Intelligent monitoring method for mine blasting vibration

Abstract

The invention relates to the technical field of vibration monitoring, in particular to an intelligent mine blasting vibration monitoring method, which comprises the following steps of: acquiring initial three-dimensional vibration waveform data of a mine blasting area by using a distributed vibration sensor, dividing the initial three-dimensional vibration waveform data into a plurality of monitoring time windows according to time sequence coordinates, and calculating the frequency spectrum energy concentration ratio in each time window; the method comprises the following steps: firstly, analyzing a waveform, determining an initial reference window, carrying out layer-by-layer analysis by utilizing a multi-layer analysis mechanism according to the determined initial reference window, optimizing the waveform of an analysis result through an optimization iteration mechanism, and carrying out accuracy verification on the optimized waveform; and constructing a mine vibration propagation model based on the three-dimensional energy distribution array verified twice, and performing intelligent monitoring on mine blasting vibration according to the constructed mine vibration propagation model. According to the method, a frequency band boundary dynamic adjustment technology of primary energy frequency band screening and multi-feature time point comparison is adopted, so that a refined capturing effect of blasting vibration high-energy features is realized.

Description

Technical Field

[0001] This invention relates to the technical field of vibration monitoring, specifically to an intelligent monitoring method for vibration in mine blasting. Background Technology

[0002] With the rapid development of technology, the demand for mine blasting is increasing. However, at present, the following challenges still exist in monitoring blasting vibrations:

[0003] The selection of vibration time windows relies on manual experience or fixed thresholds, making it difficult to accurately locate the core period of blasting, which limits the accuracy of subsequent identification.

[0004] Traditional frequency band division uses fixed bandwidth or bandpass filtering methods, which cannot adjust the frequency band boundaries according to the dynamic changes of the blasting signal, and are prone to missing key energy areas;

[0005] The lack of a comprehensive index system for evaluating waveform quality means that a single energy or correlation index cannot accurately verify whether vibration characteristics are real and effective.

[0006] Existing systems struggle to accurately predict safe distances based on propagation models, leading to reliance on empirical values ​​for blasting safety monitoring, which carries high risks. Summary of the Invention

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an intelligent monitoring method for mine blasting vibration, comprising the following steps,

[0008] Vibration waveform data from the mine blasting area were collected, and the collected waveforms were optimized using an optimization iteration mechanism. Simultaneously, the accuracy of the optimized waveforms was verified using historical blasting vibration characteristics. Specifically:

[0009] Initial three-dimensional vibration waveform data of the mine blasting area were collected using distributed vibration sensors. The data was divided into multiple monitoring time windows according to time series coordinates. The spectral energy concentration in each time window was calculated, and an initial reference window was determined. Based on the determined initial reference window, a multi-layer analysis mechanism was used for layer-by-layer analysis. At the same time, the waveform of the analysis results was optimized through an optimization iteration mechanism, and the accuracy of the optimized waveform was verified.

[0010] In addition, monitoring of mine blasting vibrations based on identified mine blasting vibration characteristics, specifically:

[0011] A three-dimensional energy distribution array is constructed based on the optimized waveform. The spatial resolution of the three-dimensional energy distribution array is calibrated and the consistency of the propagation path is verified in sequence. A mine vibration propagation model is constructed based on the three-dimensional energy distribution array that has passed the two verifications. Intelligent monitoring of mine blasting vibration is carried out based on the constructed mine vibration propagation model.

[0012] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the determination of the initial reference window is specifically as follows:

[0013] Based on the defined monitoring time windows, the spectral energy concentration for each monitoring time is calculated, and the calculation results are used as an effective indicator of the vibration signal. Therefore,

[0014] Let K be the total number of frequency components corresponding to the divided monitoring time windows. Then, perform a Fourier transform on the three-dimensional vibration signal matrix within the current monitoring time window. Simultaneously, calculate the Fourier transform of the sum of all three-dimensional vibration signal matrices across all frequency components within the monitoring time window. Based on the ratio between these two values, determine the spectral energy concentration of each monitoring time window. Then, we have...

[0015] The spectral energy concentration of each monitoring time window is calculated sequentially. After the spectral energy concentration of all monitoring time windows has been calculated, the monitoring time window with the largest spectral energy concentration is selected from all the calculation results as the initial reference window for identifying the vibration characteristics of mine blasting.

[0016] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the step of performing layer-by-layer analysis using a multi-layer analysis mechanism is as follows:

[0017] Given a defined initial parent time window, the first-generation analysis waveform is determined using a multi-feature time point comparison mechanism. Therefore,

[0018] Set an energy spectral density threshold, and filter the first-level energy frequency band region in the initial parent time window according to the set energy spectral density threshold. Based on the determined first-level energy frequency band region, determine the spectral energy difference between the initial parent time window and the adjacent time window. According to the calculated spectral energy difference, set a spectral energy difference threshold. Determine whether to execute the frequency band boundary adjustment strategy according to the set spectral energy difference threshold. Collect waveform data after each execution of frequency band boundary adjustment. After all feature time points are compared, determine the first generation analysis waveform.

