Shallow-buried tunnel blasting ground surface vibration waveform prediction method and system

Abstract

The present application provides a kind of shallow tunnel blasting ground surface vibration waveform prediction method and system, it is related to ground surface vibration waveform prediction technical field, including: collecting geological, blasting and environmental parameters of history and current blasting, by three-stage index storage, outlier screening and standardization processing, form standardization parameter set;Based on the mapping relationship of LSTM neural network between blasting parameters and vibration waveform time domain sequence, predict the vibration waveform of current blasting;Fusion geological condition influence factor, environmental sensitivity influence factor and predicted waveform characteristics, generate comprehensive safety evaluation index;According to the matching result of the index and multilevel safety threshold, the risk is divided into four levels of safety, low risk, medium risk and high risk, and differential management is executed, to realize the accurate prediction and dynamic regulation of blasting vibration risk.The present application can accurately simulate the vibration wave propagation process in shallow tunnel blasting construction, so as to better evaluate the influence of blasting.

Description

Technical Field

[0001] This invention relates to the field of surface vibration waveform prediction technology, specifically to a method and system for predicting surface vibration waveforms during shallow-buried tunnel blasting. Background Technology

[0002] Shallow tunnel blasting is a crucial step in urban underground space development and transportation infrastructure construction. The resulting surface vibrations directly impact the safety of nearby buildings, underground pipelines, and the daily lives of residents. Therefore, accurate prediction, assessment, and control of surface vibrations generated by blasting operations are core technical issues for ensuring project safety and reducing environmental and social risks. Traditional vibration assessment methods often rely on empirical formulas or simplified numerical simulations, which struggle to fully reflect the vibration response characteristics under complex geological conditions, variable blasting parameters, and the coupled effects of sensitive surrounding environments. Prediction accuracy and real-time performance need improvement, resulting in significant shortcomings in refined safety management.

[0003] In the prior art, a blasting operation control system and method based on the Internet of Things (IoT) with publication number CN119539506A integrates IoT sensing and simulation analysis technologies to achieve comprehensive assessment and dynamic control of the blasting vibration propagation range and environmental risks. However, this scheme focuses on macroscopic regional safety distance calculation and risk scoring based on multiple factors (such as geology and climate). Its vibration prediction does not delve into the waveform time-domain sequence level and lacks a fine-grained description of the characteristics of the entire vibration process. At the same time, its consideration of the impact on the surrounding environment does not establish a direct and dynamic mapping relationship with the blast source parameters, nor does it model the vibration superposition effect of multi-stage micro-differential blasting in shallow-buried tunnels. Therefore, there is still room for further improvement in terms of vibration waveform prediction accuracy, environmental sensitivity factor coupling analysis, and the targeted nature of real-time hierarchical control.

[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 predicting surface vibration waveforms during blasting in shallow-buried tunnels, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for predicting surface vibration waveforms during shallow-buried tunnel blasting, comprising the following steps:

[0008] Step 1: Collect the basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting operation; the basic parameters include geological parameters, blasting parameters and surrounding environmental parameters. Based on the parameter type, design a three-level index table to classify and store the basic parameters. Use an outlier detection algorithm to filter the effective parameters in the three-level index table, and standardize the effective parameters to obtain a standardized parameter set;

[0009] Step 2: Analyze the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences, establish the mapping relationship between the blasting parameters and the vibration waveform time-domain sequences, and determine the predicted vibration waveform time-domain sequence corresponding to the current blast based on the mapping relationship;

[0010] Step 3: Based on the geological parameters and surrounding environmental parameters in the standardized parameter set, determine the geological condition influencing factor and the environmental sensitivity influencing factor. Perform a fusion analysis on the geological condition influencing factor, the environmental sensitivity influencing factor, and the predicted vibration waveform time-domain sequence to generate a comprehensive safety assessment index for quantitatively evaluating the safety of the current blasting.

[0011] Step 4: Call the preset multi-level safety risk thresholds, match and analyze the comprehensive safety assessment index with the multi-level safety risk thresholds, and determine the risk level of the current blasting operation based on the matching results; the risk level includes four levels: safe, low risk, medium risk, and high risk;

[0012] Step 5: Implement differentiated control measures based on the determined risk level: for the safety level, implement the original plan and conduct routine monitoring simultaneously; for the low-risk level, activate real-time high-frequency vibration monitoring; for the medium-risk level, reduce the amount of single-stage detonating explosive and extend the detonation time difference proportionally on the basis of low-risk control; for the high-risk level, suspend blasting operations and take enhanced vibration reduction measures, which will be implemented after the risk is reduced.

[0013] Furthermore, the basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting, were collected. The specific data acquisition methods are as follows:

[0014] Through borehole exploration, ground-penetrating radar scanning, and on-site rock mechanics tests, the radius of the explosion source center point of the detonation section was determined. Geological parameters at multiple sampling points within the range, including rock mass elastic modulus, rock mass Poisson's ratio, rock mass density, joint and fissure density, and soil shear wave velocity, are used as the average values ​​as the geological parameter values ​​at the corresponding explosion source center point. Among them, soil shear wave velocity refers to the speed at which shear waves propagate in the soil layer and is used to evaluate the softness and looseness of the soil layer.

[0015] The blasting parameters are obtained from the blasting design scheme and initiation system, including the maximum charge per segment of each initiation section, the segment initiation sequence, the borehole depth, and the spatial coordinates of the corresponding blast source center point; wherein the segment initiation sequence... Represented as , Indicates the first The detonation moment of the explosive at the center point of the blast source. For the index of the detonation phase, , This indicates the number of detonation segments, with each detonation segment corresponding to a blast center point;

[0016] By analyzing the design drawings and conducting on-site inspections, it was determined that the center point of the detonation source in the initiation section should be the center of a circle with a radius of... The surrounding environmental parameters within the range include building burial depth, building equivalent stiffness, underground pipeline diameter, and underground pipeline burial depth. For each explosion source center point, the building burial depth and building equivalent stiffness of the nearest building to the explosion source center point are taken as representative values. The shortest vertical distance from the explosion source center point to each nearby pipeline is calculated, and the underground pipeline diameter and underground pipeline burial depth corresponding to the pipeline with the minimum shortest vertical distance are taken as representative values. If there are multiple pipelines corresponding to the minimum shortest vertical distance, the underground pipeline diameter and underground pipeline burial depth of the pipeline with the most vibration-sensitive material are taken as representative values.

[0017] The three-level index table has a first-level index for parameter categories, namely geological parameters, blasting parameters, and surrounding environment parameters; a second-level index for the specific parameter names under each parameter category; and a third-level index for the collection timestamps and spatial coordinates of each parameter.

[0018] Furthermore, when filtering valid parameters of the three-level index table using the outlier detection algorithm, a dual threshold method based on physical meaning and statistical distribution is adopted: First, a reasonable range of values ​​for each parameter is set based on engineering experience, and parameter values ​​exceeding this reasonable range are identified as outliers and removed; Second, for parameters within the reasonable range, a box plot method is used to identify parameter values ​​less than the lower quartile - 3 times the interquartile range or greater than the upper quartile + 3 times the interquartile range as discrete outliers and remove them.

[0019] The standardization process employs the range standardization method, transforming the selected parameter values ​​to the range using the following formula. Interval:

[0020] ;

[0021] in For the original value of the parameter, For the standardized values ​​of the parameters, , These are the minimum and maximum values ​​of the current parameter to be standardized in the set of valid parameters after outlier filtering;

[0022] After the above processing, a standardized parameter set containing standardized values ​​of geological parameters, blasting parameters, and surrounding environmental parameters is obtained.

