Coal seam floor water inrush probability imaging method based on microseismic tail wave signals

Abstract

The application discloses a coal seam floor water inrush probability imaging method based on microseismic tail wave signals and belongs to the field of coal mine water disaster monitoring and geophysical imaging. The method first collects and screens effective microseismic events of the floor, determines a tail wave time window and extracts frequency dispersion characteristics; a frequency dispersion response index which is positively correlated with water-bearing fracture opening and dimensionless is constructed; extreme value normalization processing is performed on the frequency dispersion response index of all effective microseismic events in the evaluation area to obtain a relative water conductivity indication index; the spatial coordinates of each effective microseismic event are associated with the corresponding relative water conductivity indication index to construct a spatial discrete data set; and a continuous floor water inrush probability distribution image is generated through spatial interpolation. The application does not need to accurately invert fracture parameters, can effectively reduce the influence of observation condition differences, converts discrete single event evaluation into regional continuous imaging, intuitively delineates a water inrush danger area and is suitable for dynamic evaluation and prevention and control of the coal seam floor water inrush probability under complex mining conditions.

Description

Technical Field

[0001] This invention relates to the field of microseismic monitoring and geophysical imaging technology for coal mine water hazards, specifically to a probabilistic imaging method for water inrush at the bottom of coal seams based on microseismic wake signals. Background Technology

[0002] During deep coal seam mining, mining-induced stress can cause floor rock mass fracturing, crack propagation, and the evolution of seepage channels. When the floor fracturing zone connects with the underlying confined aquifer, fault, or other water-conducting structures, it may form a water inrush channel and induce mine water hazards. Therefore, timely understanding of the spatial coupling relationship between floor fracturing activity and water-bearing anomalies is an important foundation for the identification and prevention of water inrush hazard areas.

[0003] Current microseismic monitoring typically analyzes foundation rupture activity based on event location, magnitude, or energy, yielding results that are mostly discrete source points or event density distributions. While these methods can reflect the degree of rock mass rupture activity, they fail to adequately utilize information on fluid fracture response, scattering attenuation, and dispersive wake waves contained after the direct wave of the microseismic waveform, making it difficult to directly generate a continuous probability distribution of foundation water inrush. Furthermore, relying solely on a single waveform parameter or a single microseismic event is easily affected by source energy, propagation distance, detector coupling, and on-site noise.

[0004] Microseismic wakes consist of scattered waves, converted waves, and the response of the fracture medium following the direct wave. Frequency-dependent time variations, wake energy proportions, and duration can reflect the development of fractures and the degree of fluid involvement near the propagation path. In particular, Krauklis waves radiating into the surrounding rock from fluid-filled fractures can be recorded as a wake response with dispersion characteristics in external geophones. Existing technologies utilize the dispersion characteristics of Krauklis waves to invert fracture aperture, but these all aim for accurate single-event inversion, resulting in the following shortcomings: First, fracture aperture inversion depends on precise propagation paths and medium parameters, leading to large inversion errors under complex field conditions; second, single-event results are affected by factors such as source energy and propagation distance, and there is a lack of a unified and comparable evaluation standard for the water conduction probability between different events; third, only discrete point evaluation results can be obtained, failing to intuitively present the continuous distribution of regional-scale water inrush probability, making it difficult to directly guide on-site water hazard prevention and control.

[0005] Therefore, it is necessary to transform the wake wave characteristics from single-event identification information into probability factors that can participate in spatial fusion, and to establish a probability imaging method for bottom plate water inrush by combining the spatial distribution of microseismic events. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for probabilistic imaging of coal seam floor water inrush based on microseismic wake signals. This method solves the problems of insufficient utilization of wake information, poor comparability of single-event evaluations, and difficulty in intuitively reflecting regional water inrush risk with discrete results in existing technologies, thereby achieving continuous, quantitative, and visual evaluation of the probability of coal seam floor water inrush.

