Goaf air leakage identification and sealing evaluation method and system based on data fusion

By using multi-source sensor networks and data fusion technology, the problems of the complexity of air leakage channels in goaf areas and the difficulty in quantifying the sealing effect have been solved. This has enabled accurate identification of air leakage areas and quantitative evaluation of sealing effects, thereby improving the reliability and efficiency of coal mine safety production.

CN123068872APending Publication Date: 2026-11-10XI'AN UNIVERSITY OF ARCHITECTURE AND TECHNOLOGY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611562012.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-10
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Under the conditions of repeated overlapping and disturbance mining of shallowly buried, closely spaced coal seams, the air leakage channels in the goaf are complex, the leakage direction is variable, and the sealing effect is difficult to quantify and evaluate. Existing technologies cannot accurately identify the leakage area and quantify the leakage intensity, and the sealing effect is difficult to quantify.

Method used

A multi-source sensor network is used to collect data on pressure field, temperature field, gas concentration field, and wind speed field. Combined with SF6 tracer gas, a multi-physics data framework is constructed through data fusion technology. The leakage contribution and sealing effect of grid cells are calculated, and multi-dimensional improvement rate and comprehensive score are defined to realize leakage area identification and quantitative evaluation of sealing effect.

Benefits of technology

It improves the accuracy of locating air leakage areas, reduces misjudgments and omissions, provides a quantitative basis for optimizing sealing schemes, reduces the subjectivity and lag of manual judgment, and makes the evaluation results more accurate and reliable. It can identify incorrect sealing locations and guide the correction of sealing schemes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123068872A_ABST
    Figure CN123068872A_ABST
Patent Text Reader

Abstract

The application discloses a goaf air leakage identification and sealing evaluation method and system based on data fusion, relates to the technical field of coal mine safety monitoring, and comprises the following steps: deploying a multi-source sensor network to collect pressure, temperature, gas concentration and wind speed data, releasing SF6 tracer gas, detecting the concentration and first arrival time of each receiving point, and obtaining surface fissure data; fusing the multi-source data to calculate the air leakage intensity factor of each grid unit, fusing the maximum value of five physical components, identifying the air leakage area, and calculating the air leakage channel identification confidence; calculating the sealing efficiency coefficient according to the change of the air leakage intensity factor after sealing; and finally coupling the air leakage intensity improvement rate with the tracer gas concentration improvement rate, coupling the temperature improvement rate with the pressure balance improvement rate, and multiplying the two to obtain a comprehensive sealing effect score. The application can identify the position of the goaf air leakage channel and quantitatively evaluate the sealing effect.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of coal mine safety monitoring technology, specifically to a method and system for identifying and assessing air leakage in goaf areas based on data fusion. Background Technology

[0002] Under the conditions of repeated overlapping and disturbance mining of shallowly buried, closely spaced coal seams, the complex air leakage channels in the goaf, the variable air leakage direction, and the difficulty in quantifying and evaluating the sealing effect are long-standing technical challenges in the field of coal mine safety production.

[0003] Currently, the identification of air leakage in goaf areas mainly relies on threshold judgments of single physical quantities. Estimating the air leakage volume in the goaf by comparing the air volume difference between the intake and return airways only provides a rough estimate of the overall leakage volume and cannot pinpoint the specific location of the leakage area. Determining the presence of leakage by checking the decrease in oxygen concentration at the return air corner is not specific enough, as oxygen concentration decreases can be caused by various factors such as the oxidation of residual coal and the influx of inert gases. Qualitatively identifying leakage channels by manually releasing SF6 tracer gas and detecting its concentration at various points is labor-intensive, inefficient, and cannot provide real-time continuous monitoring; it can only determine the existence of leakage channels, not quantify the intensity and spatial distribution of leakage. Furthermore, existing technologies mostly use qualitative descriptions or single-indicator comparisons, lacking multi-dimensional comprehensive quantitative evaluation methods.

[0004] Therefore, there is an urgent need for a method and system that can integrate multi-source sensor information to accurately identify air leaks and quantitatively evaluate the sealing effect from multiple dimensions. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address the shortcomings of the existing technology by providing a method and system for identifying and assessing air leakage in goaf areas based on data fusion.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The data fusion-based method for identifying and assessing air leakage in goaf areas includes the following steps:

[0008] Step S1: Deploy a multi-source sensor network in the goaf and surrounding roadways to collect pressure field data, temperature field data, gas concentration field data and wind speed field data of the goaf. At the same time, release SF6 tracer gas in the goaf and detect the concentration and first arrival time at each receiving point to obtain the width of the surface fissures and the distance to the working face.

[0009] Step S2: Perform time synchronization, spatial registration, outlier removal and normalization on the collected multi-source data to construct a unified multi-physics data fusion framework;

[0010] Step S3: Based on the fused multiphysics data, the goaf is divided into several grid units. Five components are calculated for each grid unit: pressure field leakage contribution, temperature field leakage contribution, gas concentration field leakage contribution, wind speed field leakage contribution, and surface fissure leakage contribution. The maximum value after adjustment by the adjustment coefficient is taken as the leakage intensity factor of the grid unit.

[0011] Step S4: Compare the leakage intensity factor of each grid cell with the preset leakage intensity judgment threshold to identify the leakage area and calculate the identification confidence of the leakage channel.

[0012] Step S5: After implementing the sealing measures, collect multiphysics data after sealing, calculate the leakage intensity factor of each grid after sealing, calculate the amplitude improvement factor based on the average leakage intensity factor before and after sealing, calculate the trend stability factor based on the time series slope of the leakage intensity factor after sealing, and multiply the amplitude improvement factor and the trend stability factor to obtain the sealing efficiency coefficient.