[0023] Furthermore, the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences are analyzed to establish a mapping relationship between the blasting parameters and the vibration waveform time-domain sequences. Based on this mapping relationship, the predicted vibration waveform time-domain sequence corresponding to the current blast is determined. The specific logic is as follows:

[0024] A machine learning model is constructed based on long short-term memory network. The standardized parameter set includes the blasting parameters and vibration waveform time-domain sequences of historical shallow tunnel blasting construction. The sets are divided into training set, validation set and test set according to the proportions of 70%, 15% and 15% respectively. The blasting parameters corresponding to each blasting segment are used as the model input, and the vibration waveform time-domain sequences of each blasting segment are used as the model output for model training.

[0025] The model has three LSTM hidden layers with 128, 64, and 32 neurons respectively. The model training uses the mean squared error (MSE) as the loss function and adopts the Adam optimizer. The learning rate is set to 0.001, the first moment estimation exponential decay rate is set to 0.9, and the training is automatically stopped when the validation set loss no longer decreases for 10 consecutive rounds to prevent overfitting.

[0026] After training, the blasting parameters corresponding to each detonation segment of the current blasting operation are input into the model, and the model outputs the waveform time-domain sequence corresponding to each detonation segment. The waveform time-domain sequences of all detonation segments are superimposed according to the detonation time sequence to form a complete predicted vibration waveform time-domain sequence.

[0027] Furthermore, based on the geological parameters and surrounding environmental parameters in the standardized parameter set, the geological condition influencing factors and environmental sensitivity influencing factors are determined;

[0028] The formula used to calculate the geological condition influence factor is as follows:

[0029] ;

[0030] In the formula, For the first Geological conditions at the center of an explosion source are a factor influencing this. For the first in the standardized parameter set Joint and fracture density at the center of the explosion source The wave impedance influence factor is, where For the first in the standardized parameter set The density of rock at the center of the explosion source For the first in the standardized parameter set Shear wave velocity of the soil layer at the center of the explosion source For the first in the standardized parameter set Elastic modulus of rock mass at the center of the explosion source To standardize the parameters, the maximum value of the rock mass elastic modulus is required. For the first in the standardized parameter set Poisson's ratio of the rock mass at the center of the explosion source. , , and For preset weights, , , and All are non-negative numbers, and satisfy the following conditions: ;

[0031] For each detonation stage The surrounding environmental parameters corresponding to the explosion source center point are extracted from the standardized parameter set, including building burial depth, building equivalent stiffness, underground pipeline diameter, and underground pipeline burial depth. Based on the above parameters, the building reflection correction coefficient and the underground pipeline attenuation correction coefficient are calculated to determine the environmental sensitivity influence factor. The calculation formula is as follows:

[0032] ;

[0033] In the formula, For the first The building reflection correction factor at the center point of the explosion source. For the first Attenuation correction factor for underground pipelines at the center point of the explosion source For the first Environmental sensitivity factors affecting the center point of an explosion source For the first in the standardized parameter set The equivalent stiffness of the building at the center point of the explosion source This represents the maximum value of the equivalent stiffness of the building within the standardized parameter set; For the first in the standardized parameter set The diameter of the underground pipeline at the center point of the explosion source, The maximum value of the underground pipeline diameter in the standardized parameter set; For the first in the standardized parameter set The burial depth of underground pipelines at the center point of the explosion source;

[0034] A comprehensive safety assessment index is generated by fusing geological condition influencing factors, environmental sensitivity influencing factors, and predicted vibration waveform time-domain sequences to quantitatively evaluate the safety of the current blasting operation. The formula used is as follows:

[0035] ;

[0036] In the formula, To form a comprehensive safety assessment index, Indicates the number of detonation stages. The peak particle vibration velocity of the predicted vibration waveform time-domain sequence for the i-th initiation segment; The dominant frequency of the predicted vibration waveform time-domain sequence for the i-th detonation segment.

[0037] Furthermore, a preset multi-level safety risk threshold is invoked, and the comprehensive safety assessment index is matched and analyzed with the multi-level safety risk threshold. The risk level of the current blasting operation is determined based on the matching results. The specific logic behind this is as follows:

[0038] like If so, the risk level of the current blasting operation is determined to be "safe";

[0039] like If so, the risk level of the current blasting operation is determined to be "low risk";

[0040] like If so, the risk level of the current blasting operation is determined to be "medium risk";

[0041] like If so, the risk level of the current blasting operation is determined to be "high risk";

[0042] In the formula, , , , These represent the preset upper limit of the safety threshold, the upper limit of the low-risk threshold, the upper limit of the medium-risk threshold, and the lower limit of the high-risk threshold, respectively, and satisfy the following conditions: .

[0043] Furthermore, the specific execution process of step 5 is as follows:

[0044] Based on the determined risk level of the current blasting operation, corresponding graded control and real-time feedback operations are implemented: If the risk level is low, real-time high-frequency vibration monitoring sensors are deployed in areas where the predicted peak particle vibration velocity is greater than 0.5 cm / s and in sensitive structures within 50 m of the blast source center point during blasting, with a sampling frequency of not less than 1000 Hz, and the monitoring data is dynamically fed back to the management platform; if the risk level is medium, based on the low-risk monitoring, the amount of explosive charge per stage is reduced to 70% of the original design value, and the detonation time difference is extended to 1.5 times the original design value; if the risk level is high, the blasting operation is immediately suspended, and enhanced vibration reduction measures are taken, and the risk level is reassessed until it drops to medium or below; at the same time, the control measures and execution results under each risk level are recorded in the database for optimizing the early warning model and threshold settings.

[0045] The present invention also provides a surface vibration waveform prediction system for shallow tunnel blasting, wherein the surface vibration waveform prediction system for shallow tunnel blasting is used to execute the above-mentioned surface vibration waveform prediction method for shallow tunnel blasting, including:

[0046] The data acquisition and preprocessing module is used to collect the basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting. The basic parameters include geological parameters, blasting parameters, and surrounding environmental parameters. A three-level index table is designed based on parameter type to classify and store the basic parameters. The effective parameters in the three-level index table are filtered by an outlier detection algorithm, and the effective parameters are standardized to obtain a standardized parameter set.

[0047] The mapping relationship determination module is used to analyze the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences, establish the mapping relationship between the blasting parameters and the vibration waveform time-domain sequences, and determine the predicted vibration waveform time-domain sequence corresponding to the current blast based on the mapping relationship.

[0048] The safety index fusion calculation module is used to determine the geological condition influence factor and the environmental sensitivity influence factor based on the geological parameters and surrounding environmental parameters in the standardized parameter set. It performs fusion analysis on the geological condition influence factor, the environmental sensitivity influence factor and the predicted vibration waveform time domain sequence to generate a comprehensive safety assessment index for quantitatively evaluating the safety of the current blasting.

[0049] The risk grading module is used to call preset multi-level safety risk thresholds, match and analyze the comprehensive safety assessment index with the multi-level safety risk thresholds, and determine the risk level of the current blasting operation based on the matching results; the risk level includes four levels: safe, low risk, medium risk, and high risk.

[0050] The graded control measures execution module is used to implement differentiated control measures based on the determined risk level: for the safety level, the original plan is implemented and routine monitoring is carried out simultaneously; for the low-risk level, real-time high-frequency vibration monitoring is activated; for the medium-risk level, the amount of single-stage detonating explosive is reduced proportionally and the detonation time difference is extended on the basis of low-risk control; for the high-risk level, blasting operations are suspended and enhanced vibration reduction measures are taken, which will be implemented after the risk is reduced.