[0007] To achieve the above objectives, the technical solution adopted by this invention is: a probabilistic imaging method for coal seam floor water inrush based on microseismic wake signals, comprising the following steps: S1. Microseismic signal acquisition and effective microseismic event screening: Microseismic detectors are installed in the roadways and floor boreholes of the coal seam working face to continuously acquire microseismic signals induced by mining. The acquired signals are preprocessed, including noise reduction, trigger detection, first arrival pickup and source location. Effective microseismic events located within the influence range of the coal seam floor are screened, and the arrival time of the direct wave of the corresponding waveform of each event is determined.

[0008] S2. Wake wave time window determination and dispersion feature extraction: The signal segment after the arrival time of the direct wave is taken as the candidate region of the wake wave. The effective wake wave time window is determined according to the continuity of wake wave energy attenuation and the on-site noise level. Continuous wavelet time-frequency analysis (CWT) is performed on the single-channel microseismic signal within the wake wave time window to obtain the time-frequency energy distribution. The arrival time of the energy peak corresponding to different frequencies is extracted to obtain the wake wave dispersion response characteristics.

[0009] S3. Construction of the Fracture Aperture Sensitive Dispersion Response Index (FDMI): The frequency and peak arrival time within the effective frequency band of the wake (i.e., Krauklis wave) are normalized respectively, the local dispersion intensity and time-frequency energy weights are calculated, and the dimensionless dispersion response index (FDMI) is obtained by fusion. The dispersion response index increases monotonically with the increase of the water-bearing fracture aperture and is used to characterize the relative intensity of the fracture water-conducting possibility.

[0010] S4, Relative water-conducting indicator I K Calculation: To reduce the impact of differences in effective frequency band, propagation distance, source energy, and observation conditions of various microseismic events, the event-level water inrush hazard index of all effective microseismic events in the evaluation area was normalized to obtain a relative water conductivity indicator I with a value range of 0 to 1. K .

[0011] S5. Spatial Imaging of Bottom Plate Water Inrush Probability: Correlating the spatial coordinates of each effective microseismic event with the corresponding relative water-conducting index value to construct a spatial discrete dataset. ;in, For the first Spatial coordinates of an effective microseismic event For the first The relative water-conductivity index values ​​of each effective microseismic event were obtained; a three-dimensional imaging grid of the coal seam floor was established, and the discrete relative water-conductivity index values ​​were mapped to the grid nodes using a spatial interpolation method to generate a continuous spatial distribution image of the relative water-conductivity index values, thereby delineating the water inrush hazard area.

[0012] Furthermore, in step S2, the specific process of continuous wavelet transform and time-frequency feature extraction is as follows: Single-channel microseismic signal within the wake time window Perform continuous wavelet transform:

[0013] In the formula, These are the wavelet transform coefficients. For observing signals, The scale parameter and , For translation parameters, To satisfy the admissibility condition of the wavelet mother function, It is the conjugate of the wavelet mother function.

[0014] Calculate the time-frequency energy distribution based on the wavelet transform coefficients:

[0015] Extracting the energy peak values ​​corresponding to different frequencies to time:

[0016] In the formula, For frequency The corresponding energy peak arrival time, For the effective wake time window.

[0017] Furthermore, step S3 specifically includes: S3.1, Frequency and peak time normalization:

[0018]

[0019] In the formula, The normalized frequency, When the normalized peak value is reached, The upper and lower limits of the effective frequency band of the wake (Krauklis wave) are determined by the dispersion energy ridge obtained by continuous tracking within the wake time window; These represent the maximum and minimum values ​​of the peak value within the effective frequency band, respectively.

[0020] S3.2 Calculate the local dispersion intensity:

[0021] S3.3 Calculate the time-frequency energy weights:

[0022] In the formula, This represents the maximum value of time-frequency energy within the effective frequency band.

[0023] S3.4 Constructing the dimensionless dispersion response index FDMI: .

[0024] Furthermore, in step S4, the formula for calculating the relative water-conducting indicator is:

[0025] In the formula, The value range is [0,1].

[0026] Furthermore, the first The relative water conductivity indicator for each effective microseismic event is calculated using the following formula:

[0027] In the formula, For the first The relative water conductivity indicator of an effective microseismic event. For the first Dispersion response index of an effective microseismic event These represent the maximum and minimum values ​​of the dispersion response index for all effective microseismic events within the evaluation area, respectively. The value range is [0,1]. The larger the value, the higher the probability of water diversion in the area corresponding to the event.

