Low-vibration multi-point delay digital twin drive intelligent electronic detonator blasting control method

Abstract

The invention discloses a low-vibration multi-point delay digital twin drive intelligent electronic detonator blasting control method, and relates to the technical field of blasting vibration control, and the method comprises the steps: 1, constructing a frequency weight curve and a weighted energy index, and unifying a time scale and three-component measurement; 2, establishing a virtual field by using a single-hole signature library, instrument transfer function deconvolution and a boundary correction propagation operator; step 3, performing delay optimization in combination with a timing characteristic function and a partition phase, and setting a keyhole read-only and weighted threshold; fourthly, programming backward reading is conducted, gating is conducted through a consistency index and an admission ratio, only one-time fine adjustment is conducted on the non-key hole on line according to an expected energy track and deviation statistics, and block assimilation updating is conducted later. According to the scheme, quantitative suppression of sensitive frequency band energy is achieved, and construction efficiency and traceability are kept. Through unified parameter and data object through identification, database building, optimization, execution and feedback, the scheme consistency is improved, the trial explosion frequency is reduced, supervision and acceptance are facilitated, and redisk tracing is supported.

Description

Technical Field

[0001] The present invention relates to the technical field of blasting vibration control, and specifically to a low-vibration multi-point delay digital twin driven intelligent electronic detonator blasting control method. Background Art

[0002] For controlled blasting scenarios in urban tunnels and underground caverns close to existing buildings or linings, construction needs to balance crushing efficiency and structural safety within a tight schedule. Supervision is based on structural response limit constraints, and traffic pipelines and equipment operation make working conditions changeable.

[0003] Existing practices generally rely on velocity peak limits and empirical delay tables at control points. Although standards emphasize frequency-based building response assessment, engineering practices still primarily focus on single peaks, ignoring the selective amplification of narrowband frequencies by structures. Traditional signature holes often directly time-superimpose single-hole waveforms without removing the influence of instrument transmission and complex boundaries. Although electronic detonators offer high-precision timing and readback, they are not systematically used for spectrum shaping. Sensor coupling, range saturation, and inconsistent clocks at the acquisition end within the monitoring link also make horizontal data comparison difficult. To reduce costs, numerical models often simplify free-surface reflections, lining interfaces, and cavern shapes using equivalent attenuation. This results in frequency-domain predictions being insensitive to delay refinement, a lack of lightweight virtual sites, and a lack of unified interfaces such as reusable frequency weight curves and expected energy trajectories in existing processes. Consequently, there are long-standing breakpoints between design, execution, and review.

[0004] Based on this, the core technical problem that needs to be solved is to quantify the sensitivity of the protected building or lining to a specific frequency band under complex boundaries and non-stationary working conditions into a unified control quantity that can directly drive the delay design and online judgment, and to maintain consistent time scales, consistent indicators and traceable parameters in each link of single-hole testing, group-hole prediction, delay scheduling, detonation execution and post-feedback, so that the energy of the sensitive frequency band can be reliably suppressed even under the worst working conditions. This problem arises in scenarios where urban tunnels are close to operating facilities, historical buildings or thin-walled linings. The causes are the lack of unified clock and attitude correction in monitoring, the failure to deconvolute single-hole signatures, the failure to include boundary reflections and modal conversion in propagation modeling, the failure to characterize the timing dispersion of electronic detonators in the frequency domain, and the lack of weighted thresholds and access criteria in the delay scheme. In structures with significant frequency sensitivity, this will cause the phase organization of the group of holes to deviate from the design, the superposition of narrowband energy toward the protected object, and the lag of online judgment due to the lack of expected trajectories and triggering criteria. The consequences are manifested in the expansion of component cracks, loosening of lining interfaces and decreased stability of surrounding rock, or excessive conservatism to avoid risks, resulting in insufficient crushing, reduced propulsion efficiency and increased costs. Summary of the Invention

[0005] (1) Technical problems solved To address the shortcomings of existing technologies, the present invention provides a low-vibration, multi-point delay digital twin-driven intelligent electronic detonator blasting control method. By constructing a frequency weight curve and weighted energy index, the method unifies time scales and three-component measurements. A virtual site is established using a single-hole signature library, instrument transfer function deconvolution, and a boundary-corrected propagation operator. Delay optimization is combined with timing characteristic functions and partitioned phases to set read-only and weighted thresholds for critical holes. Programming readback is gated using consistency indexes and access ratios. Online fine-tuning is performed only once for non-critical holes based on the expected energy trajectory and deviation statistics, followed by block-by-block assimilation updates. This method achieves quantitative energy suppression in sensitive frequency bands while maintaining construction efficiency and traceability, thus resolving the technical issues discussed in the background art.

[0006] (2) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: A low-vibration, multi-point delay digital twin-driven intelligent electronic detonator blasting control method includes deploying three-component sensors to collect responses under a unified clock reference, restoring them to a structural coordinate system based on a direction mapping matrix and a range normalization matrix, calculating the group power spectrum and modal dominance, and constructing a frequency weighting curve and weighted energy index. The three-component complex spectrum of a single hole is deconvolved on the unified clock reference by combining the instrument transfer function with the path transfer operator to obtain the source spectrum signature. The boundary correction propagation operator and the detonator timing characteristic function are introduced to form the group hole prediction complex spectrum and energy trajectory. A weighted robust target is constructed using the frequency weight curve and the timing characteristic function, a delay vector is established under the constraints of the operation time window and the detonation sequence, a delay list is obtained by updating the projected gradient, and a phase difference and a key hole redundancy window are set according to the spatial partition; The delay list is written into the electronic detonator, the clock offset is estimated according to the unified clock reference, and the consistency index and the admission ratio are calculated; the weighted coherence is used to trigger a one-time fine-tuning of only the non-critical hole online; and after detonation, the source spectrum signature, path transfer operator, boundary parameters, instrument transfer function and frequency weight curve are updated by block assimilation.

[0007] Furthermore, a unified clock reference is established, and the three-component response is restored to the structural coordinate system in combination with the installation attitude calibration matrix and range normalization. The group power spectrum is calculated according to the multi-window STFT, and the saturation and coupling degradation channels are eliminated using the coherence function to form the in-service frequency response baseline. The measurement point coordinates and component properties are written into the metadata for unified archiving.

[0008] Furthermore, the output of the measured SSI-Cov is used to identify the modal shape, poles and damping, construct the participation coefficient and the through term, and calculate the modal dominance spectrum. This is then coupled with the frequency response function amplitude according to the power exponent to form a frequency weight curve, and the weighted energy index and evaluation bandwidth are defined using the three-component Minkowski composite index and the penalty power exponent.