[0051] Compared with the prior art, the beneficial effects of the present invention are:

[0052] This invention achieves high-precision prediction of the time-domain sequence of surface vibration waveforms during shallow-buried tunnel blasting by constructing a deep learning model that integrates geological conditions, environmental sensitivity, and predicted vibration waveform characteristics, significantly improving the accuracy and reliability of vibration assessment. A comprehensive safety assessment index with multi-factor coupling quantitatively reflects the complex risk status of blasting operations and achieves refined risk classification by combining multi-level thresholds. A systematic hierarchical control mechanism dynamically adjusts monitoring strategies and blasting parameters based on real-time risk assessment results. It strengthens monitoring of key areas during low-risk periods, automatically optimizes charge quantity and detonation sequence during medium-risk periods, and decisively interrupts operations and initiates enhanced measures during high-risk periods, effectively avoiding the drawbacks of traditional methods such as delayed response and inefficient control. This invention organically combines data-driven prediction, multi-source factor fusion, and intelligent decision-making response, significantly improving the safety, controllability, and environmental friendliness of shallow-buried tunnel blasting operations, providing comprehensive technical support for blasting construction in complex urban environments. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the overall method flow of the present invention;

[0054] Figure 2 This is a schematic diagram of the overall system modules of the present invention;

[0055] Figure 3 This is a parallel coordinate image of joint and fracture density versus geological condition influencing factors;

[0056] Figure 4 This is a parallel coordinate image of the wave impedance influence factor and the geological condition influence factor;

[0057] Figure 5 A 3D bar image showing the influence of joint and fracture density on geological conditions. Detailed Implementation

[0058] 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.

[0059] 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.

[0060] Example:

[0061] Please see Figure 1 The present invention provides a technical solution:

[0062] A method for predicting surface vibration waveforms during shallow-buried tunnel blasting, comprising the following steps:

[0063] Step 1: Collect the basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting operation; the basic parameters include geological parameters, blasting parameters and surrounding environmental parameters. Based on the parameter type, design a three-level index table to classify and store the basic parameters. Use an outlier detection algorithm to filter the effective parameters in the three-level index table, and standardize the effective parameters to obtain a standardized parameter set;

[0064] In this embodiment, basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations are collected, as well as basic parameters of the current shallow-buried tunnel blasting operation. The specific data collection method is as follows:

[0065] Through borehole exploration, ground-penetrating radar scanning, and on-site rock mechanics tests, the radius of the explosion source center point of the detonation section was determined. Geological parameters at multiple sampling points within the range, including rock mass elastic modulus, rock mass Poisson's ratio, rock mass density, joint and fissure density, and soil shear wave velocity, are used as the average values ​​as the geological parameter values ​​at the corresponding explosion source center point. Among them, soil shear wave velocity refers to the speed at which shear waves propagate in the soil layer and is used to evaluate the softness and looseness of the soil layer.

[0066] The blasting parameters are obtained from the blasting design scheme and initiation system, including the maximum charge per segment of each initiation section, the segment initiation sequence, the borehole depth, and the spatial coordinates of the corresponding blast source center point; wherein the segment initiation sequence... Represented as , Indicates the first The detonation moment of the explosive at the center point of the blast source. For the index of the detonation phase, , This indicates the number of detonation segments, with each detonation segment corresponding to a blast center point;

[0067] By analyzing the design drawings and conducting on-site inspections, it was determined that the center point of the detonation source in the initiation section should be the center of a circle with a radius of... The surrounding environmental parameters within the range include building burial depth, building equivalent stiffness, underground pipeline diameter, and underground pipeline burial depth. For each explosion source center point, the building burial depth and building equivalent stiffness of the nearest building to the explosion source center point are taken as representative values. The shortest vertical distance from the explosion source center point to each nearby pipeline is calculated, and the underground pipeline diameter and underground pipeline burial depth corresponding to the pipeline with the minimum shortest vertical distance are taken as representative values. If there are multiple pipelines corresponding to the minimum shortest vertical distance, the underground pipeline diameter and underground pipeline burial depth of the pipeline with the most vibration-sensitive material are taken as representative values.

[0068] The three-level index table has a first-level index for parameter categories, namely geological parameters, blasting parameters, and surrounding environment parameters; a second-level index for the specific parameter names under each parameter category; and a third-level index for the collection timestamps and spatial coordinates of each parameter.

[0069] When filtering valid parameters of the three-level index table using outlier detection algorithms, a dual threshold method based on physical meaning and statistical distribution is adopted: First, a reasonable range of values ​​for each parameter is set based on engineering experience, and parameter values ​​that exceed the reasonable range are identified as outliers and removed; Second, for parameters within the reasonable range, a box plot method is used to identify parameter values ​​that are less than the lower quartile - 3 times the interquartile range or greater than the upper quartile + 3 times the interquartile range as discrete outliers and removed.

[0070] The standardization process employs the range standardization method, transforming the selected parameter values ​​to the range using the following formula. Interval:

[0071] ;

[0072] in For the original value of the parameter, For the standardized values ​​of the parameters, , These are the minimum and maximum values ​​of the current parameter to be standardized in the set of valid parameters after outlier filtering;

[0073] After the above processing, a standardized parameter set containing standardized values ​​of geological parameters, blasting parameters, and surrounding environmental parameters is obtained.

[0074] Step 1, serving as the data foundation and core of the overall solution, is significant because it constructs a high-quality, multi-dimensional, and spatiotemporally correlated standardized parameter set through systematic data acquisition, structured storage, and standardized processing. This step employs a three-level index table to classify and store geological, blasting, and environmental parameters, ensuring orderly management and rapid retrieval of massive heterogeneous data. Combined with a dual outlier detection algorithm based on both physical meaning and statistical distribution, it effectively eliminates invalid data caused by human error, equipment noise, and other factors, improving the reliability and consistency of the dataset. Furthermore, range standardization eliminates differences in the dimensions and numerical ranges of different parameters, providing standardized and balanced input data for subsequent machine learning models, fundamentally guaranteeing the accuracy and stability of vibration prediction model training and evaluation.

[0075] Compared with existing technologies, the advantages of this step are mainly reflected in three aspects: First, at the data acquisition level, the principle of sampling within a spatial range centered on the explosion source center point is clearly defined, and the average value of geological parameters and the most recent or most sensitive representative value of environmental parameters are taken, which enhances the spatial representativeness and engineering relevance of the parameters and avoids the problems of arbitrary environmental parameter values ​​and lack of spatial correlation in traditional methods. Second, at the data governance level, a three-level index structure and a dual outlier detection mechanism are innovatively adopted to achieve systematic and standardized management of multi-source heterogeneous data, solving the defects of scattered data storage and insufficient quality control in existing technologies. Third, at the data processing level, through a strict standardized process, high-quality input that can be directly calculated is provided for subsequent waveform prediction based on deep learning, overcoming the drawbacks of traditional methods that limit the accuracy of model prediction due to inconsistent data formats and a lot of noise.