[0019] Furthermore, the waveforms after each execution of the frequency band boundary adjustment strategy in the first-generation analysis waveform are subjected to feature enhancement processing, and the second-generation analysis waveform is constructed based on the results of the feature enhancement processing.

[0020] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the step of filtering the first-level energy frequency band region within the initial parent time window based on a set energy spectral density threshold is specifically as follows:

[0021] Based on the spectral energy E corresponding to the frequency point index sequence number k in the initial parent time window. k(W P ), and sequentially compared with the set energy spectral density threshold E T By comparison, if the comparison results satisfy formula E as the frequency point index sequence progresses... k (W P )≥E T Then, it represents the spectrum of the initial parent time window at frequency k, which is a first-level energy frequency band region. Based on the screening formula of the first-level energy frequency band, all frequency points in the initial parent time window that satisfy the first-level energy frequency band formula are combined to form a first-level energy frequency band region.

[0022] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the step of determining whether to execute a frequency band boundary adjustment strategy based on a set spectral energy difference threshold is as follows:

[0023] A characteristic time point is randomly selected from the first-level energy frequency band region. At the same time, the same characteristic time point is selected from the adjacent time windows, and the spectral energy difference ΔP between the initial parent time window and the same time point in the adjacent time windows is compared.

[0024] Set the spectral energy difference threshold ΔP T The decision to execute a frequency band boundary adjustment strategy is based on a set spectral energy difference threshold.

[0025] If the calculated spectral energy difference, when compared with the set spectral energy difference threshold, satisfies the formula ΔP≤ΔP T If the selected feature time point has not undergone feature evolution in the adjacent time window, the frequency band boundary adjustment strategy will not be executed.

[0026] If the calculated spectral energy difference, when compared with the set spectral energy difference threshold, satisfies the formula ΔP>ΔP T If the selected feature time point undergoes feature evolution in the adjacent time window, the frequency band boundary adjustment strategy is executed.

[0027] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the frequency band boundary adjustment strategy is specifically as follows:

[0028] By comparing the spectral density corresponding to the selected characteristic time point in the first-level energy frequency band region with the spectral density corresponding to the same time point in the adjacent time window, and implementing a frequency band boundary adjustment strategy based on the comparison results, we have:

[0029] If the spectral density of the same time point selected in the first-level energy frequency band region within an adjacent time window is greater than the spectral density of the same time point selected in the first-level energy frequency band region, then the frequency band boundary is expanded outward; otherwise, the frequency band boundary is contracted inward.

[0030] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the construction of the second-generation analysis waveform based on the result of feature enhancement processing is as follows:

[0031] Let T be the duration of the waveform after each execution of the frequency band boundary adjustment strategy in the first-generation analysis waveform. C And according to the midpoint T of the duration c / 2, the first generation analysis waveform is divided into two parts, the head and the tail, and the positions of the two parts are interchanged. At the same time, the waveform data after the position is interchanged is spliced ​​together. The spliced ​​waveform data is the waveform of the first generation analysis waveform after time reversal.

[0032] Based on the inverted waveform, feature enhancement is performed using enhancement coefficients. The set of all first-generation analysis waveforms after feature enhancement processing is the second-generation analysis waveform.

[0033] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the optimization of the analysis result waveform through an optimization iteration mechanism is specifically as follows:

[0034] The average energy envelope continuity of the first-generation analysis waveform is greater than the envelope smoothness index of the data within the initial parent time window.

[0035] The feature discrimination of any segment of the waveform in the second-generation analysis is better than that of the first-generation analysis.

[0036] If all the first-generation analysis waveforms can satisfy the iteration conditions for the second-generation analysis waveform, then the second-generation analysis waveform generated by iterating from the current first-generation analysis waveform will be used as the waveform for optimization.

[0037] As a preferred embodiment of the intelligent monitoring method for mine blasting vibration described in this invention, the accuracy verification of the optimized waveform is specifically as follows:

[0038] Based on the optimized waveform, the spectral energy concentration index, envelope smoothness index, and waveform cross-correlation residuals of each waveform segment are calculated, and a feature discrimination function f is constructed based on the results. D (J);

[0039] The blasting vibration characteristics of three consecutive time points were extracted from historical blasting vibration waveform data, and a blasting vibration characteristic threshold f was set. D (T), based on the set blast vibration characteristic threshold, the waveform accuracy is verified, then we have,

[0040] The blasting vibration features extracted from three consecutive time feature points are input into the constructed feature discrimination function. If the function output for each blasting vibration feature point is less than the set blasting vibration feature threshold, the accuracy verification of the optimized waveform is passed. Otherwise, the accuracy verification of the optimized waveform is not passed. The initial three-dimensional vibration waveform data of the mine blasting area is re-acquired and the waveform is optimized until the accuracy verification of the optimized waveform is completed.