[0009] Furthermore, the single-hole complex spectrum is subjected to Tikhonov regularized deconvolution under the instrument transfer function and the path transfer operator, wherein the instrument transfer function is a three-component transfer matrix; The path transfer operator is composed of direction cosine matrix, distance power attenuation, phase delay and quality factor attenuation, and a boundary correction propagation operator is used to introduce complex reflection coefficient, round-trip propagation delay and mode conversion matrix.

[0010] Furthermore, the dispersion is characterized by the timing characteristic function of the electronic detonator, the planned timing addresses the hole index set hole by hole with a unified clock reference, the uniform jitter half interval is dynamically updated by readback statistics, and is superimposed with the boundary correction propagation operator to obtain the phase consistency factor of the complex spectrum for group hole prediction.

[0011] Furthermore, a robust target with a weighted energy upper bound is constructed using the most unfavorable phase disturbance, and the uncertainty set is defined as the absolute value of the delay disturbance of each hole not exceeding a preset upper limit; under the constraint of the operation time window and the difference between the detonation sequence, the projection-gradient is used to update the delay vector. The projection is first cropped to the time window hole by hole, and then the difference constraint is alternately iterated by Euclidean projection until it no longer changes. The resulting delay list is used for programming.

[0012] Furthermore, the hole groups are clustered on the directional map according to azimuth, distance and geological attributes to obtain a partition set; the target phase difference of adjacent partitions is set around the sensitive center frequency of the frequency weight curve, the partition delay offset is limited to the respective operation time window, and a read-only redundant window and readback strategy are configured for the key holes.

[0013] Furthermore, the delay list is written into the electronic detonator and read back hole by hole, the global clock offset is estimated based on the difference between the read-back delay and the programmed delay, and the hole-by-hole consistency index is calculated according to the normalized time error, and compared with the preset regulatory tolerance threshold for determination; The admission ratio of the weighted energy index to the threshold is calculated at each measuring point and the most unfavorable value of the entire field is taken. If any condition is not met, the system falls back to adjusting the phase difference and sequence constraints.

[0014] Furthermore, the deviation statistics are constructed by the weighted complex correlation between the sliding short-time spectrum and the virtual field prediction spectrum. When the maximum deviation across the measurement points exceeds the trigger threshold and there is sufficient remaining time slot, the non-critical hole is updated with a one-time least squares update according to the first-order sensitivity of the spectrum to the delay. The truncation operator limits the correction amount to the regulatory tolerance interval, the critical hole is not updated, and all operations and timestamps are written into the event log.

[0015] Furthermore, the post-assimilation includes: taking the weighted spectral domain loss as the objective function and introducing prior regularization such as smoothness, sparsity or low rank, and updating in blocks according to the fixed order of boundary parameters and propagation operator → source spectrum signature → instrument propagation function → frequency weight curve; the update amount is obtained by the projection of the error on the Jacobian of each block, and is uniformly addressed and stored with the hole index and measurement point index set, and the parameter increment is constrained by the regularization weight.

[0016] (3) Beneficial effects The present invention provides a low-vibration, multi-point delay digital twin-driven intelligent electronic detonator blasting control method, which has the following beneficial effects: Through the frequency weight curve and weighted energy index, the risk based on frequency is converted into programmable control quantity, forming a unified interface for delay design from identification to execution, avoiding misjudgment caused by only looking at the peak value, and significantly enhancing the targeted constraint and verifiability of narrowband resonance.

[0017] With the single-hole signature library and virtual site as the core, combined with the source-channel decoupling mechanism of boundary correction propagation operator and instrument transmission function separation, a compact representation of the free surface and lined cavern reflection is achieved, making the group hole prediction and actual measurement closer to the site, reducing repeated test blasting and construction uncertainty.

[0018] The timing characteristic function is embedded in the frequency domain superposition of the group holes and coordinated with the frequency weight curve. Combined with the spatial arrangement of partition mapping and partition phase, the energy of the sensitive frequency band close to the protected object is actively thinned, while the insensitive direction remains favorable for fragmentation, realizing frequency and space dual-domain shaping.

[0019] By adopting the delay optimization with robust objectives as the core and supplemented by weighted frequency band thresholds, the most adverse effects of electronic detonator dispersion and link micro-mismatch are pre-modeled, and programmable timing is output under the constraints of operation window and detonation sequence, so as to maintain the safety margin of sensitive frequency band under the worst working conditions.

[0020] Before detonation, through programmed readback and unified clock reference correction, dual threshold gating of consistency index and admission ratio is set to complete consistency verification of time domain executable and frequency domain suppression target before triggering, timely eliminate implicit mismatches that will destroy coherent superposition and ensure full chain traceability.

[0021] During the online detonation phase, the expected energy trajectory and deviation statistics are used to trigger one-time online fine-tuning of non-critical holes only and leave traces throughout the entire chain. Critical holes are kept read-only and safe through redundant windows. After detonation, the parameters are updated by block assimilation to promote model self-learning, so that the weight curve and the virtual scene continue to approach reality and generate synergy in the next round of design. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of the digital twin-driven intelligent electronic detonator blasting control method of the present invention. DETAILED DESCRIPTION

[0023] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0024] See also Figure 1 The present invention provides a low-vibration multi-point delay digital twin driven intelligent electronic detonator blasting control method, including: Step 1: Based on the unified time scale and three-component response, establish the in-service frequency response baseline of the building-lining system to provide a repeatable and traceable data base for sensitive frequency band identification.

[0025] The protected object has a strong frequency selectivity in its response to the excitation. or peak acceleration As a single indicator, it is easy to ignore the narrowband amplification of weak modes. Therefore, it is necessary to first unify the clock reference. Under the main axis of the building-lining structure, three-component sensors are arranged and three-component vibration signals are collected. , and then through the direction mapping matrix and the range normalization matrix Restore the three-component response to the structural coordinate system , thus forming a unified entrance for subsequent frequency domain analysis.

[0026] To avoid systematic distortion caused by independent clocks, assembly deviations, and range differences at multiple acquisition terminals, an absolute time scale is first established on the trigger link, and all channels are aligned through a one-time time calibration. Direction correction is then performed using the installation attitude calibration matrix, and the range and sensitivity differences of each channel are normalized to the structural coordinate system response:

[0027] Among them: The three-component response of the structural coordinate system :As the only input for frequency domain identification and subsequent weight construction, it is a three-dimensional real vector space ; Direction Mapping Matrix : Map the sensor coordinate system to the structure coordinate system, which is a three-dimensional special orthogonal group The set of orthogonal matrices of ; that is, orthogonal and with determinant of A real matrix representing a pure rotation around the origin (without mirroring); Range normalization matrix : Linear normalization is performed based on channel sensitivity and upper range limit, which is , where the channel gain ; Sensor coordinate system three-component response : The original measurement signal is a three-dimensional real vector space ; Clock deviation :Eliminate the time drift of asynchronous acquisition, Coordinated with electronic detonator time resolution.

