CN113586157B - Extraction working face salient dangerous area rapid division method based on Kriging interpolation - Google Patents

Abstract

The invention provides a quick division method for a critical zone of a stope face based on Kerling interpolation. According to the method, a microseismic monitoring technology is applied to a coal and gas outburst mine, a coal and gas outburst mechanism and geostatistics are used as theoretical bases, a GIS technology and a Kriging algorithm are used as research tools, dynamic evaluation based on microseismic monitoring data is established, the problems that the existing coal and gas outburst prediction method is large in drilling engineering quantity, large in human factor interference, production is influenced to a certain extent generally, continuous monitoring cannot be achieved in a time domain, point evaluation forms are reflected in an airspace, regional coal and rock stress environments in a mining disturbance process and the problem of dynamic evolution process of coal and gas outburst danger are difficult to reflect, and the visualization of dangerous grade of an outburst danger area of a stoping working face is achieved.

Description

Quick division method of mining face outburst danger zone based on Kriging interpolation

Technical field

The invention relates to the technical field of safety monitoring and early warning of underground coal mine dynamic disasters, and specifically relates to a method for rapid division of mining face outburst dangerous areas based on Kriging interpolation.

Background technique

In recent years, with the increasing depth and intensity of coal mining in my country, the threat of coal and gas outburst disasters has become increasingly serious. As one of the most destructive and harmful dynamic disasters in coal mining, coal and gas outbursts have seriously threatened the lives of workers in the coal mining process. Therefore, effective and reasonable predictions of coal and gas outbursts are guaranteed. Necessary measures for the normal operation of the extraction process.

As a key part of coal and gas outburst prediction, regional prediction is more controllable than other prediction methods in terms of prediction range. The existing regional outburst risk prediction methods mainly include: single index method, outburst prediction comprehensive index method, comprehensive geological index method, gas geological regional prediction and gas geological unit method, etc.; the existing working face prediction methods mainly include: drill cuttings Index method, R index method and composite index method, etc. Most of the above methods use random inspections and fixed-point indicators, which require a large amount of drilling work, large interference from human factors, and usually affect production to a certain extent; continuous monitoring cannot be achieved in the time domain, and it is reflected in the form of point evaluation in the airspace, which is difficult to reflect. The regional coal and rock stress environment during the mining disturbance process and the dangerous dynamic evolution process of coal and gas outbursts.

In the research process of non-contact continuous outburst prediction technology, domestic and foreign research directions mainly focus on three aspects: microseismic outburst prediction technology, electromagnetic radiation outburst prediction technology and gas gush dynamic characteristics outburst prediction technology. Among the three non-contact continuous outburst prediction methods, microseismic (MS) is the vibration caused by the rupture of coal and rock masses. Microseismic monitoring technology has achieved a series of achievements in the theory and technical application of rock burst monitoring and early warning. Previous studies have revealed the spatial distribution characteristics and evolution process of microseismic activity during the occurrence of rock bursts and rock bursts, and divided them accordingly. Dangerous zone. Based on the previous foundation, how to innovatively apply microseismic monitoring technology to coal and gas outburst mines and establish dynamic evaluation based on microseismic data is of great significance for the rapid division of outburst dangerous areas on the working face.

Contents of the invention

In view of the existing prediction methods for coal and gas outbursts, there are large drilling projects, large interference from human factors, and usually affect production to a certain extent. Continuous monitoring cannot be achieved in the time domain, and it is reflected in the form of point evaluation in the airspace, which is difficult to reflect. Technical issues related to the regional coal and rock stress environment during the mining disturbance process and the dynamic evolution process of coal and gas outburst dangers. The present invention provides a method for quickly dividing the outburst danger zone of the mining working face based on Kriging interpolation.