[0076] Step 2: Analyze the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences, establish the mapping relationship between the blasting parameters and the vibration waveform time-domain sequences, and determine the predicted vibration waveform time-domain sequence corresponding to the current blast based on the mapping relationship;

[0077] In this embodiment, the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences are analyzed to establish a mapping relationship between the blasting parameters and the vibration waveform time-domain sequences. Based on this mapping relationship, the predicted vibration waveform time-domain sequence corresponding to the current blast is determined. The specific logic is as follows:

[0078] A machine learning model is constructed based on long short-term memory network. The standardized parameter set includes the blasting parameters and vibration waveform time-domain sequences of historical shallow tunnel blasting construction. The sets are divided into training set, validation set and test set according to the proportions of 70%, 15% and 15% respectively. The blasting parameters corresponding to each blasting segment are used as the model input, and the vibration waveform time-domain sequences of each blasting segment are used as the model output for model training.

[0079] The model has three LSTM hidden layers with 128, 64, and 32 neurons respectively. The model training uses the mean squared error (MSE) as the loss function and adopts the Adam optimizer. The learning rate is set to 0.001, the first moment estimation exponential decay rate is set to 0.9, and the training is automatically stopped when the validation set loss no longer decreases for 10 consecutive rounds to prevent overfitting.

[0080] After training, the blasting parameters corresponding to each detonation segment of the current blasting operation are input into the model, and the model outputs the waveform time-domain sequence corresponding to each detonation segment. The waveform time-domain sequences of all detonation segments are superimposed according to the detonation time sequence to form a complete predicted vibration waveform time-domain sequence.

[0081] Step 2, as the core prediction engine of the overall scheme, is significant because it establishes a nonlinear mapping relationship between blasting parameters and the time-domain sequence of vibration waveforms, thereby achieving accurate digital simulation of the blasting vibration process. Based on a standardized parameter set, this step utilizes a long short-term memory network to capture the temporal dependency between blasting parameters and vibration waveforms, accurately predicting the time-domain sequence of vibration waveforms generated in each detonation stage, and obtaining a complete predicted vibration waveform through temporal superposition. This process overcomes the limitation of traditional empirical formulas that can only predict peak vibration velocity, providing waveform data containing complete vibration characteristics such as amplitude, frequency, and duration, thus providing an unprecedented data foundation for subsequent safety assessments.

[0082] Compared with existing technologies, the advantages of this step are mainly reflected in three aspects: First, in terms of prediction dimension, existing technologies are mostly limited to predicting a single indicator of vibration intensity or safety distance, while this invention directly predicts the complete vibration waveform time-domain sequence, and can obtain key features such as the dominant frequency and vibration duration, achieving a qualitative leap from "intensity prediction" to "waveform reconstruction"; Second, in terms of algorithm advancement, the use of deep learning methods instead of traditional empirical formulas or statistical regression can better handle the complex temporal relationships and nonlinear characteristics of multi-stage micro-delay blasting, significantly improving prediction accuracy; Third, in terms of practicality, by superimposing the waveforms of each segment according to the detonation sequence, the vibration superposition effect in actual blasting is realistically simulated, overcoming the oversimplification problem of simplifying multi-stage blasting to a single sound source in existing technologies, and providing technical support for accurately assessing the cumulative impact of blasting vibration on the surrounding environment.

[0083] Step 3: Based on the geological parameters and surrounding environmental parameters in the standardized parameter set, determine the geological condition influencing factor and the environmental sensitivity influencing factor. Perform a fusion analysis on the geological condition influencing factor, the environmental sensitivity influencing factor, and the predicted vibration waveform time-domain sequence to generate a comprehensive safety assessment index for quantitatively evaluating the safety of the current blasting.

[0084] In this embodiment, geological condition influencing factors and environmental sensitivity influencing factors are determined based on geological parameters and surrounding environmental parameters in the standardized parameter set.

[0085] The formula used to calculate the geological condition influence factor is as follows:

[0086] ;

[0087] In the formula, For the first Geological conditions at the center of an explosion source are a factor influencing this. For the first in the standardized parameter set Joint and fracture density at the center of the explosion source The wave impedance influence factor is, where For the first in the standardized parameter set The density of rock at the center of the explosion source For the first in the standardized parameter set Shear wave velocity of the soil layer at the center of the explosion source For the first in the standardized parameter set Elastic modulus of rock mass at the center of the explosion source To standardize the parameters, the maximum value of the rock mass elastic modulus is required. For the first in the standardized parameter set Poisson's ratio of the rock mass at the center of the explosion source. , , and For preset weights, , , , .

[0088] The setting of weighting coefficients follows The principle behind this is that joint and fracture density is the most critical factor controlling the integrity of rock mass and the attenuation of blasting vibration wave energy. A large number of fractures significantly enhances the scattering and absorption effects of waves, hence it is given the highest weight. Wave impedance directly determines the propagation efficiency and energy dissipation of stress waves in soil and rock media, and is a core physical indicator affecting vibration intensity; therefore, it is a crucial factor in determining the weight of the wave. Secondly, the elastic modulus reflects the stiffness of the rock mass and has a significant impact on vibration frequency characteristics, but its sensitivity is lower than that of wave impedance, hence its weight is less important. It ranks third; Poisson's ratio mainly characterizes the lateral deformation capacity of rock masses, and its influence on vibration propagation is relatively indirect and weak, therefore it is given the lowest weight. This weighting strategy ensures that the geological condition influence factor can most effectively capture the geological characteristics that play a dominant role in the blasting vibration of shallow-buried tunnels.

[0089] In the above formula, the dependent variable As a geological condition influencing factor, specifically reflecting the first The comprehensive influence of geological conditions at the blast source center point of each detonation segment on the propagation of blasting vibration waves is a quantitative integration of key geological characteristics such as rock mass integrity and mechanical properties. Its core meaning is to characterize the comprehensive effect of geological conditions on the amplitude, frequency, and attenuation law of vibration waves through a single index. The technical effect is to transform multi-dimensional geological parameters into quantitative factors that can be directly correlated with blasting parameters.

[0090] The correlation between the independent and dependent variables stems from the direct influence of each parameter on the propagation of blasting vibration waves: joint and fracture density. The degree of rock mass fracturing determines the reflection, refraction, and energy attenuation of vibration waves; wave impedance influence factor. Reflecting the rock mass's ability to transmit vibrational waves, it determines wave velocity and energy transfer efficiency; elastic modulus ratio Differences in rock mass stiffness affect the attenuation rate of vibration waves; The brittle characteristics of the associated rock mass affect the waveform distortion of the vibration wave. These independent variables are all core parameters controlling the propagation characteristics of vibration waves in geological conditions, and therefore, together with the dependent variable characterizing the overall influence, they are crucial. There is an inherent connection between them, and their combined effect determines the ultimate impact of geological conditions on vibration propagation.

[0091] With joint and fracture density There is a negative correlation, meaning that the denser the joints in the rock mass, the weaker the overall beneficial influence of geological conditions on vibration propagation (such as stable energy transfer); and it is negatively correlated with the wave impedance influence factor. Elastic modulus ratio , All are positively correlated, that is, the greater the wave impedance, the higher the elastic modulus of the rock mass, and the smaller the Poisson's ratio, the stronger the comprehensive beneficial influence of geological conditions on vibration propagation, and the better the energy transfer and waveform stability of the vibration wave.

[0092] The rationality of this formula is reflected in three aspects: First, by using a weighted summation of individual terms, geological parameters of different dimensions, such as joint fissure density, wave impedance, elastic modulus, and Poisson's ratio, are integrated into a single influencing factor. This preserves the independent effect of each parameter on vibration propagation while reflecting the differences in their degree of influence through weight allocation, which is consistent with the actual characteristics of the combined effects of multiple factors in geological conditions. Second, the formula adopts a weighted summation of joint fissure density... In the form of Poisson's ratio, This form naturally establishes an inverse correlation between parameters and the influence of geological conditions: the denser the joints and the larger the Poisson's ratio, the weaker the beneficial influence on vibration propagation, which is consistent with the law of rock mass properties on vibration waves in engineering practice; thirdly, the elastic modulus adopts a relative value. Both the direct product form of wave impedance and the dimensionless integration of parameters after standardization ensure that the calculation conflicts between different physical dimensions are avoided, and the formula output has a stable quantitative meaning, laying a reliable foundation for the next step.