[0028] When used, attitude errors, range differences, and clock drift are eliminated all at once within the same formula, achieving seamless restoration from the signal domain to the structural domain. This ensures that subsequent frequency domain operations only depict the structure-path-boundary without mixing in equipment factors. Ultimately, a unified baseline for comparability is formed across multiple measurement points and multiple working conditions. To identify the structural consistency and boundary influence between measurement points, the coherence function is further calculated to screen out poorly coupled or saturated hidden channels:

[0029] Among them: coherence function :Measure the frequency of two measurement points The linear correlation at ; Cross spectral density :Characterization channel With channel The frequency domain correlation of ; Autospectral density : Characterizes the energy distribution of each channel, which is a real number domain .

[0030] When in use, coherent screening is used to eliminate pseudo-peak channels caused by coupling degradation, loose cover or cables; thus, subsequent modal identification is based on a clean channel set; ultimately, the stability and reproducibility of sensitive frequency band extraction are improved.

[0031] Without disturbing the structure and construction organization, the dominant modes are extracted based on the output-only frequency domain decomposition by utilizing in-service micro-vibration or slight excitation. The concentration of frequency domain energy on the group response matrix is ​​used as a quantifiable indicator of the weak mode, where:

[0032] Where: Group power spectral density matrix : Aggregate frequency domain energy and phase information of multiple measurement points, which is a real number domain ; Eigenvector matrix : Give frequency The main direction basis at is a set of orthogonal matrices; the eigenvalue diagonal matrix : gives the energy distribution intensity, which is , ; Principal singular / eigenvalue : Characterizes the dominant mode energy at this frequency, which is ; Modal dominance spectrum : Normalized main mode dominance index, .

[0033] When used, the modal dominance that is independent of the input is obtained through output-only decomposition, thereby avoiding the contamination of identification by source term uncertainty; forming a clear indication of weak modes on multi-measurement point data; and ultimately providing an objective basis for the weight allocation of sensitive frequency bands.

[0034] Combine modal dominance and frequency response into a frequency weighted curve , and define weighted energy indicators for delay optimization and digital twins , forming a parameter-level interface with the signature hole database.

[0035] It is not enough to drive the control quantity only where the mode is strong, but also to quantitatively emphasize these frequency bands and seamlessly couple them with the spectrum response of the subsequent hole group superposition. Therefore, the building-lining combination frequency response function needs to be Amplification characteristics and modal dominance spectrum of Combined with the above, a weighted expression that reflects the weak mode and can be reused across working conditions is constructed; and a unified weighted energy index is used Convert the pros and cons of different delay designs into comparable scalar targets.

[0036] The identified modal frequencies are set The kernel function is used to smooth the neighborhood, and then multiplied by the normalized frequency response amplitude to obtain the continuous weight that highlights the sensitive frequency band and suppresses the non-sensitive frequency band. ,in:

[0037] Where: Frequency weight curve : The core weight of subsequent delay optimization and prediction evaluation, ;frequency : independent variable, domain ; To evaluate the upper and lower limits of bandwidth; number of seeds : The number of weighted seed frequencies is equal to the number of sensitive modes / frequency bands to be focused on; normalization coefficient :ensure ,for ; Seed frequency : No. Sensitive frequency centers, from Or the local maximum value extraction of the modal dominance spectrum needs to fall on ; Modal weight coefficient : According to the modal dominance spectrum The weight assigned to the component importance, and ; Kernel Function :by is the smoothing kernel of the independent variable, requiring 、Follow Monotonically decaying, dimensionless. Default Gaussian kernel: ; Laplace kernel can also be used or Epanechnikov kernel; exponential parameter : Control the prominence of the modal clustering area, ; Frequency response function : The amplitude-frequency amplification characteristics of the building-lining combination are ; The identified structure-measurement point combination frequency response (output - only identification or complex frequency domain fitting); In this formula, only its amplitude is used and Normalized, dimensionless; amplitude power exponent : Adjust the intensity of the influence of the amplification effect on the weight, .

[0038] In the unified time scale Under the 3D model, collect three-component responses at key locations of floor / lining-surrounding rock. ; The group spectrum matrix is ​​obtained by multi-window STFT ; Extract modes using SSI-Cov and damping; construction:

[0039] Among them, the modal shape and participation coefficient Obtained by identification regression, the direct term Slight regularization; then normalization participate structure. For the modal poles (complex, including damping), is the output-participation vector, is the high frequency direct pass, high frequency residual / rigid direct pass term, absorbing the unmodeled high frequency modes and direct paths. , the number of modes retained; Give Initialization rules: and Recommendations on value ranges based on component importance and DIN / BS building categories; Adaptive with modal density (e.g. Start, follow Increment to ). Output discrete weight vector (same as FFT frequency) and metadata ( ).

[0040] When in use, the multiplicative coupling of the modal-amplification factors causes the weight to form a steep energy penalty in the weak frequency band; thereby quantifying the risk by frequency and making it programmable; ultimately, a control curve with stable transferability between different working conditions is obtained, which directly drives the spectrum shaping of the subsequent delay scheme.

[0041] In order to connect the weight curve with the hole group prediction, the three-component synthetic spectrum and weighted energy index are introduced as a unified measure of the quality of the delay scheme, and form a coupling interface with the spectrum output of the signature hole prediction, where:

[0042] Among them: three-component autospectral density :Power spectral density of the three components of the structural coordinate system, Components at frequencies The non-negative scalar spectral amplitude at is ; Synthetic spectrum amplitude : The three-component energy is synthesized into a single spectrum using the generalized Minkowski norm, which is ; Composite index : Adjust the focus to peak or balanced. .

[0043]

[0044] Among them: Weighted energy index : The scalar target of the delay scheme, the smaller the value, the better, ; Frequency lower limit , frequency cap : Evaluate bandwidth boundaries, 、 ; Power index : The degree of penalty for highlighting narrow-band peaks, .

[0045] When used, the synthesized spectrum amplitude losslessly compresses the three-component information into a unified amplitude metric in the spectral domain; thus, the weighted energy index can be directly used as the target and constraint of delay optimization; and finally the output spectrum predicted by the signature hole-group hole is Implement a plug-and-play parameter-level interface to ensure consistency across steps.