[0041] The beneficial effects of this invention are:

[0042] This invention achieves precise positioning of the effective time window for blasting vibration by adopting a monitoring time window screening technology based on spectral energy concentration;

[0043] By adopting a frequency band boundary dynamic adjustment technology that combines primary energy frequency band screening with multi-feature time point comparison, a refined capture effect of high-energy characteristics of blasting vibration was achieved.

[0044] By employing a feature enhancement technique that combines a time-reversal structure with an enhancement coefficient, significant enhancement of vibration waveform characteristics and noise suppression were achieved.

[0045] By adopting a feature discrimination evaluation technique that combines envelope smoothness, spectral energy concentration and cross-correlation residuals, the reliability, stability and explosion feature integrity of the optimized waveform were accurately verified.

[0046] By adopting a propagation path consistency verification technique based on the energy wavefront structure and the distance of the energy attenuation gradient, a high-precision confirmation of the spatial continuity of the three-dimensional energy distribution array was achieved.

[0047] By adopting the technology of constructing a model of mine blasting vibration propagation and calculating the safe attenuation distance, intelligent monitoring and safety early warning of blasting vibration have been achieved. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0049] Figure 1 This is a schematic diagram of the overall method steps of the intelligent monitoring method for mine blasting vibration according to the present invention. Detailed Implementation

[0050] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0051] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0052] Example 1

[0053] Reference Figure 1 This is the first embodiment of the present invention, providing a method for intelligent monitoring of blasting vibration in mines, comprising the following steps.

[0054] S1: Collect vibration waveform data of the blasting area in the mine, optimize the collected waveform using an optimization iteration mechanism, and verify the accuracy of the optimized waveform using historical blasting vibration characteristics.

[0055] Specifically, the acquisition of vibration waveform data in the mine blasting area involves using distributed vibration sensors to collect initial three-dimensional vibration waveform data of the mine blasting area, and then scanning and identifying the vibration characteristics of the mine blasting based on the acquired initial three-dimensional vibration waveform data. The specific implementation is as follows:

[0056] Set up M sensors to collect initial three-dimensional vibration waveform data X={(X1,t),(X2,t),...,(X M ,t)}, where represents the initial three-dimensional vibration waveform data of the mining blasting area collected, and (X1,t) represents the waveform data collected by the first sensor at time t, (X M ,t) represents the waveform data collected by the Mth sensor at time t. Furthermore, the waveform data collected by each sensor also includes the corresponding three-dimensional coordinate data. The three-dimensional coordinate system is established based on the three-dimensional system of the mine terrain.

[0057] Based on the collected initial three-dimensional vibration waveform data, the vibration characteristics of mine blasting are scanned and identified, as follows:

[0058] The initial three-dimensional vibration waveform data is divided into N monitoring time windows according to time series coordinates, and the spectral energy concentration within each time window is calculated. The calculated spectral energy concentration is used as an effective indicator of the vibration signal. The time window with the highest effective indicator is selected from all monitoring time windows as the initial reference window for identifying the vibration characteristics of mine blasting. Specifically:

[0059] Dividing the time series into N monitoring time windows according to the time series coordinates, we have:

[0060] W = {W t |t=1,2,..,N}

[0061] Among them, W t The three-dimensional vibration signal matrix represents the t-th monitoring time window. It is multi-dimensional data, including the waveform data collected by each vibration sensor within the current time window, and the time acquisition point L corresponding to the vibration sensor acquiring the waveform data. t N represents the total number of monitoring time windows;

[0062] Based on the defined monitoring time windows, the spectral energy concentration for each monitoring time is calculated, and the calculation results are used as an effective indicator of the vibration signal. Therefore,

[0063] Let K be the total number of frequency components corresponding to the divided monitoring time windows. Then, perform a Fourier transform on the three-dimensional vibration signal matrix within the current monitoring time window. Simultaneously, calculate the Fourier transform of the sum of all three-dimensional vibration signal matrices across all frequency components within the monitoring time window. Based on the ratio between these two values, determine the spectral energy concentration of each monitoring time window. Then, we have...

[0064]

[0065] Where i represents the sequence index of the vibration sensor in the three-dimensional vibration signal matrix, j represents the sequence index of the time sampling point of the waveform data acquired by the vibration sensor, and L t This indicates the time points at which the vibration sensor collects waveform data, M represents the total number of vibration sensors, and W represents the time points at which the vibration sensor collects waveform data. t (i,j) represents the three-dimensional vibration signal matrix within the current monitoring time window, k represents the frequency component sequence index within the current monitoring time window, K represents the total number of frequency components within the current monitoring time window, f represents the Fourier transform function, and E t The spectral energy concentration of the t-th monitoring time window is used to determine the initial reference window for identifying mine blasting vibration characteristics, specifically:

[0066] The spectral energy concentration for each monitoring time window is calculated sequentially. After the spectral energy concentration for all monitoring time windows has been calculated, the monitoring time window with the highest spectral energy concentration is selected from all the calculation results as the initial reference window W for identifying mine blasting vibration characteristics. P (Initial parent generation time window), then we have,

[0067]

[0068] Among them, E t W represents the spectral energy concentration in the t-th monitoring time window, where t represents the index of the monitoring time window. P This indicates a defined initial parent generation time window.