[0093] Table 1: Statistical Table of Geological Condition Influence Factors

[0094]

[0095] Please see Figures 3-5 Through visualization analysis from different dimensions, the physical relationship and changing trends between the geological condition influencing factors and various input parameters were verified, proving the rationality of the formula. Furthermore, based on the 15 sets of data in Table 1, a significant correlation exists between the geological condition influencing factors and various parameters: when the joint and fracture density is low, the geological condition influencing factors are generally high, reflecting the positive effect of rock mass integrity on geological conditions; while the increase in wave impedance and elastic modulus also leads to an increase in the geological condition influencing factors, demonstrating the important influence of rock mass mechanical properties on overall geological conditions. Poisson's ratio shows an inverse correlation, that is, the smaller the Poisson's ratio, the larger the geological condition influencing factor tends to be, consistent with the influence of rock mass brittleness on vibration propagation.

[0096] Overall, the changing trend of the geological condition influencing factors is consistent with the combined effect of multiple parameters, verifying the rationality of the constructed formula. When the rock mass has good integrity and excellent mechanical properties (such as low joint density and high elastic modulus), the geological condition influencing factors are at a higher level, and vice versa. This is consistent with the actual influence of geological conditions on vibration propagation in shallow tunnel blasting, indicating that the formula can effectively integrate key geological parameters and provide a reliable quantitative basis for geological conditions for the subsequent calculation of the comprehensive safety assessment index.

[0097] For each detonation stage The surrounding environmental parameters corresponding to the explosion source center point are extracted from the standardized parameter set, including building burial depth, building equivalent stiffness, underground pipeline diameter, and underground pipeline burial depth. Based on the above parameters, the building reflection correction coefficient and the underground pipeline attenuation correction coefficient are calculated to determine the environmental sensitivity influence factor. The calculation formula is as follows:

[0098] ;

[0099] In the formula, For the first The building reflection correction factor at the center point of the explosion source. For the first Attenuation correction factor for underground pipelines at the center point of the explosion source For the first Environmental sensitivity factors affecting the center point of an explosion source For the first in the standardized parameter set The equivalent stiffness of the building at the center point of the explosion source This represents the maximum value of the equivalent stiffness of the building within the standardized parameter set; For the first in the standardized parameter set The diameter of the underground pipeline at the center point of the explosion source, The maximum value of the underground pipeline diameter in the standardized parameter set; For the first in the standardized parameter set The burial depth of underground pipelines at the center point of the explosion source; and This is a preset proportional coefficient. , .

[0100] In the above formula, the building reflection correction factor and underground pipeline attenuation correction factor Its form has clear physical meaning and rationality:

[0101] For building reflection correction factor The construction of its calculation formula reflects the equivalent stiffness of the building. The effect on reflectivity. (Through...) The proportional relationship, by standardizing the stiffness at the center point of the i-th explosion source, can accurately reflect the contribution of relative intensity to reflection. Simultaneously, the exponential decay term... The influence of the burial depth of underground pipelines is taken into account. The greater the burial depth, the weaker the reflection ability may be. Therefore, the exponential decay form can more reasonably characterize this physical phenomenon.

[0102] Secondly, regarding the attenuation correction factor for underground pipelines The calculation formula emphasizes the impact of underground pipeline diameter on signal attenuation. Through... The standardization reflects the contribution of relative pipeline size to attenuation. Meanwhile, The influence of underground pipeline burial depth is partially considered; the deeper the burial, the more significant the attenuation effect. Combined with the characteristics of the exponential function, this helps to accurately describe the attenuation process of the pipeline's effect on the explosion source signal. Therefore, both formulas are reasonably constructed and consistent with engineering practice, contributing to the effective quantification and analysis of environmental factors.

[0103] Finally, regarding the factors influencing environmental sensitivity The value of this factor comprehensively reflects the sensitivity of the environment surrounding the blasting point to vibration response. A higher value indicates a higher environmental sensitivity at the blast source's center point, and a greater risk of environmental impact from blasting vibration. This factor mathematically couples the building reflection effect with the attenuation effect of underground pipelines, achieving a unified quantification of sensitivity to different types of environments and providing key environmental impact parameters for subsequent safety assessments. In the calculation formula, the independent variable The reflection effect is related to the building's equivalent stiffness and burial depth; the greater the stiffness and the shallower the burial depth, the more significant the reflection effect. (Independent variable) The vibration damping capability is related to the pipeline diameter and burial depth; the larger the diameter and the shallower the burial depth, the weaker the vibration damping capability. use and The form of adding 1 to the difference reflects both the antagonistic relationship between the building's vibration-enhancing reflection effect and the pipeline's vibration-reducing attenuation effect, and ensures that the result is positive and has a clear physical meaning through the addition of 1: Time reflection effect dominates. The time decay effect is dominant.

[0104] For the first formula in the above formula group, the dependent variable... The building reflection correction coefficient specifically reflects the degree of quantitative correction of the intensity of the building's reflection of blasting vibration waves. It means that by integrating the building's equivalent stiffness and burial depth parameters, the comprehensive influence of the building on the reflection of vibration waves is quantified. The technical effect is to introduce the correction basis of building reflection into the initial waveform, so that the waveform is closer to the real state of interference from building reflection in actual propagation.

[0105] Building equivalent stiffness Its ability to reflect vibrational waves is directly determined by its stiffness; the greater the stiffness, the stronger the reflection. Related; burial depth The building's efficiency in reflecting ground vibration waves is affected; the greater the burial depth, the weaker the reflection effect, hence the correlation is established through an exponential term. Therefore, Equivalent stiffness of building It is positively correlated with burial depth. It shows a negative correlation.

[0106] For the second formula in the above formula group, the dependent variable... The underground pipeline attenuation correction coefficient reflects the degree of quantitative correction of the intensity of the energy attenuation effect of underground pipelines on vibration waves. It means integrating pipeline diameter and burial depth parameters to quantify the attenuation effect of pipelines on vibration waves. The technical effect is to improve the environmental adaptability of waveform prediction by correcting the initial waveform, reflecting the absorption and blocking effect of underground pipelines on vibration energy.

[0107] for underground pipeline diameter The diameter determines its ability to block and attenuate vibration waves; the larger the diameter, the more significant the attenuation. Burial depth... The range of influence of pipelines on ground vibration is affected; the greater the burial depth, the weaker the attenuation effect. Therefore, both are related to... Relatedly, they jointly reflect the attenuation effect of the pipeline through the correction term. Therefore, With underground pipeline diameter It shows a negative correlation with the burial depth of underground pipelines. They are positively correlated.

[0108] A comprehensive safety assessment index is generated by fusing geological condition influencing factors, environmental sensitivity influencing factors, and predicted vibration waveform time-domain sequences to quantitatively evaluate the safety of the current blasting operation. The formula used is as follows:

[0109] ;

[0110] In the formula, To form a comprehensive safety assessment index, Indicates the number of detonation stages. The peak particle vibration velocity of the predicted vibration waveform time-domain sequence for the i-th initiation segment; The dominant frequency of the predicted vibration waveform time-domain sequence for the i-th detonation segment;

[0111] in, and The vibration waveforms of each detonation segment predicted in step 2 are obtained by performing spectral analysis.