[0046] In terms of sensor installation and coupling, the same type of three-component accelerometers are used for the top floor, foundation, and side facing the explosion source of the building. A dual insurance method of mechanical fixing and coupling agent is used to ensure that the coupling stiffness is not less than an order of magnitude lower than the estimated local stiffness of the structure; the cables are separated and isolated from strong electricity, and stress relief rings are set at the inflection points to avoid false triggering caused by pulling; the range selection uses non-saturation as the red line, combined with the estimated maximum velocity-acceleration-displacement three-domain conversion, giving priority to wide dynamic range sensors and calibrating the preamplifier margin. The time base is based on a unified clock base. As the root, through GPS / Beidou timing or high-stability local oscillator synchronization, triggering is hard-wired, and event logs and trigger timestamps are recorded for pre-readback and post-traceability. Sensitive components such as floor slabs, columns / vaults, and linings are equipped with response points to improve the resolution of the component as a whole. Their coordinates and component properties are written into the metadata together to ensure that the modal dominance spectrum Calculation and frequency weight curve The modal weight coefficient can be updated according to the importance of the component .

[0047] Step 2: Deconvolute the three-component measured waveform of a single hole on a unified time scale to restore it to the source spectrum signature, and use the boundary-path joint correction to obtain a transferable propagation operator to provide a stable primitive for the superposition of group holes.

[0048] Although single-hole blast sources are nearly reproducible, under varying hole locations and geological conditions, the measured three-component spectrum simultaneously reflects the combined effects of source characteristics, propagation attenuation, boundary scattering, and instrument transfer. Directly using the original spectrum as the basis for cluster superposition inevitably introduces the randomness of equipment and layout into the delay design.

[0049] Therefore, the single hole response needs to be based on a unified clock reference Next, the letter is transmitted through the instrument and path transfer operator Joint deconvolution to extract source spectrum signature , and the boundary influence is represented by the boundary set operator Write the propagation end instead of the source end, so that the same charging system can be compared at different locations. Since the frequency weight curve has been output in step 1 , the source-channel separation method is used to ensure that the subsequent weighted evaluation and scheme selection are targeted at the structurally weak frequency bands without being confused by instruments or accidental paths.

[0050] After a single hole is detonated, the three-component spectra of multiple measuring points are inverted in groups, and the Tikhonov type robust regularization is used to suppress the pathological amplification of weak frequency bands; by converging the constraints of multiple measuring points into the same source spectrum signature, the device-path decoupling of the source is achieved. With the measuring point index After establishing the unified set, the source spectrum signature is calculated as follows:

[0051] Where: Source spectrum signature : Single hole The three-component complex spectrum primitives in the structural coordinate system are solved independently by component and frequency, as follows: ; Sum index : The set of all measurement points involved in the inversion (including at least the top layer, foundation and three-component measurement points facing the explosion source); observation complex spectrum :Measuring point In the hole The three-component complex spectrum at the time of detonation, ; Obtained by FFT or STFT from synchronous recording; Path transfer operator : Source to measurement point The three-component propagation-geometry-attenuation-dispersion operator, ; It can be constructed as direction cosine matrix × distance power attenuation × phase delay × quality factor attenuation; Instrument transmission letter :Measuring point Frequency response of the sensor-front-end-acquisition chain, transfer matrix of the three-component sensor-acquisition chain, , manufacturer calibration curve + on-site frequency sweep micro-excitation; the main diagonal is the amplitude-phase frequency response, and the cross axis does not exceed 5% of the small coupling; Regularization weight : Suppress the inversion instability of the pathological frequency point, which is , Tikhonov regularization weight It is recommended to use L-curve or generalized cross validation (GCV) for automatic selection; Smooth operator :right The frequency smoothing prior of is like the first-order difference, which is a linear operator isomorphic to the frequency dimension; for single-frequency solutions, the default is (Ridge regression); if you want to smooth across frequencies, you can combine multiple frequency points and let is the frequency-indexed first-order difference block tridiagonal operator; conjugate transpose : Least squares closed-form term under energy consistency.

[0052] Boundary Set Operator Parameters A two-step calibration is given: using an early window to identify the direct field and lock The geometric attenuation of the late window is used to fit the main lobe of the first reflection, limiting No more than 3 for robustness.

[0053] As additional content:

[0054]

[0055] Propagation direction unit vector :From the source point coordinates With the coordinates of the measuring point get, ; dimensionless. Coordinates are given by surveying or design model; directional polarization matrix : Project the three bases of the main motion direction (the two directions orthogonal to the propagation direction) to the direction cosine matrix of the three-component coordinate system of the measuring point. Construction method: Let the local basis of the measuring point be (such as radial / tangential / vertical or X / Y / Z), first define:

[0056] is the global vertical unit vector; if , replace it with another non-collinear reference axis. Let and ,but:

[0057] Each element is a direction cosine, and the unit is dimensionless. This construction ensures Orthogonal reversible; if you need to distinguish the relative weight of P / S, you can Right multiplication of diagonal matrix , the default value is 1. Distance : Unit: m; calculated directly from geometric coordinates. Geometric attenuation : dimensionless; . Common values: Approximate spherical waves, Approximate cylindrical wave. Calibration method: linear regression of the amplitude logarithm of the early direct window of the signature hole-distance, the slope is given range of experience.

[0058] Wave speed : Unit: m / s. Non-dispersive approximation takes constant ; Weak dispersion approximation is available , . Obtained by arrival time inversion of ultrasonic / seismic lines or signature holes (P waves are common , confirmed by survey results). Quality factor : dimensionless, describes inelastic dissipation; commonly used in engineering Calibration method: Fitting the amplitude ratio of the same frequency at different distances get . Can also be given in separate bands .frequency : Unit Hz; value is within the evaluation bandwidth , imaginary unit :satisfy .

[0059] When in use, source-channel separation is achieved through multi-measurement point coupled inversion, so that the source spectrum signature does not drift with the equipment and layout; thus, smooth and reusable source characteristics can still be obtained in the weak signal-noise frequency band; ultimately, a convergent and stable primitive that can be migrated across working points is provided for the group hole superposition.

[0060] Considering the reflection, transmission and shear-compression mode conversion effects of the free surface, lining-rock interface and cavern boundary, they are uniformly mapped into the boundary set operator of the propagation end. , in order to avoid misclassifying boundary randomness as the source spectrum signature. Specifically, the propagation operator is multiplicatively corrected in the frequency domain:

[0061] Among them: Boundary correction propagation operator :The phase-amplitude effect of boundary scattering-reflection is incorporated into the propagation end, which is ; Identity matrix : Directly reach the field base, no value interval; boundary number : The number of significant boundaries participating in a scattering path, Complex reflection coefficient :boundary The frequency-dependent reflection / transmission weight of can include the impedance mismatch and incident angle effects, which is , ; Propagation delay : By border The additional round trip delay is ; Modal conversion matrix :The energy distribution and direction transformation between P wave / SV wave / SH wave is .

[0062] When used, the boundary effects are imaged and superimposed on the propagation end to achieve a compact representation of complex geometry; thereby improving the geometric adaptability of the prediction without increasing the complexity of the source spectrum signature; and ultimately providing a unified propagation model for the group hole response under different boundary conditions.