[0013] Step S6: Calculate the comprehensive score of the sealing effect. Define four improvement rates: leakage intensity improvement rate, tracer gas concentration improvement rate, temperature improvement rate, and pressure equalization improvement rate. Couple the leakage intensity improvement rate with the tracer gas concentration improvement rate to obtain the channel sealing effectiveness index. Couple the temperature improvement rate with the pressure equalization improvement rate to obtain the environmental recovery effectiveness index. Multiply the channel sealing effectiveness index with the environmental recovery effectiveness index to obtain the comprehensive score of the sealing effect.

[0014] Further, in step S1, pressure sensors are deployed inside the goaf, at the intake corner, at the return corner, and in the intake and return airway; temperature sensors are deployed inside the goaf and at the return corner; oxygen concentration sensors and SF6 tracer gas concentration sensors are deployed inside the goaf, at the return corner, and at surface fissures; wind speed sensors are deployed in the intake and return airway; barometers are deployed at corresponding surface locations; an SF6 release device is installed on the intake side or intake corner of the goaf to stably release SF6 gas; gas receiving and sampling points are set at surface fissure zones, return corners, and the sealed wall of the overlying goaf; if SF6 is detected at a receiving point, the fissure corresponding to that receiving point is determined to be an effective penetrating fissure; and the time difference from the start of release to the first detection of SF6 is recorded as the SF6 penetration time.

[0015] Furthermore, in step S2, the time synchronization processing unifies sensor data with different sampling frequencies to the same time reference through a linear interpolation algorithm; the spatial registration processing maps sensor data from different spatial locations to a unified three-dimensional grid coordinate system of the goaf; the outlier removal processing adopts the Laida criterion, and when a data point deviates from the mean by more than 3 times the standard deviation, it is judged as an outlier and removed; the data normalization processing transforms sensor data with different dimensions to the same interval through the minimum-maximum normalization method.

[0016] Further, in step S3, the contribution of pressure field leakage is obtained by normalized cross-correlation analysis of goaf pressure and surface air pressure, characterizing the intensity of the goaf breathing effect; the contribution of temperature field leakage is obtained by statistically analyzing the proportion of simultaneous decrease in oxygen concentration and increase in temperature, characterizing the thermal response synchronization rate of the goaf; the contribution of gas concentration field leakage is obtained by multiplying the SF6 penetration time ratio and peak concentration ratio, characterizing the dominance of tracer gas arrival; the contribution of wind speed field leakage is obtained by the cumulative offset of the difference between inlet and outlet air volume within the sliding time window, characterizing the intensity of cumulative leakage trend; and the contribution of surface fissure leakage is obtained by multiplying the fissure width ratio, SF6 penetration time ratio, and distance attenuation factor, characterizing the effective conduction index of the fissure.

[0017] Furthermore, in step S3, the calculation method for the pressure field leakage contribution is as follows: normalized cross-correlation analysis is performed on the average pressure of the goaf and the surface air pressure, and the maximum absolute value of the cross-correlation function is taken as the pressure field leakage contribution.

[0018] The contribution of air leakage in the temperature field is calculated as follows: calculate the temperature change rate for each temperature sensor, and sum the absolute values ​​of the temperature change rates during the period when the oxygen concentration decreases and the temperature rises, and divide by the sum of the absolute values ​​of the temperature change rates for all periods.

[0019] The calculation method for the contribution of gas concentration field leakage is as follows: For each receiving point that detects SF6, calculate the ratio of the penetration time of the receiving point to the minimum penetration time among all receiving points, multiply it by the ratio of the peak concentration of the receiving point to the maximum peak concentration among all receiving points, multiply it by the Gaussian spatial attenuation weight from the receiving point to the center of the goaf, and take the average over all receiving points.

[0020] The contribution of surface fissures to air leakage is calculated as follows: only effective through-flow fissures confirmed by the SF6 tracer gas method are counted. For each effective fissure, the ratio of fissure width to width benchmark threshold, the reciprocal of the ratio of fissure penetration time to minimum penetration time, and the exponential decay value of the distance from the fissure to the working face are calculated. The three items are multiplied together and averaged over all effective through-flow fissures.

[0021] Furthermore, in step S4, the confidence level of the air leakage channel identification is calculated as follows: the air leakage intensity factor of all grid cells is normalized by minimum and maximum, and only grid cells that exceed the judgment threshold are counted. The normalized values ​​are summed and then divided by the total number of grid cells.

[0022] Further, in step S5, the amplitude improvement factor is 1 minus the ratio of the average leakage intensity factor after sealing to the average leakage intensity factor before sealing; the trend stability factor is obtained by exponentially decaying the positive part of the linear regression slope of the leakage intensity factor after sealing with respect to time, and is penalized only when the slope is positive, and not penalized when the slope is negative or zero; the sealing efficiency coefficient is the product of the amplitude improvement factor and the trend stability factor.

[0023] Further, in step S6, the leakage intensity improvement rate is equal to the sealing efficiency coefficient; the tracer gas concentration improvement rate is equal to 1 minus the ratio of the average SF6 concentration after sealing to the average SF6 concentration before sealing; the temperature improvement rate is 1 minus the ratio of the heating rate after sealing to the heating rate before sealing when the heating rate before sealing exceeds the effective threshold, and takes the default value of 1 when the heating rate before sealing does not exceed the effective threshold; the pressure equalization improvement rate is equal to 1 minus the ratio of the response lag time before sealing to the response lag time after sealing.

[0024] Furthermore, in step S6, the channel sealing effectiveness index is the product of the leakage intensity improvement rate and the tracer gas concentration improvement rate. When both the leakage intensity improvement rate and the tracer gas concentration improvement rate are synchronized, the channel sealing effectiveness index is high. When the leakage intensity improvement rate is high but the tracer gas concentration improvement rate is low, the channel sealing effectiveness index is low to penalize the situation where the sealing position deviates from the main tracer gas channel. The environmental recovery effectiveness index is the product of the temperature improvement rate and the pressure balance improvement rate. When both are improved simultaneously, the environmental recovery effectiveness index is high. The comprehensive sealing effect score is the product of the channel sealing effectiveness index and the environmental recovery effectiveness index.