[0069] Furthermore, based on the determined initial benchmark window for identifying mine blasting vibration characteristics, a multi-layer analysis and identification mechanism is used to perform layer-by-layer analysis, thereby completing the identification of mine blasting vibration characteristics. The specific implementation is as follows:

[0070] For a given initial parent time window W P The first-generation analysis waveform is determined using a multi-feature time point comparison mechanism, specifically as follows:

[0071] For the high-energy frequency band region of the initial parent time window, multiple characteristic time points are randomly selected, and the selected characteristic time points are compared with the corresponding regions in adjacent time windows for spectral energy. The frequency band boundaries are adjusted according to the comparison results. After all characteristic time points have been compared, the resulting waveform set is the first-generation analysis waveform. Therefore,

[0072] Based on a defined initial parent time window W P Set adjacent time windows W P-1 and W P+1 The energy spectral density threshold E was set based on historical blasting vibrations. T The energy spectrum is set based on the historical blasting vibrations. The specific values ​​are determined by the implementers according to the actual application scenario. The first-level energy frequency band region is filtered within the initial parent time window based on the set energy spectral density threshold. Specifically:

[0073] Based on the spectral energy E corresponding to the frequency point index sequence number k in the initial parent time window. k (W P The frequency index sequence is then compared sequentially with the set energy spectral density threshold. If the comparison result satisfies formula E as the frequency point index sequence progresses... k (W P )≥E TThen, it means that the spectrum of the initial parent time window at frequency k is a first-level energy frequency band region. Based on the screening formula of the first-level energy frequency band, all frequency points in the initial parent time window that satisfy the first-level energy frequency band formula are combined to form the first-level energy frequency band region A1.

[0074] Based on the defined first-level energy frequency band region, the spectral energy difference between the initial parent time window and adjacent time windows is determined, specifically as follows:

[0075] Randomly select a characteristic time point L from the first-level energy frequency band region. t Meanwhile, by selecting time points with the same characteristics from adjacent time windows and comparing the spectral energy difference between the initial parent time window and the same time points in adjacent time windows, we have:

[0076]

[0077] Where A1 represents a defined first-order energy frequency band region, L t E represents a characteristic time point randomly selected from the first-order energy frequency band. k (W P ,L t ) represents the selected characteristic time point L in the first-order energy frequency band region. t The corresponding spectral density, E k (W P-1 ,L t E represents the spectral density corresponding to the same characteristic time point selected in the first-order energy frequency band region within the preceding adjacent time window. k (W P+1 ,L t The spectral density at the same time point as the selected characteristic time point in the first-level energy frequency band region within the next adjacent time window is represented by ΔP, which represents the spectral energy difference between the selected characteristic time point and the same time point in the adjacent time window. This difference is used to determine whether to execute the frequency band boundary adjustment strategy. Specifically:

[0078] Set the spectral energy difference threshold ΔP T The decision to execute a frequency band boundary adjustment strategy is based on a set spectral energy difference threshold.

[0079] If the calculated spectral energy difference, when compared with the set spectral energy difference threshold, satisfies the formula ΔP≤ΔP T If the selected feature time point has not undergone feature evolution in the adjacent time window, the frequency band boundary adjustment strategy will not be executed.

[0080] If the calculated spectral energy difference, when compared with the set spectral energy difference threshold, satisfies the formula ΔP>ΔP TThis indicates that the selected feature time point undergoes feature evolution within an adjacent time window, and a frequency band boundary adjustment strategy is executed, specifically:

[0081] By comparing the spectral density corresponding to the selected characteristic time point in the first-level energy frequency band region with the spectral density corresponding to the same time point in the adjacent time window, and implementing a frequency band boundary adjustment strategy based on the comparison results, we have:

[0082] If the spectral density of the same time point selected in the first-level energy frequency band region within an adjacent time window is greater than the spectral density of the same time point selected in the first-level energy frequency band region, then the frequency band boundary is expanded outward; otherwise, the frequency band boundary is contracted inward.

[0083] Based on the implemented frequency band boundary adjustment strategy, the first-generation analysis waveform is determined as follows:

[0084] Waveform data is collected after each frequency band boundary adjustment. After comparing all characteristic time points, the first-generation analysis waveform is determined, and then...

[0085]

[0086] Where I represents the sequence number index for executing the frequency band boundary adjustment strategy, and K1 represents the total number of times the frequency band boundary adjustment strategy is executed. G1 represents the waveform data after the I-th execution of the frequency band boundary adjustment strategy, and G1 represents the determined first-generation analysis waveform.