[0063] The source spectrum signature is superimposed with the electronic detonator timing characteristic function through a boundary correction propagation operator in phase-amplitude consistency, and the group hole prediction interface and expected energy trajectory oriented to the frequency weight curve are output.

[0064] The core of the group-hole response lies in phase organization. The high timing accuracy of electronic detonators provides a strict phase reference for frequency-domain superposition, but slight dispersion still exists. If the timing dispersion is not explicitly introduced in the frequency domain to weaken the phase consistency, the design of the pressure energy in the narrow sensitive frequency band will be biased.

[0065] Therefore, the source spectrum signature Propagation operator through boundary correction Mapped to each measurement point and used as a timing characteristic function The electronic detonator dispersion is reduced in the spectral domain and then compared with the frequency weight curve. Only by using coupling as a unified indicator can we achieve comparability across design schemes and optimization for sensitive frequency bands.

[0066] Group hole collection Each hole in the Around a unified clock reference The slight dispersion of the electronic detonator is realized as a zero-mean random delay, and the consistency of the characteristic function is superimposed in the frequency domain attenuation.

[0067] Measuring point The three-component complex spectrum of the hole group prediction is as follows:

[0068] Among them: The three-component complex spectrum of the group hole prediction :Measuring point The frequency domain prediction output is ; Boundary Correction Propagation Operator : The propagation map including boundary scattering is ; Instrument transmission letter :Measuring point The measurement link frequency response is ; Source spectrum signature :hole The three-component source spectrum of ; Scheduled :hole The programmed detonation time relative to the unified clock reference is ; Timing characteristic function : The frequency domain consistency factor of the electronic detonator dispersion is defined as ,for , which does not increase monotonically with frequency; or, gives a uniform jitter within the manufacturer's tolerance Closed form when:

[0069] in is the half-interval tolerance (dynamically updated by the readback statistics ), dispersion :hole The random delay error is Limited by device specifications.

[0070] When used, the additive expression explicitly separates the effects of planned timing and dispersion, thereby preserving the dominance of delay design in the phase domain. This in turn allows for a reasonable reduction in band consistency at high frequencies, avoiding overly optimistic peak cancellation estimates. Ultimately, it provides an interpretable and sensitivity-analyzable frequency-domain framework for different delay schemes.

[0071] is the composite spectrum amplitude with step 1 With weighted energy index Seamlessly connect, calculate the generalized Minkowski composite spectrum at each measurement point and perform weighted integration using the frequency weight curve; at the same time, perform inverse Fourier transform on the square root of the weighted spectrum to obtain the expected energy trajectory for online comparison in step 4, where:

[0072] Among them: weighted energy index of measuring point :Measuring point The weighted evaluation quantity, ; Frequency lower limit , frequency cap : Evaluate bandwidth boundaries, 、 ;Frequency weight curve : Weight function of sensitive frequency band, ; Complex spectral components :Measuring point No. The component complex spectrum is ; Composite index : Control the focusing of the three-component synthesis, ; Power index : The degree of penalty for highlighting narrow-band peaks, .

[0073]

[0074] Where: expected energy trajectory :Measuring point The time domain reference trajectory is used for online comparison and threshold discrimination, which is ; Inverse Fourier operator : Frequency domain to time domain transformation; frequency weight curve :Elevate the weak frequency band to the power trajectory observable in the time domain, .

[0075] When used, through weighted synthesis and integration, the predicted amount and the evaluation amount in step 1 are isomorphic at the parameter level; thus, the delay scheme can be Perform sorting and constraint checking; finally, the expected energy trajectory Convert frequency domain risks into time domain executable signals to provide a gold standard for real-time verification in step four.

[0076] To ensure the source spectrum signature and boundary-corrected propagation operator Traceability, strict reuse of unified clock reference in single hole test , and record the trigger timestamp and event log before and after detonation; the instrument link uses the readable range and sensitivity calibration, and explicitly writes the instrument transmission function parameter; The path transfer operator is established on the path side using the borehole coordinates, hole depth, charge column section and lithology layering The inversion only modifies a few aggregation parameters to avoid overfitting; the boundary number is determined by the cave geometry and lining parameters on the boundary side. and the modal conversion matrix The complex reflection coefficient is then calibrated by the test waveform and propagation delay ; Timing side uses the planned timing of the programmed readback Combining with device self-test statistics to form timing characteristic function parameter library.

[0077] All objects are unified as a hole index collection and the measurement point index collection Perform addressing to ensure that delay optimization can be performed directly by hole-partition in step three.

[0078] Constructing a set of delay decision variables for sensitive frequency bands Together with the robust constraint system, the weighted energy of the sensitive frequency band is minimized and programmable redundancy is reserved for the execution link under the premise of meeting the operation window and detonation sequence, so that the delay design revolves around the sensitive frequency band and has a provable safety margin for electronic detonator dispersion and link mismatch.

[0079] A delay optimization model is established with the goal of minimizing the weighted sensitive frequency bands and is robust to timing dispersion and boundary uncertainty. A solution process that can be directly connected with on-site constraints is given.

[0080] The delay variable only works in the phase domain, but the frequency band sensitivity is determined by the frequency weighting curve The coherence of the hole cluster is determined by the uncertainty, and the coherence of the hole cluster will be reduced by the small dispersion of the electronic detonator and the occasional mismatch in the communication. If the objective function does not explicitly model these uncertainties, the ideal cancellation of the narrowband will be overestimated. Therefore, the frequency domain superposition expression of the hole cluster in step 2 is coupled with the timing characteristic function to construct a robust objective with upper bound significance. Then, the solution is obtained under the constraints of the operation window and process priority. This makes the intention of pressure-sensitive frequency band energy executable, verifiable, and traceable within the actual boundaries.

[0081] Predicting three-component complex spectra in a cluster of holes On this basis, the delay disturbance caused by the dispersion of electronic detonators is taken as the uncertainty, and the upper bound target is defined by the most unfavorable phase disturbance, so that the delay scheme has the ability to resist deviation in narrow-band sensitive areas. The uncertain set , construct a robust objective:

[0082] Among them: Robust target : The upper bound of the weighted energy under the most unfavorable dispersion is ; Delay disturbance :hole Random offset of is used to characterize the dispersion and link micro-mismatch, Upper limit of dispersion : The upper bound of the time uncertainty given by the device and the communication link, ;Measurement point index : Enumeration of monitoring locations on the protected object, no value range; frequency lower limit , frequency cap :Evaluation bandwidth boundary, 、 ;Frequency weight curve : The weight function that highlights the sensitive frequency band is ; Complex spectral components :Measuring point No. The frequency domain prediction output of the component is ; Composite index and power index : The functions are three-component synthesis and narrowband penalty intensity, 、 .