[0112] Dependent variable It is a dimensionless, comprehensive quantitative index used to fully reflect the overall safety risk level of the current blasting operation. The higher the value, the higher the potential risk to the surrounding environment and structures caused by the blasting operation. By integrating three major categories of factors—geological conditions, environmental sensitivity, and vibration waveform characteristics—this index achieves a multi-dimensional comprehensive evaluation of blasting safety, providing a scientific and objective quantitative basis for risk level classification and overcoming the limitations of single-indicator assessment.

[0113] Independent variables include geological condition influencing factors. Environmental sensitivity influencing factors Peak particle vibration velocity and main frequency . and These respectively reflect the amplification or attenuation effect of the geological conditions and surrounding environment at the blasting point on vibration propagation; It reflects the vibration intensity, and its square form highlights the impact of high-intensity vibration; This reflects the frequency characteristics of the vibration; vibrations of different frequencies have varying degrees of impact on structures. These independent variables, coupled through multiplication and summation, jointly determine the potential risk level of blasting vibrations.

[0114] In the formula, Value and , , A positive correlation exists: the practical reason is that when the geological conditions are worse, the environment is more sensitive, or the vibration velocity is greater, The value will increase accordingly; Value and The correlation is positive, but the growth is relatively slow, indicating that although high-frequency vibrations pose a higher risk, their impact is milder than that of vibration intensity. Furthermore, the use of logarithmic and square root functions not only compresses the dynamic range of the data, avoiding the dominance of extreme values, but also maintains the positive correlation between various influencing factors, making the final index more practical for engineering applications.

[0115] Step 3, as the core of the overall risk quantification scheme, is significant because it enables a refined and quantitative assessment of blasting vibration risk by constructing geological condition influencing factors and environmental sensitivity influencing factors. Based on standardized parameters, this step comprehensively considers key geological factors such as rock mass integrity (joint and fracture density), wave impedance characteristics (rock density and shear wave velocity), rock mass stiffness (elastic modulus), and deformation characteristics (Poisson's ratio), as well as environmental sensitivity factors such as building reflection effects and underground pipeline attenuation characteristics. Through multi-factor coupled mathematical modeling, complex geological environmental conditions are transformed into quantifiable influencing factors. Furthermore, the geological conditions, environmental sensitivity, and predicted vibration waveform characteristics are integrated and analyzed to generate a comprehensive safety assessment index that fully reflects the risk level of blasting operations, providing a scientific and objective quantitative basis for risk level classification.

[0116] Compared with existing technologies, the advantages of this step are mainly reflected in three aspects: First, in terms of assessment dimensions, existing technologies mostly adopt a single risk scoring method, while this invention innovatively proposes a dual assessment mechanism of geological condition influence factors and environmental sensitivity influence factors, which conducts refined assessments from the two dimensions of medium propagation characteristics and bearing sensitivity, respectively, realizing multi-dimensional deepening of risk assessment; Second, in terms of factor construction, existing technologies usually use simple linear weighting methods, while this invention establishes a physical model that includes nonlinear relationships such as exponential decay and logarithmic transformation, which more accurately reflects the propagation law of vibration waves in actual media and the interaction mechanism with environmental structures; Third, in terms of index fusion methods, existing technologies mostly remain at the simple data level of integration, while this invention adopts a method combining logarithmic compression and square root transformation, which not only highlights the influence of high-intensity vibrations, but also balances the numerical differences between parameters of different dimensions, making the final comprehensive safety assessment index more engineering interpretable and practical.

[0117] Step 4: Call the preset multi-level safety risk thresholds, match and analyze the comprehensive safety assessment index with the multi-level safety risk thresholds, and determine the risk level of the current blasting operation based on the matching results; the risk level includes four levels: safe, low risk, medium risk, and high risk;

[0118] In this embodiment, a preset multi-level safety risk threshold is invoked, and the comprehensive safety assessment index is matched and analyzed with the multi-level safety risk threshold. The risk level of the current blasting operation is determined based on the matching result. The specific logic is as follows:

[0119] like If so, the risk level of the current blasting operation is determined to be "safe";

[0120] like If so, the risk level of the current blasting operation is determined to be "low risk";

[0121] like If so, the risk level of the current blasting operation is determined to be "medium risk";

[0122] like If so, the risk level of the current blasting operation is determined to be "high risk";

[0123] In the formula, , , , These represent the preset upper limit of the safety threshold, the upper limit of the low-risk threshold, the upper limit of the medium-risk threshold, and the lower limit of the high-risk threshold, respectively, and satisfy the following conditions: .

[0124] The method for determining the multi-level safety risk thresholds is as follows: Based on a historical blasting case database, the distribution characteristics of the comprehensive safety assessment index Z-value and the correspondence between actual vibration effects are statistically analyzed, and threshold intervals are divided in conjunction with expert experience. Cluster analysis is used to identify the natural distribution boundaries of the Z-value, dividing the risk level into four typical intervals: safe, low risk, medium risk, and high risk. Machine learning algorithms are then used to optimize the threshold boundaries, ensuring that each threshold satisfies a strict numerical increasing relationship and matches the actual risk level of the project. After the thresholds are determined, they need to be verified through field tests and adjusted based on feedback to ultimately form a multi-level safety risk threshold system suitable for specific engineering conditions.

[0125] Step 4 will comprehensively evaluate the index With preset multi-level thresholds , , , By matching continuous risk quantification values, this method transforms them into four distinct levels: "safe," "low-risk," "medium-risk," and "high-risk," providing a direct decision-making basis for the subsequent precise implementation of differentiated control measures. Compared to the simple "safe / dangerous" dichotomy or the coarse judgment relying on a single empirical threshold in existing technologies, this method achieves refined differentiation of risk levels through a multi-level threshold system optimized based on historical data, significantly improving the pertinence and operability of risk control.

[0126] Step 5: Implement differentiated control measures based on the determined risk level: for the safety level, implement the original plan and conduct routine monitoring simultaneously; for the low-risk level, activate real-time high-frequency vibration monitoring; for the medium-risk level, reduce the amount of single-stage detonating explosive and extend the detonation time difference proportionally on the basis of low-risk control; for the high-risk level, suspend blasting operations and take enhanced vibration reduction measures, which will be implemented after the risk is reduced.

[0127] In this embodiment, the specific execution process of step 5 is as follows:

[0128] Based on the determined risk level of the current blasting operation, corresponding graded control and real-time feedback operations are implemented: If the risk level is low, real-time high-frequency vibration monitoring sensors are deployed in areas where the predicted peak particle vibration velocity is greater than 0.5 cm / s and in sensitive structures within 50 m of the blast source center point during blasting, with a sampling frequency of not less than 1000 Hz, and the monitoring data is dynamically fed back to the management platform; if the risk level is medium, based on the low-risk monitoring, the amount of explosive charge per stage is reduced to 70% of the original design value, and the detonation time difference is extended to 1.5 times the original design value; if the risk level is high, the blasting operation is immediately suspended, and enhanced vibration reduction measures are taken, and the risk level is reassessed until it drops to medium or below; at the same time, the control measures and execution results under each risk level are recorded in the database for optimizing the early warning model and threshold settings.