[0028] Further, in step S5, the spatial interpolation method includes any one of inverse distance weighted interpolation, Kriging interpolation, and radial basis function interpolation; the calculation formula for spatial interpolation is:

[0029] In the formula, The location of the grid node to be calculated, For the first The spatial weight of each effective microseismic event to that node This represents the total number of effective microseismic events.

[0030] Furthermore, after step S5, a result verification step is also included: combining the time shift anomaly of the apparent resistivity of the bottom plate, the borehole exposure results, the location of the aquifer, and the mine hydrogeological data to comprehensively verify and correct the water inrush probability imaging results.

[0031] Compared with the prior art, the present invention has the following advantages: 1. This invention constructs a dimensionless dispersion response index FDMI that is positively correlated with the aperture of water-bearing fractures. It can transform the dispersion characteristics of microseismic wake waves into relative evaluation parameters of the water conductivity of fractures without directly inverting the specific geometric parameters of the fractures. This avoids the accumulation of errors in parameter inversion under complex media conditions, and the evaluation results are stable and reliable.

[0032] 2. This invention constructs a relative water-conductivity indicator by normalizing regional extreme values, which effectively reduces the impact of differences in source energy, propagation distance, and observation conditions between different events, realizes a unified quantitative comparison of the water-conductivity probability at different spatial locations, and eliminates the limitations of single-event evaluation.

[0033] 3. This invention converts the evaluation results of discrete microseismic events into a continuous three-dimensional probability distribution image through spatial interpolation, which intuitively presents the location, range and spatial continuity of the high-risk area of ​​water inflow in the foundation slab. It can be directly used for the delineation of the inrush danger zone and the formulation of prevention and control plans on site.

[0034] 4. This invention can directly utilize the data collected by the existing mine microseismic monitoring system without the need for additional special monitoring equipment. At the same time, it can be fused and verified with multi-source data such as electrical resistivity and borehole data, and is suitable for dynamic evaluation of the probability of water inrush in the floor under complex mining conditions. Attached Figure Description

[0035] Figure 1 This is a flowchart illustrating the overall technical process of the present invention.

[0036] Figure 2 The diagram shows a comparison of the time-domain and time-frequency characteristics of three typical microseismic records, where Type I has no obvious coda, Type II has a coda but no dispersion, and Type III has a coda and dispersion.

[0037] Figure 3 This is a spatial imaging result of the probability of water inrush at the bottom of the coal seam obtained by the present invention. Detailed Implementation

[0038] The present invention will be further described below.

[0039] This embodiment uses the 1105 longwall mining face of a certain mine as the engineering application object. The average burial depth of this working face is 820m, mainly mining the No. 3 coal seam with an average thickness of 3.2m. The coal seam floor is mainly composed of siltstone and sandy mudstone, with an underlying Ordovician limestone confined aquifer with a water pressure of about 6.2MPa. The F12 normal fault is developed in the middle of the working face with a drop of 0-8m, which is a typical working face with a high-pressure floor threatened by water inrush. The strike length of the working face is 1200m and the dip length is 200m. During the mining process, the method of this invention is used to dynamically evaluate the probability of floor water inrush.

[0040] The specific implementation steps of this embodiment are as follows: S1. Microseismic signal acquisition and effective microseismic event screening: Four three-component microseismic geophones are installed in the working face transport roadway and return air roadway, and six geophones are installed in the advance detection borehole of the floor plate, for a total of 14 monitoring nodes. The main frequency of the geophones is 100Hz, the sampling rate is set to 1000Hz, and the monitoring range fully covers the working face and the 100m depth range below the floor plate.

[0041] Raw microseismic data were continuously collected during the 30-day mining period. After denoising using a 5-300Hz bandpass filter, valid microseismic events were detected using the STA / LTA triggering algorithm, P-wave first arrivals were manually picked up, and three-dimensional seismic source localization was performed using the Geiger method, resulting in the location of 1286 valid microseismic events. Based on the predicted failure depth of the working face floor, 427 microseismic events within the range of 0-80m below the floor were selected as valid evaluation events. The arrival time of the P-wave for each event was also determined.