[0083] Replace with the upper bound of the worst phase perturbation: , the upper bound of its amplitude is 1, but the upper bound of energy after superposition is available Make conservative corrections: The phase coherence factor of each hole in is multiplied by Get the upper bound spectrum and then calculate the robust target , target phase difference : Inverted / semi-inverted target at the center of the sensitive frequency band; When used, the above bounded target replaces the nominal condition minimization, thereby having a priori tolerance for the coherence degradation caused by dispersion in the sensitive frequency band; thereby avoiding overfitting to the ideal cancellation; and ultimately making the delay scheme still maintain a low robustness target within the realistic error envelope. spectral suppression effect.

[0084] To ensure the feasibility of the solution, a unified operation window and detonation sequence map are introduced, and gradient descent with projection is used to find the optimal solution within the feasible domain. The decision variable projection update formula is defined as:

[0085] Where: Iteration Index : Optimize iteration count; projection operator : An operator that projects the candidate delay vector onto the feasible domain, which consists of the job time window constraint and the sequence difference constraint; Job window lower limit , upper limit : The allowed programming time range, ; Constraint set : A set of detonation sequence and safety isolation; step length : Update amplitude control, .

[0086] The projection operator is a one-step rule for adjusting candidate delay vectors to the feasible domain. Given a candidate delay for each borehole and two hard constraints—one that each borehole must fall within the upper and lower bounds of the operational time window; and two that a directed graph defines the order and minimum interval (one borehole must be at least a certain number of milliseconds later than another). The output of the projection is a set of delays that are closest to the original vector while satisfying both constraints. Implementation involves first performing component-level pruning to ensure that each borehole falls within its own time window. Then, a Euclidean projection is performed on the difference constraint (which can be understood as evenly distributing the adjustment amount along the violating edges until all differences are within the specified interval). These two steps are iterated alternately until no further changes occur. The resulting delay list is neither window-crossing nor out-of-order, and can be used directly as a feasible step in delay optimization or as a validation step after online fine-tuning.

[0087] At the same time, the edge set of the directed sequence relationship graph To impose an order constraint:

[0088] Among them: the edge set of the preceding and following relationship : A set of directed edges that specifies the detonation sequence; minimum safe interval : The minimum distance under process and safety requirements, ; Decision delay :hole The programming detonation moment, .

[0089] When used, the projection-gradient framework is used to reduce the robust target while satisfying the window and order constraints. ; This avoids the ineffective detour of first satisfying the spectrum and then repairing the feasibility; and ultimately obtains a programmable and readable delay sequence through limited iterations, which is directly transmitted to the electronic detonator programming list.

[0090] The hole group is partitioned and phase-arranged from the perspective of space-frequency coupling, and frequency band energy thinning is implemented for the unfavorable radiation direction close to the protected object. Redundant windows and readback strategies are injected into the key holes to ensure that they do not cross the boundary even under the worst working conditions.

[0091] The nominally optimal time sequence may not be uniform in space, especially the hole clusters close to the protected object will form directional superposition peaks in the sensitive frequency band. Therefore, it is necessary to use spatial partitioning As the skeleton, in the center of the sensitive frequency band The phases between the nearby partitions are intentionally reversed and peak-shifted, so that the synthetic spectrum in the unfavorable direction is actively thinned; at the same time, the key hole set is considered Due to its high importance and irreplaceability, a redundant time window and a readback-logging strategy that are not affected by online fine-tuning are configured for it to achieve dual robustness in design and execution.

[0092] Map hole clusters to partition sets based on spatial relationships , set the reverse phase difference between the partition close to the protected object and the partition far away, so that The main lobe is cancelled near the core, and the frequency weight curve is taken into account by core bandwidth modulation. The width of the partition delay bias is , for two adjacent partitions Regulation:

[0093] Where: Partition delay offset : Partition The unified phase reference, ; , lower time limit (earliest programmable / allowable delay), , time upper limit (latest programmable / allowed delay); Target phase difference : Inverting / semi-inverting target at the center of the sensitive frequency band, ; Center of sensitive frequency band :By frequency weight curve The center frequency is determined by the peak value of ; Adjacent partitions : Adjacent spatial subdomains in the main radiation direction.

[0094] The directional map is constructed by the normal direction of the protected object and the visibility of the hole-measurement point; K-means clustering is obtained in the three-dimensional space of azimuth-distance-geological partitioning. ; Set the target in the forward area , side rear area . Broadband weight compatibility: When The peak width is narrow, and strict compliance is preferred When the peak width is wide, a small random phase code (±2–5ms) is introduced within the partition to improve the average suppression within the band.

[0095] When used, the frequency domain weights are explicitly written back to the spatial arrangement using partition-phase difference, so that the energy close to the direction of the protected object is in the center of the sensitive frequency band. Active cancellation occurs; then by selecting the target phase difference The frequency weight curve is realized by introducing low amplitude jitter in the partition Broadband compatibility; ultimately significantly reducing weighted energy in unfavorable directions without sacrificing distal fragmentation efficiency.

[0096] In order to reduce the frequency band energy to a verifiable index layer, a weighted frequency band threshold is set at each measuring point, and this threshold drives the linkage between the critical hole redundancy window and the readback strategy. A subset of sensitive frequency bands constraint:

[0097] Among them: sensitive frequency band subset :Measurement point Selection of local sensitive frequency bands, ; is the lower frequency limit, is the upper frequency limit; Threshold :The upper bound of weighted energy under regulatory and self-defined safety requirements, ; The high percentile (such as 90%) of the weighted index distribution of historical qualified working conditions and the mandatory requirements of the standard red line are taken as the more stringent value, and the different measurement points of floor slabs / columns / linings are distinguished. Subband, complex spectral component : The predicted complex spectrum of a given time series is .

[0098] In terms of execution strategy, the key hole set Allocate redundancy window with read-only fine-tuning permissions, i.e. the key hole Delay The non-critical hole set is not modified during the online phase. According to the expected energy trajectory The deviation is adjusted slightly once within the tolerance allowed by the regulations. All bias actions are recorded with timestamps and readback check values ​​and logged.

[0099] When in use, the weighted frequency band threshold is used to solidify the appearance criteria of a good design into an acceptable inequality; thereby, the redundancy of key holes and online fine-tuning have a basis and each performs its own function; ultimately, without changing the charge and the total window, robust and strict control of the sensitive frequency band is achieved.