[0025] A data fusion-based goaf leakage identification and sealing assessment system is provided to implement any of the data fusion-based goaf leakage identification and sealing assessment methods described above, including:

[0026] The multi-source sensing monitoring subsystem is used to deploy pressure sensor arrays, temperature sensor arrays, gas concentration sensor arrays, and wind speed sensor arrays in the goaf and surrounding roadways to collect pressure field data, temperature field data, gas concentration field data, and wind speed field data in the goaf in real time.

[0027] The surface fissure dynamic monitoring module is used to obtain the width of the surface fissures and their distance to the working face, and to screen effective through fissures based on the SF6 tracer gas detection results.

[0028] The data acquisition and transmission module is connected to the multi-source sensing and monitoring subsystem. It is used to amplify, filter, and perform analog-to-digital conversion on the signals from each sensor, and transmit the acquired data to the data processing center via wired or wireless network.

[0029] The data fusion processing module, connected to the data acquisition and transmission module, is used to perform time synchronization, spatial registration, outlier removal and normalization on the received multi-source heterogeneous data, and to build a unified multi-physics data fusion framework.

[0030] The air leakage identification module is connected to the data fusion processing module. It is used to calculate the air leakage intensity factor of each grid unit in the goaf based on the fused multi-physics field data, and to identify the air leakage area and calculate the confidence level of the air leakage channel identification based on the air leakage intensity factor and the preset threshold.

[0031] The sealing assessment module, connected to the air leakage identification module, is used to collect multiphysics data after sealing measures are implemented, and to calculate the sealing efficiency coefficient and the comprehensive score of sealing effect based on the data before and after sealing.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] 1. This invention fuses the information from five dimensions—pressure field, temperature field, gas concentration field, wind speed field, and surface fissure—by maximizing the maximum value. It can simultaneously utilize multi-physical field information for cross-validation, avoiding misjudgment and omission caused by judging a single physical quantity. This makes the positioning accuracy of the air leakage area significantly better than the traditional single-index threshold judgment method.

[0034] 2. This invention defines four indicators: leakage intensity improvement rate, tracer gas concentration improvement rate, temperature improvement rate, and pressure balance improvement rate. It uses a pairwise coupled product structure to obtain a comprehensive score, which improves the sealing effect from a qualitative description to a quantitative value, providing a quantitative basis for the selection and optimization of sealing schemes.

[0035] 3. This invention automatically distinguishes between centralized and decentralized air leakage by recognizing confidence levels, and outputs differentiated treatment suggestions for local fixed-point sealing and large-area sealing respectively, reducing the subjectivity and lag of manual judgment.

[0036] 4. This invention eliminates the influence of seasonal temperature changes on temperature assessment in the temperature improvement rate; and eliminates the influence of changes in ventilation system operating conditions on pressure assessment in the pressure balance improvement rate, making the assessment results more accurate and reliable.

[0037] 5. The channel sealing effectiveness index of the present invention couples the improvement of air leakage intensity with the improvement of tracer gas concentration. When the air leakage intensity decreases but the tracer gas concentration does not decrease synchronously, the comprehensive score automatically decreases, which can effectively identify the situation of incorrect sealing location and guide the correction of subsequent sealing schemes. Attached Figure Description

[0038] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0039] Figure 1 This is a flowchart illustrating an embodiment of the present invention;

[0040] Figure 2 This is a system schematic diagram according to an embodiment of the present invention;

[0041] Figure 3 This is a flowchart illustrating the identification confidence decision process according to an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] like Figure 1 As shown, the data fusion-based method for identifying and assessing air leakage in goaf areas includes the following steps:

[0044] Step S1: Deploy a multi-source sensor network in the goaf and surrounding roadways to collect pressure field data, temperature field data, gas concentration field data and wind speed field data of the goaf. At the same time, release SF6 tracer gas in the goaf and detect the concentration and first arrival time at each receiving point to obtain the width of the surface fissures and the distance to the working face.

[0045] Step S2: Perform time synchronization, spatial registration, outlier removal and normalization on the collected multi-source data to construct a unified multi-physics data fusion framework;

[0046] Step S3: Based on the fused multiphysics data, the goaf is divided into several grid units. Five components are calculated for each grid unit: pressure field leakage contribution, temperature field leakage contribution, gas concentration field leakage contribution, wind speed field leakage contribution, and surface fissure leakage contribution. The maximum value after adjustment by the adjustment coefficient is taken as the leakage intensity factor of the grid unit.

[0047] Step S4: Compare the leakage intensity factor of each grid cell with the preset leakage intensity judgment threshold to identify the leakage area and calculate the identification confidence of the leakage channel.

[0048] Step S5: After implementing the sealing measures, collect multiphysics data after sealing, calculate the leakage intensity factor of each grid after sealing, calculate the amplitude improvement factor based on the average leakage intensity factor before and after sealing, calculate the trend stability factor based on the time series slope of the leakage intensity factor after sealing, and multiply the amplitude improvement factor and the trend stability factor to obtain the sealing efficiency coefficient.

[0049] Step S6: Calculate the comprehensive score of the sealing effect. Define four improvement rates: leakage intensity improvement rate, tracer gas concentration improvement rate, temperature improvement rate, and pressure equalization improvement rate. Couple the leakage intensity improvement rate with the tracer gas concentration improvement rate to obtain the channel sealing effectiveness index. Couple the temperature improvement rate with the pressure equalization improvement rate to obtain the environmental recovery effectiveness index. Multiply the channel sealing effectiveness index with the environmental recovery effectiveness index to obtain the comprehensive score of the sealing effect.

[0050] To achieve comprehensive monitoring of air leakage in goaf areas, a multi-layered monitoring network covering pressure, temperature, gas concentration, wind speed, and surface fissures needs to be established.