[0087] Furthermore, based on the determined first-generation analysis waveform, a layer-by-layer analysis mechanism is used to identify the vibration characteristics of mine blasting, specifically as follows:

[0088] For the determined first-generation analysis waveform, a two-layer analysis mechanism is used to identify the characteristics of mine blasting vibrations, including a feature enhancement layer and an iterative optimization layer, specifically:

[0089] The waveforms obtained after each frequency band boundary adjustment strategy in the first-generation analysis waveform are subjected to feature enhancement processing. A second-generation analysis waveform is then constructed based on the results of this feature enhancement processing. Simultaneously, by iterating between the first and second-generation analysis waveforms, the two generations of analysis waveforms are optimized, thereby enabling the identification of mine blasting vibration characteristics. The specific implementation is as follows:

[0090] Let T be the duration of the waveform after each execution of the frequency band boundary adjustment strategy in the first-generation analysis waveform. C And if the waveform is divided according to the midpoint of time, then we have,

[0091] Based on the set waveform duration, the waveform splitting point for the first generation of analysis is determined, then...

[0092] The first-generation analysis waveform is analyzed according to the midpoint T of its duration. c / 2, the first generation analysis waveform is divided into two parts, the head and the tail, and the positions of the two parts are interchanged. At the same time, the waveform data after the position is interchanged is spliced ​​together. The spliced ​​waveform data is the waveform of the first generation analysis waveform after time reversal.

[0093] Based on the inverted waveform, feature enhancement is performed using enhancement coefficients, resulting in:

[0094]

[0095] in, This represents time point L in the waveform data after the I-th execution of the frequency band boundary adjustment strategy in the first-generation analysis waveform. t The corresponding data, where α represents the enhancement coefficient, is set by the implementers based on the actual application scenario. This represents time point L in the waveform data after the I-th execution of the frequency band boundary adjustment strategy in the second-generation analysis waveform. t The corresponding data.

[0096] It should be noted that the second-generation waveform is constructed by performing feature enhancement processing on all the first-generation analysis waveforms, thus, we have,

[0097]

[0098] Where I represents the sequence number index for executing the frequency band boundary adjustment strategy, and K1 represents the total number of times the frequency band boundary adjustment strategy is executed. G1 represents the result of feature enhancement processing on the waveform after the I-th execution of the frequency band boundary adjustment strategy, and G2 represents the determined second-generation analysis waveform.

[0099] Furthermore, an iterative optimization mechanism is used to iterate between the first-generation and second-generation analysis waveforms to optimize the waveforms, thereby enabling the identification of mine blasting vibration characteristics. Specifically:

[0100] The iteration between the first-generation analysis waveform and the data within the initial parent time window is as follows:

[0101] The average energy envelope continuity (envelope smoothness index) of the first-generation analyzed waveform is greater than the envelope smoothness index of the data within the initial parent generation time window.

[0102] The iteration between the first-generation and second-generation analysis waveforms is as follows:

[0103] The feature discrimination of any segment of the waveform in the second-generation analysis is better than that of the first-generation analysis.

[0104] Optimize the iteration termination condition:

[0105] If all the first-generation analysis waveforms satisfy the iteration conditions for the second-generation analysis waveform, then the second-generation analysis waveform generated iteratively from the current first-generation analysis waveform will be used as the optimized waveform. Based on the optimized waveform, waveform accuracy will be verified, specifically as follows:

[0106] Based on the optimized waveform, the spectral energy concentration index, envelope smoothness index, and waveform cross-correlation residual are calculated for each waveform segment. A feature discrimination function is then constructed based on the results.

[0107] f D (J)=β1·E(J)+β2·S(J)+β3·R(J)

[0108] Where β1, β2, and β3 represent weighting coefficients, which are set by the implementers according to the actual application scenario; E(J) represents the spectral energy concentration of the J-th waveform segment; S(J) represents the smoothness of the J-th waveform segment; R(J) represents the waveform cross-correlation residual of the J-th waveform segment; and f D (J) represents the result of the waveform segment J in the characteristic discrimination function. The accuracy is verified using historical mine blasting vibration characteristics. Specifically:

[0109] From historical blasting vibration waveform data, blasting vibration characteristics at three consecutive time points are extracted, and a blasting vibration characteristic threshold f is set. D (T), based on the set blast vibration characteristic threshold, the waveform accuracy is verified, then we have,

[0110] The blasting vibration features extracted from three consecutive time feature points are input into the constructed feature discrimination function. If the function output for each blasting vibration feature point is less than the set blasting vibration feature threshold, the accuracy verification of the optimized waveform is passed. Otherwise, the accuracy verification of the optimized waveform is not passed. The initial three-dimensional vibration waveform data of the mine blasting area is re-acquired and the waveform is optimized until the accuracy verification of the optimized waveform is completed.

[0111] It should be noted that the calculation of the smoothness index in the feature discrimination function is used to evaluate the continuity and stationarity of the energy envelope of the vibration event in the waveform, thereby verifying the integrity of the blast time. Specifically:

[0112] The waveform is subjected to Hilbert transform to extract the corresponding envelope. Simultaneously, a moving average filter is used to smooth the envelope, generating a smooth envelope curve. Peak and valley detection are then performed on the smoothed envelope curve to mark all energy peaks and valleys. Finally, it is determined whether the attenuation between each adjacent peak and valley follows an exponential decay law.