[0100] Step 4: Unify the clock reference As an anchor, put the delay list After writing into the electronic detonator, it is read back hole by hole - consistency check and energy self-test; during the detonation process, the expected energy trajectory is used. As a reference, the measured-predicted deviation is determined by weighted coherence statistics, and only the non-critical hole set Within regulatory tolerances A one-time minor adjustment is made within the detonation window without changing the total window and charge. After the detonation is completed, the source spectrum signature is -Path transfer operator -Boundary parameters -Instrument transmission letter -Frequency weight curve Block assimilation and update to maintain unique mapping and traceability of parameters and indicators throughout the entire link.

[0101] Partition-hole level delay list Reliably sent to the detonation network, based on a unified clock reference The system completes hole-by-hole readback, consistency check and energy self-check, and uses weighted index gating to form the feasibility criteria and access threshold before detonation.

[0102] Step 3 has given the delay timing after robust optimization With weighted threshold However, if there is no digital signature, readback and gating before detonation, any link jitter may eliminate the advantage of spectrum shaping. Therefore, it is necessary to unify the clock reference. Under this, the series closed loop of hole-by-hole programming readback-clock correction-cyclic redundancy check CRC / encryption verification-energy self-test-threshold verification is completed, thereby converting the design optimization into a safe and executable optimization.

[0103] In order to eliminate the phase distortion of the hole group caused by clock drift and programming gap, a global clock offset estimation-hole-by-hole consistency index-tolerance gate triplet of the initiation network is established: first, the readback delay With programmed delay The global clock offset is estimated by the difference of , and then the hole-by-hole residual is normalized and gated. If it exceeds the limit, the detonation is rejected and the write-back is triggered, where:

[0104] Where: Global clock offset : Correction of readback time stamp relative to a unified clock reference The offset is ; Readback delay : The detonation time of the electronic detonator read back hole by hole is ;Programming delay : Download the target detonation time of the detonator, ; Hole index set : Hole cluster, offset to be optimized : scalar variable; searches for the value in the objective function that minimizes the overall deviation.

[0105] Define well-by-well consistency indices and gate them with regulatory tolerances:

[0106] Among them: consistency index : Characterizes the normalized time error of the worst hole, which is ,when Passed; regulatory tolerance : The maximum time amplitude that can be adjusted online and the upper limit of the time scale deviation allowed by the detonation network are .

[0107] When in use, the distributed readback is compressed into an unambiguous quantity of whether it can be detonated through global clock offset estimation and hole-by-hole consistency gating; thereby eliminating the implicit mismatch that will destroy the coherent superposition in advance; and finally, the time domain executability and the frequency domain shaping target are firmly aligned before the order is issued.

[0108] After the consistency gate is passed, the predicted spectrum of the group holes output by the virtual field is Calculate the weighted index of the measuring point and compare it with the threshold Compare the values; use the most unfavorable ratio of the entire field as the admission threshold. If the threshold is exceeded, return to step 3 to rearrange the partition phase and sequence constraints, where:

[0109] Including: admission ratio : The threshold ratio of the most unfavorable measurement point in the whole field is ,when Determined to be passed; measurement point set : Monitoring location set of the protected object; Weighted energy index of the measuring point :Measurement point evaluation quantity consistent with step 2-3, ; Threshold : The upper bound of the weighted energy of the sensitive frequency band is .

[0110] When in use, the frequency band sensitivity is solidified as the entry threshold before detonation with a single dimensionless indicator; thereby avoiding the one-sidedness of relying solely on peak velocity; and ultimately ensuring the a priori correctness of strictly suppressing the sensitive frequency band upon detonation.

[0111] Detonation online phase with expected energy trajectory As a reference, the deviation is determined by weighted coherence statistics, and only the non-critical hole set In tolerance A one-time fine-tuning is performed within the system; after detonation, the source-path-boundary-instrument-weight curve is updated by block assimilation with the weighted spectral domain loss as the target, closing the data-model-control self-learning loop.

[0112] In real-world scenarios, the transient wavefields at the free surface and cavern boundaries can deviate from the nominal parameters of the virtual field. If these deviations are not assessed online and fine-tuned once within the legally permitted timeframe, coherent suppression of sensitive frequency bands may be weakened. Excessive online adjustments can violate the safety margins and timing discipline of critical holes. Therefore, a minimal intervention strategy of strict triggering, one-time fine-tuning, and full-chain traceability is required. Afterward, the deviations are decomposed back into the source term, path, boundary, instrument, and weight curve parameters for self-consistent updates.

[0113] Sliding time window to measure short time spectrum and predicted spectrum Do weighted complex correlation to get dimensionless deviation statistics ; Only when the maximum deviation across the measurement points exceeds the trigger threshold and the remaining time slot is sufficient to perform a safe fine-tuning, the non-critical hole set is Calculate the one-time update amount, where:

[0114] Where: Deviation statistic : Weighted coherence loss measure for observed-predicted, ; Measured short-time spectrum :Measuring point No. Weight at the moment The short-time spectrum of ; Prediction spectrum : The reference complex spectrum of the virtual scene output belongs to the complex set ;Frequency weight curve ,bandwidth : Evaluation weight and upper and lower limits of sensitive frequency bands, , .

[0115] The maximum value of the span point With trigger threshold Compare, if And the remaining time slot allows, then calculate the non-critical hole update amount Keyhole Collection Read-only.

[0116] Solve the problem in one go using the least squares approximation of the first-order sensitivity in the frequency domain:

[0117] Among them: fine-tuning amount :hole One-time delay correction, ; Truncation operator : Project the solution to the regulatory tolerance interval, without value interval; error spectrum : The instantaneous deviation complex spectrum is ; Sensitivity kernel : spectrum to hole The linearized response of the delay is ; Trigger threshold : The deviation trigger threshold for online fine-tuning is .

[0118] When used, the deviation statistic measures the lack of coherence in the spectrum-vector space, avoiding the hysteresis of only looking at the time domain peak; thus making the fine-tuning trigger both sensitive and controllable to the sensitive frequency band; ultimately achieving the minimum necessary intervention with a one-step approximation, and limiting the non-critical holes to ensure a safe margin.

[0119] After detonation, the source spectrum signature is calculated with the weighted spectrum domain loss function as the target. -Path transfer operator -Boundary parameters -Instrument transmission letter -Weight curve parameters Perform block-wise updates; the updates are driven by the projection of the prediction-measurement error onto the Jacobian of each block, and are stabilized using a priori regularization, where:

[0120] Among them: parameter block set : The model and weight parameter family to be assimilated is the feasible domain ; Measured spectrum : The complex spectrum normalized to the unified time scale and range after detonation is ; Prediction spectrum : The complex spectrum generated by the parameter block is ; Regularization term :The prior constraints on parameter increments such as smoothness / sparseness / low rank are ; Regularization weight : The trade-off between data fitting and priors is . , take the real part and ensure that the step size is a real number; When used, the error tracing of weight curve-propagation-source term-instrument is incorporated into the same objective function to avoid parameter drift caused by segmented patching; then, the prior regularization is used to suppress overfitting and maintain transferability; finally, the next round of virtual scene is consistent with the expected energy trajectory. Closer to reality.