[0051] Pressure sensors are deployed at key locations within the goaf, the intake and return air corners, and the intake and return air roadways; temperature sensors are deployed within the goaf and the return air corners; oxygen concentration sensors and SF6 tracer gas concentration sensors are deployed within the goaf, the return air corners, and at surface fissures; wind speed sensors, using ultrasonic anemometers or mechanical wind gauges, are deployed in the intake and return air roadways, installed at the center of the roadway cross-section to measure the average wind speed; barometers are also deployed at corresponding surface locations; all sensors are synchronized using a unified clock, with a sampling frequency of no less than 1Hz.

[0052] SF6 release devices are installed on the intake side or inlet corner of the goaf. SF6 gas is stably released through a steel pipe extending into the goaf, with the release flow rate controlled within the range of 0.5~2L / min by a mass flow controller. Gas receiving and sampling points are set up at surface fracture zones, return air corners, and the sealed walls of the overlying goaf. Gas samples are collected periodically using vacuum sampling bags and brought back to the ground for SF6 concentration analysis using gas chromatography. The time difference from the start of SF6 release to the first detection of SF6 at each receiving point is recorded as the SF6 penetration time. The shorter the penetration time, the more unobstructed the fracture channel. If SF6 gas is detected at a receiving point, the corresponding surface fracture is determined to be a valid through fracture. If no SF6 is detected, it is determined to be an invalid fracture and is not included in subsequent statistics.

[0053] Drones were used to conduct full-coverage aerial photography of the corresponding surface of the goaf area to identify the distribution and orientation of surface fissures. Ground-based measurement methods were used to measure each fissure identified by the drone aerial photography, and the width of each fissure and the horizontal distance from the fissure to the working face boundary were recorded.

[0054] The raw data collected by the sensors comes from different types of devices with different sampling frequencies and different spatial locations. They differ in time reference, spatial coordinate system and physical dimensions, and need to be preprocessed and fused by the system before entering the air leakage identification process.

[0055] Time synchronization processing unifies sensor data from different sampling frequencies onto the same time reference using a linear interpolation algorithm; spatial registration processing maps sensor data from different spatial locations to a unified three-dimensional grid coordinate system of the goaf area. A spatial rectangular coordinate system is established with the geometric center of the goaf area as the origin. The goaf area is divided into several uniformly sized grid units in the horizontal and vertical directions. Each grid unit is 5m×5m×2m in size. The inverse distance weighting method is used to interpolate the data from surrounding sensors to the center point of each grid; outlier removal processing uses the Laida criterion to calculate the mean and standard deviation of the historical data sequence of each sensor. When a data point deviates from the mean by more than 3 times the standard deviation, it is judged as an outlier and removed, and replaced with linear interpolation of the data from the previous and next time times; data normalization processing transforms sensor data of different dimensions to the [0,1] interval using the min-max normalization method to eliminate the influence of dimensional differences on the fusion results.

[0056] The goaf is divided into M grid cells, and the air leakage intensity factor of each grid cell is taken as the maximum value of its five components. The specific formula is as follows:

[0057]

[0058] in, This represents the air leakage intensity factor of a grid cell in the goaf. A larger value indicates more severe air leakage at that location. These represent the contributions of air leakage from the pressure field, temperature field, gas concentration field, wind speed field, and surface fissures, respectively. These represent the adjustment coefficients for the five components, all positive numbers, used to adjust each component to the same order of magnitude for comparison. The calibration method involves taking historical data for each component under normal mine ventilation conditions for seven consecutive days without any abnormalities, calculating the 95th percentile of each component under normal conditions, and then taking the reciprocal. For example, under normal conditions... The 95th percentile is 0.35, then Take 1 / 0.35 ≈ 2.86;

[0059] The specific calculation methods for each fusion component are as follows:

[0060] 1. Contribution of air leakage in the pressure field

[0061] Shallowly buried goaf areas are affected by diurnal variations in surface air pressure, resulting in a periodic breathing effect between the goaf area and the surface. This component characterizes the conductivity of air leakage channels by analyzing the degree of lag in goaf pressure response to surface air pressure changes. The specific formula is as follows:

[0062]

[0063] in, This represents the contribution of air leakage to the pressure field, with a value range of [0, 1]. These represent the start and end times of the data window participating in the cross-correlation analysis, respectively, taking data from the most recent hour. The average pressure inside the goaf at time t is obtained by averaging the data from various pressure sensors within the goaf. Indicates time The surface air pressure is obtained by collecting data from a surface barometer. This represents the lag time variable, ranging from 0 to 3600 s, with a step size of 10 s. It iterates through all values ​​within this range, finding the value that maximizes the entire formula on the right side of the equation. The value, the maximum value of the formula is. For example, if If the formula on the right side reaches its maximum value at 120 seconds, then the response lag time is 120 seconds, indicating that the pressure change in the goaf lags behind the surface air pressure change by 120 seconds. If the formula value on the right side is 0.82, then... Take 0.82;

[0064] 2. Contribution of air leakage in the temperature field

[0065] Leaking air carries oxygen into the goaf, promoting the oxidation and release of heat in residual coal. A typical characteristic is a decrease in oxygen concentration accompanied by a rise in temperature. This component characterizes the activity level of air leakage oxidation by statistically analyzing the proportion of simultaneous decreases in oxygen concentration and increases in temperature. The specific formula is as follows:

[0066]

[0067] in, This represents the contribution of air leakage to the temperature field, with a value range of [0, 1]. This indicates the total number of temperature sensors. Indicates the number index of the temperature sensor. This represents the temperature at the location of the nth sensor. This represents the oxygen concentration at the nth sensor location. This represents an indicator function, which takes the value 1 if the condition within the parentheses is true, and 0 otherwise. To represent a very small positive number, take 10. -6 This is used to prevent the denominator from being zero. The numerator is the sum of the absolute values ​​of the rate of temperature change of all temperature sensors during the period when the oxygen concentration decreases and the temperature increases, and the denominator is the sum of the absolute values ​​of the rate of temperature change of all time periods. The larger the value, the higher the proportion of simultaneous temperature rise and oxygen concentration decrease, meaning the more active the air leakage oxidation.