[0113] If the envelope curve monotonically decreases at the peak and the decay rate is within a predetermined range, it indicates that the envelope is smooth. The ratio between the total duration of all smooth envelopes in the waveform and the total duration of the waveform is calculated, and the result is the envelope smoothness index.

[0114] For the cross-correlation residuals in the feature discrimination function, which are suitable for measuring the similarity between the current waveform and adjacent waveforms, we have:

[0115] Select the current waveform and the waveforms of adjacent spatial position sensors, calculate the normalized cross-correlation coefficients between the previous adjacent waveform, the next adjacent waveform and the current waveform respectively. Then, calculate the mean between the two normalized cross-correlation coefficients, and take the result of subtracting the mean between the two normalized cross-correlation coefficients from 1 as the cross-correlation residual of the current waveform.

[0116] S2: Monitoring of mine blasting vibration based on identified mine blasting vibration characteristics.

[0117] Specifically, the monitoring of mine blasting vibration based on identified mine blasting vibration characteristics involves constructing a three-dimensional energy distribution array based on optimized waveforms, sequentially calibrating the spatial resolution of the three-dimensional energy distribution array and verifying the consistency of the propagation path, constructing a mine vibration propagation model based on the three-dimensional energy distribution array that has passed the two verifications, and then performing intelligent monitoring of mine blasting vibration based on the constructed mine vibration propagation model. The specific implementation is as follows:

[0118] For the optimized waveforms, the waveforms are arranged sequentially according to the time order of the initial reference window, and the rearranged result is used as the initial three-dimensional energy distribution array, specifically:

[0119] The optimized waveform is used to determine the energy density at the corresponding spatial location and frequency of each time node according to the time sequence of the initial reference window. After the energy density corresponding to all time nodes is determined, all energy densities are combined to form an initial three-dimensional energy distribution array.

[0120] Furthermore, the constructed three-dimensional energy distribution array is sequentially subjected to spatial resolution calibration and propagation path consistency verification. This includes spatial resolution calibration using a three-dimensional spatial interpolation mechanism and propagation path consistency verification using an energy propagation continuity detection mechanism, specifically implemented as follows:

[0121] Spatial resolution calibration using a three-dimensional spatial interpolation mechanism involves ensuring consistent spatial resolution between sampling points of each sensor based on the generated initial three-dimensional energy distribution array. The interpolated array is then used to verify the consistency of propagation paths. Specifically:

[0122] The sensor spacing of the initial three-dimensional energy distribution array is detected, and three-dimensional spatial interpolation is performed based on the detection results, specifically as follows:

[0123] For the generated initial 3D energy distribution array, sensor nodes in the array are detected. Three sensor locations are randomly selected, and the ratio between the actual spatial distance and the data network distance between any two locations is calculated. Spatial resolution consistency is verified based on the calculation results. Specifically:

[0124] If the ratio between the actual spatial distance between any two locations is consistent with the ratio between the data network distance, it indicates that the spatial resolution calibration of the initial three-dimensional energy distribution array is complete. Otherwise, it indicates that the spatial resolution calibration of the initial three-dimensional energy distribution array has failed, and the sensor array should be redeployed for data acquisition until the spatial resolution calibration is completed.

[0125] The consistency verification of propagation paths for a three-dimensional energy distribution array calibrated by spatial resolution is performed as follows:

[0126] The calibrated three-dimensional energy distribution array is divided into R propagation sub-regions. A region R0 is arbitrarily selected from these sub-regions as the initial region for propagation path verification. Simultaneously, the adjacent regions of the initial region are located, and their energy wavefront structures are determined. The energy wavefront structure corresponding to the initial region is also determined, and the energy attenuation gradient distance between the energy wavefront structures of the two adjacent regions is calculated. Based on the calculation results, propagation path consistency verification is performed, specifically as follows:

[0127] By setting an energy attenuation gradient threshold and verifying the consistency of the propagation path based on this threshold, we can conclude that...

[0128] If the calculated energy attenuation gradient threshold is less than the set energy attenuation gradient threshold, it means that the energy propagation path between the initial region and the connected region is continuous; otherwise, it means that the propagation path between the two regions is discontinuous, and the number of continuous propagation paths and the number of discontinuous propagation paths are recorded.

[0129] The region connected to the initial region is used as the initial region for the second verification (second-generation initial region). The energy decay gradient distance between the second-generation initial region and its neighboring regions is calculated again until the traversed neighboring regions are the initial region R0. Based on the statistical number of discontinuous propagation paths, the consistency of the propagation path is verified, specifically as follows:

[0130] Based on the number of discontinuous propagation paths and the total number of path consistency verifications, the discontinuous propagation path rate is calculated. If the calculated discontinuous propagation path exceeds 22.8%, it means that the propagation path verification of the current three-dimensional energy distribution array has failed. The sensor array is redeployed to collect data until the verification is passed. Conversely, if the calculated discontinuous propagation path verification of the current three-dimensional energy distribution array is passed, the previous three-dimensional energy distribution array is constructed as a mine blasting vibration propagation model.

