set, propagation feature set, and vibration data are summarized, and training samples of the corresponding explosion feature set-propagation feature set-vibration data are formed according to the generation relationship. All training samples are summarized, and the training set and validation set are divided in an 8:2 ratio for training in a fully connected neural network model. The Adam optimizer is used, the learning rate is set to 0.001, the number of hidden layers is 3, and the number of units is 128.

[0041] Step 402: Before the actual tunnel blasting operation, obtain the three-dimensional model of the tunnel and the surrounding rock data of each sampling point in the manner of Step 1. At the same time, form a propagation feature set for each location based on the distance from each location in the tunnel to the tunnel face. At this time, there is more than one propagation feature set. The distance from the target location of the vibration data to be obtained to the tunnel face is the distance in the propagation feature set. Obtain multiple explosion feature sets corresponding to multiple blasting plans. Input the propagation feature set and the explosion feature set into the vibration prediction model to obtain the amplitude of each location in the tunnel corresponding to each blasting plan. Construction workers select appropriate blasting schemes based on the amplitude of vibrations at various locations within the tunnel.

[0042] In a preferred embodiment, construction personnel form multiple explosion feature sets based on multiple blasting plans. They mark the sampling points of the tunnel and the surrounding rock data obtained at each sampling point in the three-dimensional model of the tunnel to be blasted. They select the location to be evaluated for vibration or all locations in the tunnel, using these locations to represent the positions of the vibration pickups. Following step 3, they obtain the propagation feature set for each location. A prediction is made using a single blasting plan and a single location. For example, if the prediction is for the vibration of point a in plan A, an explosion feature set is constructed using plan A, and a propagation feature set is constructed using point a. These propagation and explosion feature sets are then input into the vibration prediction model to obtain the amplitude of point a in plan A. The amplitude of point b in plan A is obtained in the same way, thus obtaining the amplitude generated by plan A for each target point in the tunnel. This is then changed to plan B, and the amplitude of plan B for each target point in the tunnel is obtained in the same way. Finally, the amplitude of all plans for all target points in the tunnel is obtained, and the construction personnel select the required plan based on the amplitude.

[0043] Prior to actual tunnel blasting operations, 3D models and surrounding rock data are continuously acquired, and multiple sets of propagation feature sets are formed based on the distance from the target location to the tunnel face. Simultaneously, multiple sets of explosion feature sets are compiled. These feature sets are input into a vibration prediction model to obtain amplitude prediction results for multiple blasting schemes at various locations within the tunnel. This design allows for parallel comparison and evaluation of vibration responses under different locations and blasting schemes within the same tunnel. This helps construction personnel quickly select the blasting scheme with the minimum amplitude and risk in a given scenario, improving project safety and efficiency. By providing comparative predictions of multiple schemes, the scientific rigor and flexibility of decision-making are enhanced, and the adaptability and scalability of the entire system to changes in the actual engineering environment are improved.

[0044] Meanwhile, this application discloses a tunnel blasting vibration reduction system for implementing the aforementioned tunnel blasting vibration reduction method, comprising: 3D model building module: Based on historical blasting operations, obtain surrounding rock data, explosive data and vibration data for each blasting operation. The surrounding rock data is obtained through sampling points. Build a 3D model of the tunnel and working face for each blasting operation, and mark the sampling points and corresponding surrounding rock data in the 3D model. Data correspondence module: Organizes explosive data into an explosion feature set. The vibration data is acquired by vibration pickups arranged in the tunnel. The vibration data and the corresponding positions of the vibration pickups are marked in a three-dimensional model. Vibration propagation analysis module: Based on the position of each vibration pickup relative to the working face, it acquires surrounding rock data within a certain range between the two, and determines the geological discontinuities in the vibration propagation path based on the changes in the surrounding rock data within this range. At the same time, it acquires the mean value of the surrounding rock data within this range and the distance from the vibration pickup to the working face to form a propagation feature set. Prediction model building module: The vibration data of each vibration pickup is used as the output, and the explosion feature set and the corresponding propagation feature set of each vibration pickup are used as the input. The model is trained in a fully connected neural network model to obtain a vibration prediction model. The propagation feature set and explosion feature set are established according to the actual application scenario and input into the vibration prediction model to obtain vibration data of various locations in the tunnel corresponding to various contingency plans.

[0045] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0046] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0047] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0048] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for reducing vibration during tunnel blasting, characterized in that, The specific steps include: Step 1: Obtain the surrounding rock data, explosive data and vibration data for each blasting operation based on historical blasting operations. The surrounding rock data is obtained through sampling points. Establish a three-dimensional model of the tunnel and working face for each blasting operation, and mark the sampling points and corresponding surrounding rock data in the three-dimensional model. Step 2: Organize the explosive data into an explosion feature set. The vibration data is obtained through vibration pickups placed in the tunnel. The vibration data and the corresponding positions of the vibration pickups are marked in a three-dimensional model. Step 3: Based on the position of each vibration pickup relative to the working face, obtain the surrounding rock data within a certain range between the two, and determine the geological discontinuities in the vibration propagation path based on the changes in the surrounding rock data within this range. At the same time, obtain the mean value of the surrounding rock data within this range and the distance from the vibration pickup to the working face to form a propagation feature set. Step 4: Using the vibration data of each vibration pickup as the output, and the explosion feature set and propagation feature set corresponding to each vibration pickup as the input, train in a fully connected neural network model to obtain a vibration prediction model. Then, based on the actual application scenario, establish the propagation feature set and explosion feature set as inputs to the vibration prediction model to obtain vibration data of various locations in the tunnel corresponding to multiple contingency plans.