[0068] 3. Contribution of air leakage in the gas concentration field

[0069] SF6 tracer gas migrates from the release point to each receiving point. The shorter the penetration time and the higher the peak concentration, the more unobstructed the leakage channel in that direction. The specific formula is as follows:

[0070]

[0071] in, This represents the contribution of air leakage to the gas concentration field, with a value range of [0, 1]. This represents the total number of gas receiving points where SF6 was detected; receiving points where SF6 was not detected are not included in the statistics. This indicates that the receiver point number index of SF6 has been detected. Indicates the first The SF6 penetration time at a receiving point is the time difference between the start of SF6 release and the first detection of SF6 at that receiving point. A shorter penetration time indicates a more unobstructed leakage path. This represents the minimum penetration time among all receivers that detected SF6, indicating the fastest leakage path. Indicates the first The peak SF6 concentration detected at each receiving point indicates that the higher the peak concentration, the greater the air leakage in that channel. This represents the maximum peak concentration of SF6 across all receiving points. Indicates the first The Euclidean distance from each receiving point to the center of the goaf area. The parameter representing the spatial attenuation scale of gas diffusion is calibrated from field measurement data and is usually taken as 50m.

[0072] 4. Contribution of air leakage in the wind speed field

[0073] The instantaneous difference between the inlet and return air volumes fluctuates significantly. The total air leakage is characterized by the cumulative trend of the inlet and return air volume difference over a recent period. The specific formula is as follows:

[0074]

[0075] in, This indicates the contribution of air leakage in the wind speed field. This indicates the length of the sliding time window, set to 3600 seconds. Represents the time integral variable, Indicates time The measured air volume of the intake airway. Indicates time The measured air volume in the return airway. This represents the reference air volume for the intake airway under normal ventilation conditions, calculated as the average intake air volume over seven consecutive days of normal ventilation. This represents the reference air volume of the return airway under normal ventilation conditions, and is taken as the average return air volume during 7 consecutive days of normal ventilation.

[0076] 5. Contribution of air leakage from surface fissures

[0077] Two fractures with the same width may have different actual conductivity due to differences in the degree of filling and tortuosity within the fracture. The SF6 penetration time comprehensively reflects the actual conductivity of the fracture, and the specific formula is as follows:

[0078]

[0079] in, This indicates the contribution of air leakage from surface fissures. This represents the total number of valid through fractures confirmed by the SF6 tracer gas method. Indicates the index of the effective fracture number. This represents the measured width of the s-th crack. The baseline threshold representing the crack width is the statistical maximum value of all valid through crack widths. This represents the minimum SF6 penetration time across all effectively penetrating fractures. This represents the SF6 penetration time at the s-th fracture. This represents the horizontal distance from the s-th crack to the working face. The attenuation characteristic length is determined by exponential fitting from field data and is generally taken as 30~80m.

[0080] Each grid cell has a corresponding air leakage intensity factor. First, the number of grid cells whose air leakage intensity factor exceeds the judgment threshold is counted. If no grid cell exceeds the threshold, it is determined that there is no air leakage in the goaf, no sealing measures are triggered, and the system continues to maintain the monitoring status.

[0081] If there are grid cells exceeding the threshold, calculate the confidence score for identifying the air leakage channel using the following formula:

[0082]

[0083] in, This indicates the confidence level in identifying the air leakage channel. This indicates the total number of grid cells in the goaf area. Indicates the index of the grid cell. This represents the air leakage intensity factor of the k-th grid cell. This represents the minimum value of the air leakage intensity factor among all grid cells. This represents the maximum value of the air leakage intensity factor across all grid cells. The threshold for determining the air leakage intensity factor is 1.5 times the 95th percentile of the air leakage intensity factor for each grid under normal conditions, and is determined statistically from historical data. The condition in parentheses is met when... The value is 1 if it is not 1, otherwise it is 0.

[0084] when When the value is greater than or equal to 0.6, the leakage area is highly concentrated, and a few grid cells contribute the majority of the leakage intensity, indicating that the leakage channel is a concentrated channel. At this time, the leakage channel is clearly identified and the identification confidence is high. It is recommended to adopt a local fixed-point sealing strategy and output the geometric center coordinates of the leakage grid as the sealing target location.

[0085] when When the value is less than 0.6, the leakage intensity is evenly distributed across multiple grid cells, indicating that the leakage channels are scattered or the grid division is too fine. At this time, the adaptive re-identification process is triggered, merging the original grid into larger grid cells and re-executing steps S3 and S4 to determine whether there are large-scale leakage areas that are subdivided and covered. If it is still less than 0.6, it is determined to be a scattered leakage, and it is recommended to adopt a large-area leakage plugging strategy. It is also suggested that the sensor deployment needs to be densified, because scattered leakage means that there may be missed areas in the current monitoring network.

[0086] After implementing the sealing measures, multiphysics data were collected to calculate the air leakage intensity factor of each grid after sealing, and the sealing efficiency coefficient was calculated based on the time series change of the air leakage intensity factor before and after sealing.

[0087] The specific formulas for the average leakage intensity factor before and after sealing are as follows:

[0088]

[0089] in, This represents the average air leakage intensity factor over a period of time prior to the sealing. This indicates the length of the assessment window before the blockade, taken as 72 hours. This indicates the start time of the pre-closure assessment window, taking the starting point as 72 hours before the closure implementation. Indicates the time when the blockade was implemented. This represents the air leakage intensity factor as a function of time t. This represents the average air leakage intensity factor over a period of time after sealing. This indicates the length of the assessment window after the blockade, taken as 72 hours. This indicates the end time of the post-blocking assessment window, which is the end point 72 hours after the blockade is implemented;

[0090] The trend stability factor characterizes whether the leakage intensity remains stable without rebounding after sealing. Penalty is applied only when the leakage intensity shows an upward trend. The specific formula is as follows:

[0091]

[0092] in, This represents the trend stability factor of the air leakage intensity factor after sealing. This represents the slope obtained by linear regression of the leakage intensity factor over time within the evaluation window after sealing. A positive slope indicates that the leakage intensity increases over time, and a penalty is only imposed when the leakage intensity shows an upward trend. If the slope is negative or zero, no penalty is imposed.