[0131] It should be noted that the energy front structure is based on the gradient field at the peak of the energy spatial distribution in the divided region, and the direction is towards the direction of the fastest energy decay (the perpendicular direction of wavefront propagation).

[0132] The energy decay gradient distance is calculated by constructing matching points between two regions and calculating a weighted combination of the directional and magnitude differences of the corresponding gradient vectors. The result is the energy decay gradient distance between the two regions.

[0133] Furthermore, intelligent monitoring of mine blasting vibrations is carried out based on a defined mine blasting vibration propagation model, specifically as follows:

[0134] Based on historical blasting vibration characteristics, the energy sequence corresponding to the blasting vibration characteristics in the three-dimensional energy distribution array is captured and located. Based on the located energy sequence and the energy gradient attenuation distance in the mine blasting vibration propagation model, the energy gradient attenuation distance of the located energy sequence is calculated. Simultaneously, a safe energy level for blasting vibration is set, and intelligent monitoring of blasting vibration is performed based on this set safe energy level. Therefore,

[0135] Based on the energy sequence of the location, the energy corresponding to the location energy sequence is calculated, and the distance required for it to decay to a safe energy level is determined. The calculation result is the currently identified safe distance for blasting vibration characteristics. Intelligent monitoring of mine blasting vibrations is then performed based on this identified safe distance.

[0136] A safety warning distance is set based on the identified blasting vibration characteristics, and intelligent monitoring of blasting vibration is performed according to the set safety warning distance. Specifically:

[0137] If the energy attenuation distance corresponding to the monitored blasting vibration exceeds the set safety warning distance, it indicates that the currently monitored blasting vibration is an emergency vibration, and relevant personnel should be notified to take safety precautions.

[0138] If the energy attenuation distance corresponding to the monitored blasting vibration is lower than the set safety warning distance, it indicates that the currently monitored blasting vibration is a safe vibration, and the monitoring of mine blasting vibration will continue.

[0139] Furthermore, if the aforementioned function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0140] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0141] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0142] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent monitoring of vibrations during mine blasting, characterized in that: Includes the following steps, Vibration waveform data from the mine blasting area were collected, and the collected waveforms were optimized using an optimization iteration mechanism. Simultaneously, the accuracy of the optimized waveforms was verified using historical blasting vibration characteristics. Specifically: Initial three-dimensional vibration waveform data of the mine blasting area were collected using distributed vibration sensors. The data was divided into multiple monitoring time windows according to time series coordinates. The spectral energy concentration in each time window was calculated, and an initial reference window was determined. Based on the determined initial reference window, a multi-layer analysis mechanism was used for layer-by-layer analysis. At the same time, the waveform of the analysis results was optimized through an optimization iteration mechanism, and the accuracy of the optimized waveform was verified. In addition, monitoring of mine blasting vibrations based on identified mine blasting vibration characteristics, specifically: A three-dimensional energy distribution array is constructed based on the optimized waveform. The spatial resolution of the three-dimensional energy distribution array is calibrated and the consistency of the propagation path is verified in sequence. A mine vibration propagation model is constructed based on the three-dimensional energy distribution array that has passed the two verifications. Intelligent monitoring of mine blasting vibration is carried out based on the constructed mine vibration propagation model.

2. The intelligent monitoring method for mine blasting vibration as described in claim 1, characterized in that: The determination of the initial reference window is as follows: Based on the defined monitoring time windows, the spectral energy concentration for each monitoring time is calculated, and the calculation results are used as an effective indicator of the vibration signal. Therefore, Let K be the total number of frequency components corresponding to the divided monitoring time windows. Then, perform a Fourier transform on the three-dimensional vibration signal matrix within the current monitoring time window. Simultaneously, calculate the Fourier transform of the sum of all three-dimensional vibration signal matrices across all frequency components within the monitoring time window. Based on the ratio between these two values, determine the spectral energy concentration of each monitoring time window. Then, we have... The spectral energy concentration of each monitoring time window is calculated sequentially. After the spectral energy concentration of all monitoring time windows has been calculated, the monitoring time window with the largest spectral energy concentration is selected from all the calculation results as the initial reference window for identifying the vibration characteristics of mine blasting.

3. The intelligent monitoring method for mine blasting vibration as described in claim 2, characterized in that: The specific details of using a multi-layer analysis mechanism for layer-by-layer analysis are as follows: Given a defined initial parent time window, the first-generation analysis waveform is determined using a multi-feature time point comparison mechanism. Therefore, Set an energy spectral density threshold, and filter the first-level energy frequency band region in the initial parent time window according to the set energy spectral density threshold. Based on the determined first-level energy frequency band region, determine the spectral energy difference between the initial parent time window and the adjacent time window. According to the calculated spectral energy difference, set a spectral energy difference threshold. Determine whether to execute the frequency band boundary adjustment strategy according to the set spectral energy difference threshold. Collect waveform data after each execution of frequency band boundary adjustment. After all feature time points are compared, determine the first generation analysis waveform. Furthermore, the waveforms after each execution of the frequency band boundary adjustment strategy in the first-generation analysis waveform are subjected to feature enhancement processing, and the second-generation analysis waveform is constructed based on the results of the feature enhancement processing.