[0129] Step 5 translates the risk assessment results into specific control actions, achieving closed-loop management from prediction and early warning to precise execution. Differentiated measures are set for different risk levels: for low risk, focus on monitoring high-frequency vibration areas; for medium risk, quantitatively adjust blasting parameters: reduce explosive charge to 70%, extend the time difference to 1.5 times; for high risk, immediately halt work and implement enhanced measures. Compared to the vague expressions of "strengthening monitoring" or "adjusting parameters" in existing technologies, this solution provides clear quantitative execution standards and operating procedures, ensuring that control measures are implementable and verifiable. Simultaneously, through full-process data recording, feedback is provided for model optimization, forming a virtuous cycle of continuous improvement.

[0130] Please see Figure 2 A surface vibration waveform prediction system for shallow tunnel blasting includes:

[0131] The data acquisition and preprocessing module is used to collect the basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting. The basic parameters include geological parameters, blasting parameters, and surrounding environmental parameters. A three-level index table is designed based on parameter type to classify and store the basic parameters. The effective parameters in the three-level index table are filtered by an outlier detection algorithm, and the effective parameters are standardized to obtain a standardized parameter set.

[0132] The mapping relationship determination module is used to analyze the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences, establish the mapping relationship between the blasting parameters and the vibration waveform time-domain sequences, and determine the predicted vibration waveform time-domain sequence corresponding to the current blast based on the mapping relationship.

[0133] The safety index fusion calculation module is used to determine the geological condition influence factor and the environmental sensitivity influence factor based on the geological parameters and surrounding environmental parameters in the standardized parameter set. It performs fusion analysis on the geological condition influence factor, the environmental sensitivity influence factor and the predicted vibration waveform time domain sequence to generate a comprehensive safety assessment index for quantitatively evaluating the safety of the current blasting.

[0134] The risk grading module is used to call preset multi-level safety risk thresholds, match and analyze the comprehensive safety assessment index with the multi-level safety risk thresholds, and determine the risk level of the current blasting operation based on the matching results; the risk level includes four levels: safe, low risk, medium risk, and high risk.

[0135] The graded control measures execution module is used to implement differentiated control measures based on the determined risk level: for the safety level, the original plan is implemented and routine monitoring is carried out simultaneously; for the low-risk level, real-time high-frequency vibration monitoring is activated; for the medium-risk level, the amount of single-stage detonating explosive is reduced proportionally and the detonation time difference is extended on the basis of low-risk control; for the high-risk level, blasting operations are suspended and enhanced vibration reduction measures are taken, which will be implemented after the risk is reduced.

[0136] 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.

[0137] 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.

[0138] 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.

[0139] 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 predicting surface vibration waveforms during blasting in shallow-buried tunnels, characterized in that, The specific steps include: Step 1: Collect the basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting operation; the basic parameters include geological parameters, blasting parameters and surrounding environmental parameters. Based on the parameter type, design a three-level index table to classify and store the basic parameters. Use an outlier detection algorithm to filter the effective parameters in the three-level index table, and standardize the effective parameters to obtain a standardized parameter set; Step 2: Analyze the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences, establish the mapping relationship between the blasting parameters and the vibration waveform time-domain sequences, and determine the predicted vibration waveform time-domain sequence corresponding to the current blast based on the mapping relationship; Step 3: Based on the geological parameters and surrounding environmental parameters in the standardized parameter set, determine the geological condition influencing factor and the environmental sensitivity influencing factor. Perform a fusion analysis on the geological condition influencing factor, the environmental sensitivity influencing factor, and the predicted vibration waveform time-domain sequence to generate a comprehensive safety assessment index for quantitatively evaluating the safety of the current blasting. Step 4: Call the preset multi-level safety risk thresholds, match and analyze the comprehensive safety assessment index with the multi-level safety risk thresholds, and determine the risk level of the current blasting operation based on the matching results; the risk level includes four levels: safe, low risk, medium risk, and high risk; Step 5: Implement differentiated control measures based on the determined risk level: for the safety level, implement the original plan and conduct routine monitoring simultaneously; for the low-risk level, activate real-time high-frequency vibration monitoring; for the medium-risk level, reduce the amount of single-stage detonating explosive and extend the detonation time difference proportionally on the basis of low-risk control; for the high-risk level, suspend blasting operations and take enhanced vibration reduction measures, which will be implemented after the risk is reduced. Based on the geological parameters and surrounding environmental parameters in the standardized parameter set, the geological condition influencing factors and environmental sensitivity influencing factors are determined. The formula used to calculate the geological condition influence factor is as follows: In the formula, For the first Geological conditions at the center of an explosion source are a factor influencing this. For the first in the standardized parameter set Joint and fracture density at the center of the explosion source The wave impedance influence factor is, where For the first in the standardized parameter set The density of rock at the center of the explosion source For the first in the standardized parameter set Shear wave velocity of the soil layer at the center of the explosion source For the first in the standardized parameter set Elastic modulus of rock mass at the center of the explosion source To standardize the parameters, the maximum value of the rock mass elastic modulus is required. For the first in the standardized parameter set Poisson's ratio of the rock mass at the center of the explosion source. , , and For preset weights, , , and All are non-negative numbers, and satisfy the following conditions: ; For each detonation stage The surrounding environmental parameters corresponding to the explosion source center point are extracted from the standardized parameter set, including building burial depth, building equivalent stiffness, underground pipeline diameter, and underground pipeline burial depth. Based on the above parameters, the building reflection correction coefficient and the underground pipeline attenuation correction coefficient are calculated to determine the environmental sensitivity influence factor. The calculation formula is as follows: In the formula, For the first The building reflection correction factor at the center point of the explosion source. For the first Attenuation correction factor for underground pipelines at the center point of the explosion source For the first Environmental sensitivity factors affecting the center point of an explosion source For the first in the standardized parameter set The equivalent stiffness of the building at the center point of the explosion source This represents the maximum value of the equivalent stiffness of the building within the standardized parameter set; For the first in the standardized parameter set The diameter of the underground pipeline at the center point of the explosion source, The maximum value of the underground pipeline diameter in the standardized parameter set; For the first in the standardized parameter set The burial depth of underground pipelines at the center point of the explosion source; A comprehensive safety assessment index is generated by fusing geological condition influencing factors, environmental sensitivity influencing factors, and predicted vibration waveform time-domain sequences to quantitatively evaluate the safety of the current blasting operation. The formula used is as follows: In the formula, To form a comprehensive safety assessment index, Indicates the number of detonation stages. For the index of the detonation phase, The peak particle vibration velocity of the predicted vibration waveform time-domain sequence for the i-th initiation segment; The dominant frequency of the predicted vibration waveform time-domain sequence for the i-th detonation segment.

2. The method for predicting surface vibration waveforms during shallow tunnel blasting according to claim 1, characterized in that: The basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting, were collected. The specific data acquisition method is as follows: Through borehole exploration, ground-penetrating radar scanning, and on-site rock mechanics tests, the radius of the explosion source center point of the detonation section was determined. Geological parameters at multiple sampling points within the range, including rock mass elastic modulus, rock mass Poisson's ratio, rock mass density, joint and fissure density, and soil shear wave velocity, are used as the average values ​​as the geological parameter values ​​at the corresponding explosion source center point. Among them, soil shear wave velocity refers to the speed at which shear waves propagate in the soil layer and is used to evaluate the softness and looseness of the soil layer. The blasting parameters are obtained from the blasting design scheme and initiation system, including the maximum charge per segment of each initiation section, the segment initiation sequence, the borehole depth, and the spatial coordinates of the corresponding blast source center point; wherein the segment initiation sequence... Represented as , Indicates the first The detonation moment of the explosive at the center point of the blast source. For the index of the detonation phase, , This indicates the number of detonation segments, with each detonation segment corresponding to a blast center point; By analyzing the design drawings and conducting on-site inspections, it was determined that the center point of the detonation source in the initiation section should be the center of a circle with a radius of... The surrounding environmental parameters within the range include building burial depth, building equivalent stiffness, underground pipeline diameter, and underground pipeline burial depth. For each explosion source center point, the building burial depth and building equivalent stiffness of the nearest building to the explosion source center point are taken as representative values. The shortest vertical distance from the explosion source center point to each nearby pipeline is calculated, and the underground pipeline diameter and underground pipeline burial depth corresponding to the pipeline with the minimum shortest vertical distance are taken as representative values. If there are multiple pipelines corresponding to the minimum shortest vertical distance, the underground pipeline diameter and underground pipeline burial depth of the pipeline with the most vibration-sensitive material are taken as representative values. The three-level index table has a first-level index for parameter categories, namely geological parameters, blasting parameters, and surrounding environment parameters; a second-level index for the specific parameter names under each parameter category; and a third-level index for the collection timestamps and spatial coordinates of each parameter.