[0093] The formula for calculating the blocking efficiency coefficient is:

[0094]

[0095] in, This represents the blocking efficiency coefficient; a value closer to 1 indicates a better and more stable blocking effect.

[0096] A sealing efficiency coefficient of 0.9 indicates that the average air leakage has decreased by 90% after sealing, and the trend is stable, indicating excellent sealing effect; a sealing efficiency coefficient of 0.6 indicates that the average air leakage has decreased by 60% after sealing, and the trend is stable, indicating good sealing effect; a sealing efficiency coefficient of 0.7 but a trend stability factor of 0.2 indicates that the average leakage has decreased by 70%, but the steep slope in the later stage indicates that it is rebounding rapidly, and the overall efficiency is reduced to 0.14, requiring secondary sealing.

[0097] The comprehensive evaluation of plugging effectiveness is based on the plugging efficiency coefficient, further integrating the improvement in three dimensions: tracer gas concentration, temperature, and pressure. This comprehensive evaluation quantifies the overall effectiveness of plugging measures, defining four improvement rates as follows:

[0098] Air leakage intensity improvement rate ,in Indicates the improvement rate of air leakage intensity;

[0099] The specific formula for the improvement rate of tracer gas concentration is:

[0100]

[0101] in, Indicates the improvement rate of tracer gas concentration. This indicates the average SF6 concentration at each receiving point within 72 hours prior to the sealing operation. This indicates the average SF6 concentration at each receiving point within 72 hours after the sealing.

[0102] The specific formula for the temperature improvement rate is:

[0103]

[0104] in, This indicates the temperature improvement rate; a higher value indicates a better effect of the sealing on eliminating temperature anomalies in the goaf. This represents the absolute value of the average temperature rise rate in the goaf within the 72 hours prior to sealing, obtained by taking the absolute value of the slope of a linear regression of temperature data over time within the pre-sealing assessment window. This represents the absolute value of the average temperature rise rate in the goaf within 72 hours after sealing, obtained by taking the absolute value of the slope of a linear regression of temperature data over time within the post-sealing evaluation window. The effective threshold for the heating rate is 0.001 K / min. When the heating rate before sealing is lower than this threshold, it is considered that there is no temperature anomaly in the goaf, and the temperature improvement rate is set to the default value of 1. Conversely, if the leakage of air and oxygen supply before sealing leads to the oxidation and heat release of residual coal, the temperature will show an upward trend. At this time, the percentage decrease in the heating rate is calculated as the improvement rate. This component evaluates the sealing effect by the rate of temperature change rather than the absolute value of temperature, which can effectively eliminate the interference of slow seasonal changes in ambient temperature.

[0105] The specific formula for the pressure balance improvement rate is:

[0106]

[0107] in, Indicates the rate of improvement in pressure balance. This represents the lag time of the pressure in the goaf to changes in surface air pressure before sealing, calculated through cross-correlation analysis. This indicates the response lag time of the goaf pressure to changes in surface air pressure after sealing. Before sealing, the air leakage channels are unobstructed, and the goaf pressure changes quickly with the surface air pressure, resulting in a short lag time. After sealing, the channels are blocked, and the goaf pressure is less affected by surface air pressure, resulting in a longer lag time. The greater the increase in lag time, the better the sealing effect on cutting off the pressure transmission channels.

[0108] The specific formula for the coupled comprehensive score is as follows:

[0109]

[0110] in, This represents the overall score for the sealing effect, with a value ranging from [0, 1]. The closer the score is to 1, the more comprehensive and effective the sealing effect. The coupling index is represented by a sum of 2, usually taken as 1. Improvements in leakage intensity and tracer gas concentration originate from the same physical process: the leakage channel is blocked, and both should change synchronously. If the leakage intensity decreases significantly but the tracer gas concentration does not, it indicates that although the sealing measures reduced leakage in a certain area, the SF6 tracer gas can still find other unblocked paths to reach the receiving point, meaning the sealing did not hit the main channel. A small product of this factor lowers the overall score. Temperature improvement indicates reduced oxygen supply from leakage and weakened oxidation of residual coal, a direct effect of the sealing. Improved pressure response indicates that changes in surface air pressure can no longer be transmitted to the goaf through the leakage channel, cutting off the leakage driving force, the fundamental reason for the sealing. If only temperature improves but pressure response does not, it may be a false effect caused by seasonal cooling, and the sealing has not truly cut off the leakage driving force. If only pressure response improves but temperature does not, it indicates that although the leakage driving force has been cut off, oxidation in the goaf continues, and the sealing effect needs verification. For example... A value of 0.85 indicates that the blockade has hit the main channel and has addressed both the symptoms and the root cause, resulting in an excellent blocking effect. A value of 0.40 indicates that although the average air leakage has decreased, the blockage location was not correct, or although the location was correct, the pressure source was not cut off, and the sealing effect was average. A value close to 0 indicates that the sealing is ineffective, the wrong location was sealed, or there is air leakage and rebound after sealing, and the sealing plan needs to be re-evaluated.

[0111] like Figure 2 As shown, the goaf leakage identification and sealing assessment system based on data fusion is used to implement any of the goaf leakage identification and sealing assessment methods based on data fusion, including:

[0112] The multi-source sensing monitoring subsystem is used to deploy pressure sensor arrays, temperature sensor arrays, gas concentration sensor arrays, and wind speed sensor arrays in the goaf and surrounding roadways to collect pressure field data, temperature field data, gas concentration field data, and wind speed field data in the goaf in real time.