[0042] S2. Wake wave time window determination and dispersion feature extraction: The wake wave start time is defined as 0.02 seconds after the arrival of the P-wave. The wake-end time is defined as the moment when the wake energy decays to twice the background noise. The effective wake time window is determined for each channel. In this embodiment, the wake time window length is generally distributed in the range of 0.2~0.8s.

[0043] Using the Morlet wavelet as the wavelet mother function, a continuous wavelet transform is performed on the single-channel microseismic signal within the wake time window to obtain the time-frequency energy distribution. Tracing continuous dissipative energy ridges within the 50–250 Hz frequency band, and extracting the energy peak values ​​corresponding to different frequencies to time... The wake dispersion response characteristics were obtained.

[0044] like Figure 2 As shown, the on-site microseismic records can be divided into three categories: Type I has no obvious coda and no guided wave evidence; Type II has a coda but no obvious dispersion and insufficient evidence of K-wave radiation; Type III has a clear frequency-time positive correlation dispersion trend, corresponding to a stable Krauklis wave radiation response, which is a typical feature of water-bearing fractures.

[0045] S3. Calculation of Dispersion Response Index (FDMI): For each effective microseismic event, the 50-200Hz range from which the dispersion energy ridge is continuously traceable is selected as the effective frequency band of the Krauklis wave, i.e. , .

[0046] First, the frequency and peak arrival times are normalized to eliminate systematic differences in propagation distances of different events; second, the normalized local dispersion intensity is calculated. This reflects the steepness of the dispersion curve; then the time-frequency energy weight is calculated. The contribution of high-energy frequency bands is highlighted; finally, the comprehensive dispersion energy is calculated by trapezoidal numerical integration, and the dispersion response index of each effective microseismic event is obtained by substituting it into the dispersion response index FDMI formula.

[0047] In this embodiment, the FDMI values ​​of the 427 valid microseismic events ranged from 0.21 to 0.89; among them, the FDMI values ​​of events located near the F12 fault were generally higher than 0.7, corresponding to larger fracture aperture, higher fluid filling degree, and stronger water conductivity, which is consistent with the trend that the dispersion response index FDMI increases with the increase of fracture aperture.

[0048] S4, Relative water-conducting indicator I K Calculate: The maximum and minimum values ​​of the dispersion response index for all effective microseismic events within the statistical evaluation area, respectively. Minimum value The relative water conductivity index for each event is calculated using the relative water conductivity index formula. The value is normalized to the range of [0,1]. The closer the relative water conductivity indicator value is to 1, the higher the relative probability of water conduction from the fracture at that location.

[0049] S5. Spatial Imaging and Result Analysis of Water Inrush Probability in the Working Face: A three-dimensional imaging grid was established for the working face's bottom plate, with a grid step size of 5m×5m×5m, covering the entire working face area. Inverse distance weighted interpolation (IDW) was used, with a distance power of 2, to map the relative water-conductivity index values ​​of discrete microseismic events to each grid node, generating a continuous three-dimensional imaging map of the probability of water inrush in the bottom plate. The results are as follows: Figure 3 As shown. Hazard levels are classified based on the relative magnitude of the water-conducting indicator value: I K A value >0.7 indicates a Level 1 water inrush hazard zone, and a value ≤0.4 indicates a Level 1 water inrush hazard zone. K ≤0.7 is a level II water inrush hazard zone, I K A value less than 0.4 indicates a relatively safe zone.

[0050] Imaging results show that there is a continuous distribution of high I-values ​​in the middle of the working face, within a range of 480-620m from the cut eye. K The area with the highest water inrush risk level coincides closely with the spatial location of the F12 fault and is identified as a Level I water inrush hazard zone; strip-shaped Level II hazard zones exist in localized areas at both ends of the working face; the remaining majority of the area is classified as Level I. K The value is low, indicating a relatively safe zone.