3. The method for predicting surface vibration waveforms during shallow tunnel blasting according to claim 2, characterized in that: When filtering valid parameters of the three-level index table using the outlier detection algorithm, a dual threshold method based on physical meaning and statistical distribution is adopted: First, a reasonable range of values ​​for each parameter is set based on engineering experience, and parameter values ​​that exceed the reasonable range are judged as outliers and removed. Secondly, for parameters within a reasonable range, the box plot method is used to identify and remove parameter values ​​that are less than the lower quartile minus 3 times the interquartile range or greater than the upper quartile plus 3 times the interquartile range as discrete outliers. The standardization process employs the range standardization method, transforming the selected parameter values ​​to the range using the following formula. Interval: in For the original value of the parameter, For the standardized values ​​of the parameters, , These are the minimum and maximum values ​​of the current parameter to be standardized in the set of valid parameters after outlier filtering; After the above processing, a standardized parameter set containing standardized values ​​of geological parameters, blasting parameters, and surrounding environmental parameters is obtained.

4. The method for predicting surface vibration waveforms during shallow tunnel blasting according to claim 3, characterized in that: The blasting parameters and their corresponding vibration waveform time-domain sequences for each initiation segment in the standardized parameter set are analyzed. A mapping relationship between the blasting parameters and the vibration waveform time-domain sequences is established, and the predicted vibration waveform time-domain sequence for the current blast is determined based on this mapping relationship. The specific logic is as follows: A machine learning model is constructed based on long short-term memory network. The standardized parameter set includes the blasting parameters and vibration waveform time-domain sequences of historical shallow tunnel blasting construction. The sets are divided into training set, validation set and test set according to the proportions of 70%, 15% and 15% respectively. The blasting parameters corresponding to each blasting segment are used as the model input, and the vibration waveform time-domain sequences of each blasting segment are used as the model output for model training. The model has three LSTM hidden layers with 128, 64, and 32 neurons respectively. The model training uses the mean squared error (MSE) as the loss function and adopts the Adam optimizer. The learning rate is set to 0.001, the first moment estimation exponential decay rate is set to 0.9, and the training is automatically stopped when the validation set loss no longer decreases for 10 consecutive rounds to prevent overfitting. After training, the blasting parameters corresponding to each detonation segment of the current blasting operation are input into the model, and the model outputs the waveform time-domain sequence corresponding to each detonation segment. The waveform time-domain sequences of all detonation segments are superimposed according to the detonation time sequence to form a complete predicted vibration waveform time-domain sequence.

5. The method for predicting surface vibration waveforms during shallow tunnel blasting according to claim 1, characterized in that: The system invokes preset multi-level safety risk thresholds, matches the comprehensive safety assessment index with the multi-level safety risk thresholds, and determines the risk level of the current blasting operation based on the matching results. The specific logic behind this is as follows: like If so, the risk level of the current blasting operation is determined to be "safe"; like If so, the risk level of the current blasting operation is determined to be "low risk"; like If so, the risk level of the current blasting operation is determined to be "medium risk"; like If so, the risk level of the current blasting operation is determined to be "high risk"; In the formula, , , , These represent the preset upper limit of the safety threshold, the upper limit of the low-risk threshold, the upper limit of the medium-risk threshold, and the lower limit of the high-risk threshold, respectively, and satisfy the following conditions: .

6. The method for predicting surface vibration waveforms during shallow tunnel blasting according to claim 1, characterized in that: The specific execution process of step 5 is as follows: Based on the determined risk level of the current blasting operation, corresponding graded control and real-time feedback operations are implemented: If the risk level is low, real-time high-frequency vibration monitoring sensors are deployed in areas where the predicted peak particle vibration velocity is greater than 0.5 cm / s and in sensitive structures within 50 m of the blast source center point during blasting, with a sampling frequency of not less than 1000 Hz, and the monitoring data is dynamically fed back to the management platform; if the risk level is medium, based on the low-risk monitoring, the amount of explosive charge per stage is reduced to 70% of the original design value, and the detonation time difference is extended to 1.5 times the original design value; if the risk level is high, the blasting operation is immediately suspended, and enhanced vibration reduction measures are taken, and the risk level is reassessed until it drops to medium or below; at the same time, the control measures and execution results under each risk level are recorded in the database for optimizing the early warning model and threshold settings.

7. A surface vibration waveform prediction system for shallow-buried tunnel blasting, characterized in that: The shallow tunnel blasting surface vibration waveform prediction system is used to execute the shallow tunnel blasting surface vibration waveform prediction method according to any one of claims 1-6, including: The data acquisition and preprocessing module is used to collect the basic parameters and corresponding vibration waveform time-domain sequences of multiple shallow-buried tunnel blasting operations, as well as the basic parameters of the current shallow-buried tunnel blasting. The basic parameters include geological parameters, blasting parameters, and surrounding environmental parameters. A three-level index table is designed based on parameter type to classify and store the basic parameters. The effective parameters in the three-level index table are filtered by an outlier detection algorithm, and the effective parameters are standardized to obtain a standardized parameter set. The mapping relationship determination module is used to analyze the blasting parameters of each detonation segment in the standardized parameter set and their corresponding vibration waveform time-domain sequences, establish the mapping relationship between the blasting parameters and the vibration waveform time-domain sequences, and determine the predicted vibration waveform time-domain sequence corresponding to the current blast based on the mapping relationship. The safety index fusion calculation module is used to determine the geological condition influence factor and the environmental sensitivity influence factor based on the geological parameters and surrounding environmental parameters in the standardized parameter set. It performs fusion analysis on the geological condition influence factor, the environmental sensitivity influence factor and the predicted vibration waveform time domain sequence to generate a comprehensive safety assessment index for quantitatively evaluating the safety of the current blasting. The risk grading module is used to call preset multi-level safety risk thresholds, match and analyze the comprehensive safety assessment index with the multi-level safety risk thresholds, and determine the risk level of the current blasting operation based on the matching results; the risk level includes four levels: safe, low risk, medium risk, and high risk. The graded control measures execution module is used to implement differentiated control measures based on the determined risk level: for the safety level, the original plan is implemented and routine monitoring is carried out simultaneously; for the low-risk level, real-time high-frequency vibration monitoring is activated; for the medium-risk level, the amount of single-stage detonating explosive is reduced proportionally and the detonation time difference is extended on the basis of low-risk control; for the high-risk level, blasting operations are suspended and enhanced vibration reduction measures are taken, which will be implemented after the risk is reduced.

Images