[0113] The surface fissure dynamic monitoring module is used to obtain the width of the surface fissures and their distance to the working face, and to screen effective through fissures based on the SF6 tracer gas detection results.

[0114] The data acquisition and transmission module is connected to the multi-source sensing and monitoring subsystem. It is used to amplify, filter, and perform analog-to-digital conversion on the signals from each sensor, and transmit the acquired data to the data processing center via wired or wireless network.

[0115] The data fusion processing module, connected to the data acquisition and transmission module, is used to perform time synchronization, spatial registration, outlier removal and normalization on the received multi-source heterogeneous data, and to build a unified multi-physics data fusion framework.

[0116] The air leakage identification module is connected to the data fusion processing module. It is used to calculate the air leakage intensity factor of each grid unit in the goaf based on the fused multi-physics field data, and to identify the air leakage area and calculate the confidence level of the air leakage channel identification based on the air leakage intensity factor and the preset threshold.

[0117] The sealing assessment module, connected to the air leakage identification module, is used to collect multiphysics data after sealing measures are implemented, and to calculate the sealing efficiency coefficient and the comprehensive score of sealing effect based on the data before and after sealing.

[0118] like Figure 3 As shown, after the air leakage identification begins, the number of grid cells whose air leakage intensity factor exceeds the judgment threshold is counted first. If no grid cell exceeds the threshold, it is determined that there is no air leakage in the goaf, and the system continues to monitor without triggering any sealing measures.

[0119] If there are grid cells exceeding the threshold, the confidence score for identifying the leakage channel is calculated. When the confidence score is ≥0.6, it is determined to be a centralized leakage mode, indicating that a few grid cells contribute the majority of the leakage intensity, and the leakage channel is clear and concentrated. In this case, the geometric center coordinates of the leakage grid are output as the blocking target location. When the confidence score is <0.6, it indicates that the leakage intensity is evenly distributed in multiple grid cells, which may be due to decentralized leakage or excessively fine grid division. In this case, the adaptive re-identification process is triggered: the original grid is merged into a larger grid cell and steps S3 and S4 are re-executed to calculate the confidence score. The merge re-identification is executed a maximum of 2 times. If the confidence score after merging is ≥0.6, it is treated as a centralized leakage. If the confidence score is still less than 0.6 after 2 executions, it is finally determined to be a decentralized leakage, and it is also indicated that there may be missed areas in the current monitoring network and that the deployment of encrypted sensors is required.

[0120] After the above decision is made, the process proceeds to step S5, the evaluation stage of the blocking effect.

[0121] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A method for identifying and assessing air leakage in goaf areas based on data fusion, characterized in that, Includes the following steps: Step S1: Deploy a multi-source sensor network in the goaf and surrounding roadways to collect pressure field data, temperature field data, gas concentration field data and wind speed field data of the goaf. At the same time, release SF6 tracer gas in the goaf and detect the concentration and first arrival time at each receiving point to obtain the width of the surface fissures and the distance to the working face. Step S2: Perform time synchronization, spatial registration, outlier removal and normalization on the collected multi-source data to construct a unified multi-physics data fusion framework; Step S3: Based on the fused multiphysics data, the goaf is divided into several grid units. Five components are calculated for each grid unit: pressure field leakage contribution, temperature field leakage contribution, gas concentration field leakage contribution, wind speed field leakage contribution, and surface fissure leakage contribution. The maximum value after adjustment by the adjustment coefficient is taken as the leakage intensity factor of the grid unit. Step S4: Compare the leakage intensity factor of each grid cell with the preset leakage intensity judgment threshold to identify the leakage area and calculate the identification confidence of the leakage channel. Step S5: After implementing the sealing measures, collect multiphysics data after sealing, calculate the leakage intensity factor of each grid after sealing, calculate the amplitude improvement factor based on the average leakage intensity factor before and after sealing, calculate the trend stability factor based on the time series slope of the leakage intensity factor after sealing, and multiply the amplitude improvement factor and the trend stability factor to obtain the sealing efficiency coefficient. Step S6: Calculate the comprehensive score of the sealing effect. Define four improvement rates: leakage intensity improvement rate, tracer gas concentration improvement rate, temperature improvement rate, and pressure equalization improvement rate. Couple the leakage intensity improvement rate with the tracer gas concentration improvement rate to obtain the channel sealing effectiveness index. Couple the temperature improvement rate with the pressure equalization improvement rate to obtain the environmental recovery effectiveness index. Multiply the channel sealing effectiveness index with the environmental recovery effectiveness index to obtain the comprehensive score of the sealing effect.

2. The method according to claim 1, characterized in that, In step S1, pressure sensors are deployed inside the goaf, at the intake and return air corners, and in the intake and return airway. Temperature sensors are deployed inside the goaf and at the return air corner. Oxygen concentration sensors and SF6 tracer gas concentration sensors are deployed inside the goaf, at the return air corner, and at surface fissures. Wind speed sensors are deployed in the intake and return airway. Barometers are deployed at corresponding surface locations. An SF6 release device is installed on the intake side or intake corner of the goaf to stably release SF6 gas. Gas receiving and sampling points are set up in the surface fissure zone, at the return air corner, and at the sealed wall of the overlying goaf. If SF6 is detected at a receiving point, the fissure corresponding to that receiving point is determined to be an effective penetrating fissure. The time difference from the start of release to the first detection of SF6 is recorded as the SF6 penetration time.

3. The method according to claim 2, characterized in that, In step S2, time synchronization processing unifies sensor data with different sampling frequencies to the same time reference through a linear interpolation algorithm; spatial registration processing maps sensor data from different spatial locations to a unified three-dimensional grid coordinate system of the goaf area; outlier removal processing adopts the Laida criterion, and when a data point deviates from the mean by more than 3 times the standard deviation, it is judged as an outlier and removed; data normalization processing transforms sensor data with different dimensions to the same interval through the minimum-maximum normalization method.