[0121] During the programming phase, hole-by-hole addressing, CRC, encryption handshake, energy self-test, and readback snapshot links are performed at the bus layer. All readback values ​​are based on the same clock reference. The corresponding relationship is written into the event log; the consistency gate is based on the consistency index Ratio to admission Double threshold judgment, if any one exceeds the limit, the detonation will be rejected and the process will return to step 3; in the online stage, only the non-critical hole set Within regulatory tolerances Perform one-time fine-tuning within the key hole collection Redundancy window Only reflected in the nominal scheme; fine-tuning actions and trigger thresholds Stored with timestamp; subsequent assimilation updates follow the fixed order of source-path-boundary-instrument-weight, with priority given to updating boundary reflections and propagation delay , then update the source spectrum signature , and finally adjust the weight curve parameters with a low amplitude to ensure that the physical meaning of sensitive frequency bands is not diluted.

[0122] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0123] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0124] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only for some logical functions. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0125] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0126] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. Low-vibration, multi-point delay digital twin-driven intelligent electronic detonator blasting control method, characterized by: include, Three-component sensors are deployed under a unified clock reference to collect responses. The responses are restored to the structural coordinate system based on the direction mapping matrix and the range normalization matrix. The group power spectrum and modal dominance are calculated, and the frequency weight curve and weighted energy index are constructed. The three-component complex spectrum of a single hole is deconvolved on the unified clock reference by combining the instrument transfer function with the path transfer operator to obtain the source spectrum signature. The boundary correction propagation operator and the detonator timing characteristic function are introduced to form the group hole prediction complex spectrum and energy trajectory. A weighted robust target is constructed using the frequency weight curve and the timing characteristic function, a delay vector is established under the constraints of the operation time window and the detonation sequence, a delay list is obtained by updating the projected gradient, and a phase difference and a key hole redundancy window are set according to the spatial partition; Writing the delay list into the electronic detonator, estimating the clock offset according to the unified clock reference, and calculating the consistency index and admission ratio; Triggering a one-time fine-tuning of only the non-critical hole online with weighted coherence; After detonation, the source spectrum signature, path transfer operator, boundary parameters, instrument transfer function and frequency weight curve are updated by block assimilation.

2. The digital twin-driven intelligent electronic detonator blasting control method according to claim 1 is characterized by: A unified clock reference is established, and the three-component response is restored to the structural coordinate system by combining the installation attitude calibration matrix and range normalization. The group power spectrum is calculated according to the multi-window STFT, and the saturation and coupling degradation channels are eliminated using the coherence function to form the in-service frequency response baseline. The measurement point coordinates and component properties are written into the metadata for unified archiving.

3. The digital twin-driven intelligent electronic detonator blasting control method according to claim 2 is characterized by: The output is used to identify the modal shape, poles and damping, construct the participation coefficient and the through term, and calculate the modal dominance spectrum. This is then coupled with the frequency response function amplitude according to the power exponent to form a frequency weight curve, and the weighted energy index and evaluation bandwidth are defined using the three-component Minkowski composite exponent and the penalty power exponent.

4. The digital twin-driven intelligent electronic detonator blasting control method according to claim 3 is characterized by: Perform Tikhonov regularized deconvolution on the single-hole complex spectrum under the instrument transfer function and the path transfer operator, wherein the instrument transfer function is a three-component transfer matrix; The path transfer operator is composed of direction cosine matrix, distance power attenuation, phase delay and quality factor attenuation, and a boundary correction propagation operator is used to introduce complex reflection coefficient, round-trip propagation delay and mode conversion matrix.

5. The digital twin-driven intelligent electronic detonator blasting control method according to claim 4 is characterized in that: The dispersion is characterized by the timing characteristic function of the electronic detonator. The planned timing addresses the hole index set hole by hole with a unified clock reference. The uniform jitter half interval is dynamically updated by readback statistics and is superimposed with the boundary correction propagation operator to obtain the phase consistency factor of the complex spectrum for group hole prediction.

6. The digital twin-driven intelligent electronic detonator blasting control method according to claim 5 is characterized by: A robust target with a weighted energy upper bound is constructed using the most unfavorable phase disturbance. The uncertainty set is defined as the absolute value of the delay disturbance of each hole not exceeding a preset upper limit. Under the constraints of the operation time window and the difference between the detonation sequence, the delay vector is updated using projection-gradient. The projection is first clipped to the time window hole by hole, and then the difference constraint is alternately iterated by Euclidean projection until it no longer changes. The resulting delay list is used for programming.

7. The digital twin-driven intelligent electronic detonator blasting control method according to claim 6 is characterized by: The pore groups are clustered according to azimuth, distance and geological attributes on the directional map to obtain partition sets; The target phase difference of adjacent partitions is set around the sensitive center frequency of the frequency weight curve, the partition delay offset is limited to each operation time window, and a read-only redundant window and a readback strategy are configured for the critical hole.

8. The digital twin-driven intelligent electronic detonator blasting control method according to claim 7 is characterized in that: The delay list is written into the electronic detonator and read back hole by hole, the global clock offset is estimated based on the difference between the read-back delay and the programmed delay, and the hole-by-hole consistency index is calculated according to the normalized time error, and compared with the preset regulatory tolerance threshold for determination; The admission ratio of the weighted energy index to the threshold is calculated at each measuring point and the most unfavorable value of the entire field is taken. If any condition is not met, the system falls back to adjusting the phase difference and sequence constraints.

9. The digital twin-driven intelligent electronic detonator blasting control method according to claim 8, characterized in that: Deviation statistics are constructed using the weighted complex correlation between the sliding short-time spectrum and the virtual field prediction spectrum. When the maximum deviation across measurement points exceeds the trigger threshold and there are sufficient remaining time slots, a one-time least squares update is performed on the non-critical holes according to the first-order sensitivity of the spectrum to the delay. The truncation operator limits the correction amount to the regulatory tolerance interval, the critical holes are not updated, and all operations and timestamps are written to the event log.

10. The digital twin driven intelligent electronic detonator blasting control method according to claim 9, characterized in that: Post-hoc assimilation includes: using weighted spectral domain loss as the objective function and introducing prior regularization such as smoothness, sparseness or low rank, and updating in blocks according to the fixed order of boundary parameters and propagation operator → source spectrum signature → instrument propagation function → frequency weight curve; The update amount is obtained by projecting the error onto the Jacobian of each block, and is uniformly addressed and stored using a set of hole indices and measurement point indices. The parameter increment is constrained by the regularization weight.

Images