2. The tunnel blasting vibration reduction method according to claim 1, characterized in that: Based on historical blasting operations, obtain the surrounding rock data, explosive data and vibration data corresponding to each blast. The surrounding rock data is obtained by drilling and sampling in the surrounding rock, and the surrounding rock data includes rock hardness and rock content. A three-dimensional model of the tunnel and the working face is established, and an orthogonal coordinate system is established. The working face is set to be located in the YOZ plane, and the positive direction of the X-axis is perpendicular to the working face and towards the completed tunnel. The unit length is the same as the actual unit length. The sampling points for obtaining the surrounding rock data and the surrounding rock data of each sampling point are marked in the three-dimensional model.

3. The tunnel blasting vibration reduction method according to claim 2, characterized in that: The explosive data includes explosive equivalent per unit area, explosion area, and average explosion depth. The explosion area is the area of ​​the region where explosives are buried in the working face, and the average explosion depth is the average depth of all the pits where explosives are buried. The explosive equivalent, explosion area, and average explosion depth are summarized to form an explosion feature set.

4. The tunnel blasting vibration reduction method according to claim 3, characterized in that: The vibration data is obtained by vibration pickups buried in the surrounding rock of the tunnel. The position of each vibration pickup is marked in a three-dimensional model. The vibration data represents the maximum amplitude detected at the position of the vibration pickup during blasting.

5. A method for reducing vibration during tunnel blasting according to claim 4, characterized in that: Draw a perpendicular line from each vibration pickup to the working face, and record the length of the perpendicular line to represent the distance from the vibration pickup to the working face. Define a radius around this perpendicular line. A cylindrical space was constructed, and the surrounding rock data of all sampling points within this space were identified. Starting from the working face and moving towards the positive X-axis, the sampling points within the cylindrical space were numbered from smallest to largest. The rock hardness change rate and rock content change rate of adjacent sampling points were calculated using the following formulas: in, Indicates the first The rate of change in rock hardness at each sampling point Indicates the first The rate of change of rock content at each sampling point and They represent the first The sampling point and the first Rock hardness at each sampling point and They represent the first The sampling point and the first Rock content at each sampling point Represents the retrieval variables for the sampling points. , , This represents the total number of sampling points within the cylindrical space.

6. The tunnel blasting vibration reduction method according to claim 5, characterized in that: Set a rate of change threshold. When the rate of change of rock hardness or rock content at a certain sampling point exceeds the rate of change threshold, this point is marked as a geological discontinuity. Simultaneously, the average rock hardness and average rock content of all sampling points within the cylindrical space corresponding to each pickup are obtained. Combined with the number of geological discontinuities and the distance from the pickup to the working face, a propagation feature set is formed, based on the following formula: in, Represents the propagation feature set, This represents the average rock hardness. This represents the average rock content. Indicates the number of discontinuities. Indicates distance; Obtain the propagation feature set corresponding to each vibration pickup.

7. A method for reducing vibration during tunnel blasting according to claim 6, characterized in that: The explosion feature set generated by each blasting operation is mapped to the propagation feature set of each vibration pickup. The amplitude of each vibration pickup in the blasting operation is set as a label. The explosion feature set and the propagation feature set are input into a fully connected neural network model for training to obtain a vibration prediction model. The output is the amplitude.

8. The tunnel blasting vibration reduction method according to claim 7, characterized in that: Before the actual tunnel blasting operation, the three-dimensional model of the tunnel and the surrounding rock data of each sampling point are obtained according to step 1. At the same time, a propagation feature set is formed for each location based on the distance from each location in the tunnel to the tunnel face. At this time, there is more than one propagation feature set. The distance from the target location of the vibration data to be obtained to the tunnel face is the distance in the propagation feature set. Multiple explosion feature sets corresponding to multiple blasting plans are obtained. The propagation feature set and the explosion feature set are input into the vibration prediction model to obtain the amplitude of each location in the tunnel corresponding to each blasting plan.

9. A method for reducing vibration during tunnel blasting according to claim 8, characterized in that: Construction workers select appropriate blasting schemes based on the amplitude of vibrations at various locations within the tunnel.

10. A tunnel blasting vibration reduction system, used to perform a tunnel blasting vibration reduction method according to any one of claims 1-9, characterized in that: 3D model building module: Based on historical blasting operations, obtain surrounding rock data, explosive data and vibration data for each blasting operation. The surrounding rock data is obtained through sampling points. Build a 3D model of the tunnel and working face for each blasting operation, and mark the sampling points and corresponding surrounding rock data in the 3D model. Data correspondence module: Organizes explosive data into an explosion feature set. The vibration data is acquired by vibration pickups arranged in the tunnel. The vibration data and the corresponding positions of the vibration pickups are marked in a three-dimensional model. Vibration propagation analysis module: Based on the position of each vibration pickup relative to the working face, it acquires surrounding rock data within a certain range between the two, and determines the geological discontinuities in the vibration propagation path based on the changes in the surrounding rock data within this range. At the same time, it acquires the mean value of the surrounding rock data within this range and the distance from the vibration pickup to the working face to form a propagation feature set. Prediction model building module: The vibration data of each vibration pickup is used as the output, and the explosion feature set and the corresponding propagation feature set of each vibration pickup are used as the input. The model is trained in a fully connected neural network model to obtain a vibration prediction model. The propagation feature set and explosion feature set are established according to the actual application scenario and input into the vibration prediction model to obtain vibration data of various locations in the tunnel corresponding to various contingency plans.

Images