4. The method according to claim 3, characterized in that, In step S3, the contribution of air leakage in the pressure field is obtained by normalized cross-correlation analysis of the pressure in the goaf and the surface air pressure, which characterizes the intensity of the breathing effect in the goaf; the contribution of air leakage in the temperature field is obtained by statistically analyzing the proportion of simultaneous decrease in oxygen concentration and increase in temperature, which characterizes the thermal response synchronization rate of the goaf; the contribution of air leakage in the gas concentration field is obtained by multiplying the SF6 penetration time ratio and the peak concentration ratio, which characterizes the dominance of the tracer gas arrival. The contribution of wind speed field leakage is obtained by the cumulative offset of the difference between inlet and outlet air volume within the sliding time window, which characterizes the intensity of the cumulative leakage trend; the contribution of surface fissure leakage is obtained by the product of fissure width ratio, SF6 penetration time ratio and distance attenuation factor, which characterizes the effective conduction index of fissures.

5. The method according to claim 4, characterized in that, In step S3, the calculation method for the air leakage contribution of the pressure field is as follows: normalized cross-correlation analysis is performed on the average pressure of the goaf and the surface air pressure, and the maximum absolute value of the cross-correlation function is taken as the air leakage contribution of the pressure field. The contribution of air leakage in the temperature field is calculated as follows: calculate the temperature change rate for each temperature sensor, and sum the absolute values ​​of the temperature change rates during the period when the oxygen concentration decreases and the temperature rises, and divide by the sum of the absolute values ​​of the temperature change rates for all periods. The calculation method for the contribution of gas concentration field leakage is as follows: For each receiving point that detects SF6, calculate the ratio of the penetration time of the receiving point to the minimum penetration time among all receiving points, multiply it by the ratio of the peak concentration of the receiving point to the maximum peak concentration among all receiving points, multiply it by the Gaussian spatial attenuation weight from the receiving point to the center of the goaf, and take the average over all receiving points. The contribution of surface fissures to air leakage is calculated as follows: only effective through-flow fissures confirmed by the SF6 tracer gas method are counted. For each effective fissure, the ratio of fissure width to width benchmark threshold, the reciprocal of the ratio of fissure penetration time to minimum penetration time, and the exponential decay value of the distance from the fissure to the working face are calculated. The three items are multiplied together and averaged over all effective through-flow fissures.

6. The method according to claim 5, characterized in that, In step S4, the confidence level of the air leakage channel identification is calculated as follows: the air leakage intensity factor of all grid cells is normalized by minimum and maximum, and only grid cells that exceed the judgment threshold are counted. The normalized values ​​are summed and then divided by the total number of grid cells.

7. The method according to claim 6, characterized in that, In step S5, the magnitude improvement factor is 1 minus the ratio of the average air leakage intensity factor after sealing to the average air leakage intensity factor before sealing. The trend stability factor is obtained by exponentially decaying the positive part of the linear regression slope of the leakage intensity factor over time after closure. It is penalized only when the slope is positive, and not penalized when the slope is negative or zero. The closure efficiency coefficient is the product of the amplitude improvement factor and the trend stability factor.

8. The method according to claim 7, characterized in that, In step S6, the air leakage intensity improvement rate is equal to the sealing efficiency coefficient; the tracer gas concentration improvement rate is equal to 1 minus the ratio of the average SF6 concentration after sealing to the average SF6 concentration before sealing; the temperature improvement rate is 1 minus the ratio of the heating rate after sealing to the heating rate before sealing when the heating rate before sealing exceeds the effective threshold, and takes the default value of 1 when the heating rate before sealing does not exceed the effective threshold; the pressure equalization improvement rate is equal to 1 minus the ratio of the response lag time before sealing to the response lag time after sealing.

9. The method according to claim 8, characterized in that, In step S6, the channel sealing effectiveness index is the product of the leakage intensity improvement rate and the tracer gas concentration improvement rate. When the leakage intensity improvement rate and the tracer gas concentration improvement rate are synchronized, the channel sealing effectiveness index is high. When the leakage intensity improvement rate is high but the tracer gas concentration improvement rate is low, the channel sealing effectiveness index is low to penalize the situation where the sealing position deviates from the main tracer gas channel. The environmental recovery effectiveness index is the product of the temperature improvement rate and the pressure balance improvement rate. When both are improved simultaneously, the environmental recovery effectiveness index is high. The comprehensive sealing effect score is the product of the channel sealing effectiveness index and the environmental recovery effectiveness index.

10. A data fusion-based goaf leakage identification and sealing assessment system, used to implement the data fusion-based goaf leakage identification and sealing assessment method as described in any one of claims 1-9, characterized in that, include: The multi-source sensing monitoring subsystem is used to deploy pressure sensor arrays, temperature sensor arrays, gas concentration sensor arrays, and wind speed sensor arrays in the goaf and surrounding roadways to collect pressure field data, temperature field data, gas concentration field data, and wind speed field data in the goaf in real time. The surface fissure dynamic monitoring module is used to obtain the width of the surface fissures and their distance to the working face, and to screen effective through fissures based on the SF6 tracer gas detection results. The data acquisition and transmission module is connected to the multi-source sensing and monitoring subsystem. It is used to amplify, filter, and perform analog-to-digital conversion on the signals from each sensor, and transmit the acquired data to the data processing center via wired or wireless network. The data fusion processing module, connected to the data acquisition and transmission module, is used to perform time synchronization, spatial registration, outlier removal and normalization on the received multi-source heterogeneous data, and to build a unified multi-physics data fusion framework. The air leakage identification module is connected to the data fusion processing module. It is used to calculate the air leakage intensity factor of each grid unit in the goaf based on the fused multi-physics field data, and to identify the air leakage area and calculate the confidence level of the air leakage channel identification based on the air leakage intensity factor and the preset threshold. The sealing assessment module, connected to the air leakage identification module, is used to collect multiphysics data after sealing measures are implemented, and to calculate the sealing efficiency coefficient and the comprehensive score of sealing effect based on the data before and after sealing.