[0051] S6. Result Verification and Check: The time-shifted apparent resistivity of the bottom plate obtained from parallel electrical resistivity monitoring of the working face was used for verification. The negative apparent resistivity anomaly area (decreased resistivity, corresponding to increased water content in the rock mass) was compared with the high I area delineated by this method. K The spatial locations of the danger zones highly overlap; at the same time, drilling into the bottom plate confirmed that cracks had developed in the bottom plate within the first-level danger zone, and water-conducting channels existed in some areas, verifying the accuracy and reliability of the method of the present invention.

Claims

1. A probabilistic imaging method for water inrush at the bottom of a coal seam based on microseismic wake signals, characterized in that, Includes the following steps: S1. Collect microseismic signals from the coal seam working face, preprocess the microseismic signals, screen the effective microseismic events located within the influence range of the coal seam floor, and determine the arrival time of the direct wave of the waveform corresponding to each event. S2. The signal segment after the arrival of the direct wave is taken as the candidate region of the wake wave. The effective wake wave time window is determined according to the continuity of wake wave energy attenuation and the on-site noise level. The single-channel microseismic signal within the wake wave time window is subjected to continuous wavelet transform to obtain the time-frequency energy distribution. The arrival time of the energy peak corresponding to different frequencies is extracted to obtain the wake wave dispersion response characteristics. S3. Normalize the frequency and peak arrival time of the effective frequency band of the wake wave, calculate the local dispersion intensity and time-frequency energy weight, and fuse them to obtain the dispersion response index. S4. Perform extreme value normalization on the dispersion response index of all effective microseismic events in the evaluation area to obtain the relative water conductivity index. S5. Associate the spatial coordinates of each effective microseismic event with the corresponding relative water-conducting indicators to construct a spatial discrete dataset; use spatial interpolation to map the discrete relative water-conducting indicators into the coal seam floor imaging grid to generate a continuous spatial distribution image of the relative water-conducting indicators and delineate the water inrush hazard area.

2. The method according to claim 1, characterized in that, In step S2, the specific process of continuous wavelet transform and time-frequency feature extraction is as follows: Single-channel microseismic signal within the wake time window Perform continuous wavelet transform: In the formula, These are the wavelet transform coefficients. For observing signals, The scale parameter and , For translation parameters, To satisfy the admissibility condition of the wavelet mother function, It is the conjugate of the wavelet mother function; Calculate the time-frequency energy distribution based on the wavelet transform coefficients: Extracting the energy peak values ​​corresponding to different frequencies to time: In the formula, For frequency The corresponding energy peak arrival time, For the effective wake time window.

3. The method according to claim 2, characterized in that, Step S3 specifically includes: S3.1, Frequency and peak time normalization: In the formula, The normalized frequency, When the normalized peak value is reached, The upper and lower limits of the effective frequency band of the wake wave are determined by the dispersion energy ridge obtained by continuous tracking within the wake wave time window. These represent the maximum and minimum values ​​of the peak value within the effective frequency band, respectively. S3.2 Calculate the local dispersion intensity: S3.3 Calculate the time-frequency energy weights: In the formula, This represents the maximum time-frequency energy within the effective frequency band. S3.4 Constructing the dimensionless dispersion response index FDMI: 。 4. The method according to claim 3, characterized in that, In step S4, the formula for calculating the relative water-conducting indicator is: In the formula, For the first The relative water conductivity indicator of an effective microseismic event. For the first Dispersion response index of an effective microseismic event These represent the maximum and minimum values ​​of the dispersion response index for all valid microseismic events within the evaluation area.

5. The method according to claim 4, characterized in that, In step S5, the spatial interpolation method includes any one of inverse distance weighted interpolation, Kriging interpolation, and radial basis function interpolation; the calculation formula for spatial interpolation is: In the formula, The location of the grid node to be calculated, For the first The spatial weight of each effective microseismic event to that node This represents the total number of effective microseismic events.

6. The method according to claim 1, characterized in that, The step S5 is followed by a result verification step: combining the time shift anomaly of the apparent resistivity of the base plate, the borehole exposure results, the location of the aquifer, and the mine hydrogeological data, to comprehensively verify and correct the water inrush probability imaging results.

Images