4. The intelligent monitoring method for mine blasting vibration as described in claim 3, characterized in that: The specific steps for filtering the first-level energy frequency band region within the initial parent time window based on the set energy spectral density threshold are as follows: Based on the spectral energy E corresponding to the frequency point index sequence number k in the initial parent time window. k (W P ), and sequentially compared with the set energy spectral density threshold E T By comparison, if the comparison results satisfy formula E as the frequency point index sequence progresses... k (W P )≥E T Then, it represents the spectrum of the initial parent time window at frequency k, which is a first-level energy frequency band region. Based on the screening formula of the first-level energy frequency band, all frequency points in the initial parent time window that satisfy the first-level energy frequency band formula are combined to form a first-level energy frequency band region.

5. The intelligent monitoring method for mine blasting vibration as described in claim 4, characterized in that: The specific steps for determining whether to execute the frequency band boundary adjustment strategy based on the set spectral energy difference threshold are as follows: A characteristic time point is randomly selected from the first-level energy frequency band region. At the same time, the same characteristic time point is selected from the adjacent time windows, and the spectral energy difference ΔP between the initial parent time window and the same time point in the adjacent time windows is compared. Set the spectral energy difference threshold ΔP T The decision to execute a frequency band boundary adjustment strategy is based on a set spectral energy difference threshold. If the calculated spectral energy difference, when compared with the set spectral energy difference threshold, satisfies the formula ΔP≤ΔP T If the selected feature time point has not undergone feature evolution in the adjacent time window, the frequency band boundary adjustment strategy will not be executed. If the calculated spectral energy difference, when compared with the set spectral energy difference threshold, satisfies the formula ΔP>ΔP T If the selected feature time point undergoes feature evolution in the adjacent time window, the frequency band boundary adjustment strategy is executed.

6. The intelligent monitoring method for mine blasting vibration as described in claim 5, characterized in that: The frequency band boundary adjustment strategy is as follows: By comparing the spectral density corresponding to the selected characteristic time point in the first-level energy frequency band region with the spectral density corresponding to the same time point in the adjacent time window, and implementing a frequency band boundary adjustment strategy based on the comparison results, we have: If the spectral density of the same time point selected in the first-level energy frequency band region within an adjacent time window is greater than the spectral density of the same time point selected in the first-level energy frequency band region, then the frequency band boundary is expanded outward; otherwise, the frequency band boundary is contracted inward.

7. The intelligent monitoring method for mine blasting vibration as described in claim 6, characterized in that: The second-generation analysis waveform constructed based on the result of feature enhancement processing is as follows: Let T be the duration of the waveform after each execution of the frequency band boundary adjustment strategy in the first-generation analysis waveform. C And according to the midpoint T of the duration c / 2, the first generation analysis waveform is divided into two parts, the head and the tail, and the positions of the two parts are interchanged. At the same time, the waveform data after the position is interchanged is spliced ​​together. The spliced ​​waveform data is the waveform of the first generation analysis waveform after time reversal. Based on the inverted waveform, feature enhancement is performed using enhancement coefficients. The set of all first-generation analysis waveforms after feature enhancement processing is the second-generation analysis waveform.

8. The intelligent monitoring method for mine blasting vibration as described in claim 7, characterized in that: The optimization of the waveform of the analysis result through the optimization iteration mechanism is as follows: The average energy envelope continuity of the first-generation analysis waveform is greater than the envelope smoothness index of the data within the initial parent time window. The feature discrimination of any segment of the waveform in the second-generation analysis is better than that of the first-generation analysis. If all the first-generation analysis waveforms can satisfy the iteration conditions for the second-generation analysis waveform, then the second-generation analysis waveform generated by iterating from the current first-generation analysis waveform will be used as the waveform for optimization.

9. The intelligent monitoring method for mine blasting vibration as described in claim 8, characterized in that: The accuracy verification of the optimized waveform is performed as follows: Based on the optimized waveform, the spectral energy concentration index, envelope smoothness index, and waveform cross-correlation residuals of each waveform segment are calculated, and a feature discrimination function f is constructed based on the results. D (J); The blasting vibration characteristics of three consecutive time points were extracted from historical blasting vibration waveform data, and a blasting vibration characteristic threshold f was set. D (T), based on the set blast vibration characteristic threshold, the waveform accuracy is verified, then we have, The blasting vibration features extracted from three consecutive time feature points are input into the constructed feature discrimination function. If the function output for each blasting vibration feature point is less than the set blasting vibration feature threshold, the accuracy verification of the optimized waveform is passed. Otherwise, the accuracy verification of the optimized waveform is not passed. The initial three-dimensional vibration waveform data of the mine blasting area is re-acquired and the waveform is optimized until the accuracy verification of the optimized waveform is completed.

Images