In order to solve the above technical problems, the present invention adopts the following technical solutions:

A quick division method for outburst dangerous areas in mining working faces based on Kriging interpolation includes the following steps:

S1. Microseismic data monitoring and preprocessing: Microseismic monitoring data from the mining working face are collected through the data monitoring system, and the Holt index smoothing method is used to complete the data processing of the microseismic monitoring data to accurately eliminate the data generated by the mining construction in the mining working face. The interference anomaly data is used to obtain mine safety monitoring data with time series attributes. The microseismic data monitoring includes the spatial distribution characteristics of microseismic events and the time series change characteristics of microseismic indicators;

S2. Quantification of microseismic monitoring data: Based on the probability characteristics of the normal distribution function, establish a normal distribution multi-level early warning outburst model, perform quantitative processing on the microseismic monitoring data, and calculate the outburst risk grade division value f of all microseismic monitoring data. And use the scores to quantitatively divide each protrusion risk grade division value f. The discrimination formula of the protrusion risk grade division value f is as follows:

in,

In the formula, x represents the random error risk discrimination coefficient, m _i represents the microseismic parameter, represents the overall sample mean of microseismic parameters, and S _m is the standard deviation of microseismic parameters;

S3. Perform kriging interpolation on the quantified microseismic monitoring data: Use the built-in geostatistics module of ArcGIS to perform kriging interpolation on the quantified microseismic monitoring data to obtain the kriging estimate of any point to be estimated. The specific calculation formula is as follows:

In _{the formula ,} contribution;

S4. Division of outburst dangerous areas: The spatial distribution characteristics of microseismic monitoring data obtained after Kriging interpolation are input into the ArcGIS geographical data processing system to form a visual cloud map of the classification of coal and gas outburst dangerous areas in the mining working face.

Compared with the existing technology, the method for quickly dividing mining face outburst dangerous areas based on Kriging interpolation provided by the present invention has the following beneficial effects: 1). The coal and gas outburst microseismic monitoring system included in the present invention has the ability to automatically eliminate noise function, the Holt exponential smoothing method is used to complete the microseismic monitoring data, and the interference information generated by the mining construction in the mining working face is accurately eliminated, thereby greatly improving the monitoring accuracy of coal and rock fracture information; 2), This invention establishes dynamic coal and gas outburst continuous monitoring technology early warning indicators based on microseismic monitoring data, and realizes the function of non-contact real-time monitoring of the danger of coal and gas outbursts in mining working faces. The normal distribution multi-level early warning outburst model is adopted. The energy and time series change characteristics of microseismic events are mainly used as judgment indicators to ensure the accuracy and efficiency of early warning; 3). The method for quickly dividing the outburst dangerous area of the mining working face based on Kriging interpolation provided by the present invention solves the existing problem. The amount of technical drilling projects is large, the interference from human factors is large, and it affects production to a certain extent. It can dynamically analyze the stress field distribution characteristics near the geological structures before and during the mining process of the coal mining face, and highlight dangerous locations and The moment of occurrence facilitates risk analysis; 4), this method applies microseismic monitoring technology to coal and gas outburst mines, based on the theoretical basis of coal and gas outburst mechanism and geostatistics, and cooperates with GIS technology and Kriging algorithm to establish a The dynamic evaluation of microseismic data can realize the visualization of the danger level of the prominent danger zone in the mining working face.

Furthermore, the data monitoring system in step S1 includes a microseismic probe, a monitoring substation and a ground center station. The microseismic probe is installed in the tunnel to transmit monitoring data to the monitoring substation. The monitoring substation communicates through a network switch. Monitoring data is transmitted to the ground center station.

Furthermore, the specific principle formula of using the Holt exponential smoothing method to complete the data processing of microseismic monitoring data in step S1 is as follows:

In the formula, α and γ are smoothing parameters, X _t is the monitoring data at time t, S _t is the smooth value at time t after correction using the smooth value S _t-1 of the previous period and the trend value b _t-1 , b _t is the modified trend value at time t obtained by combining the two adjacent smoothing values with the previous period's trend value b _t-1 , and T is the number of prediction periods.

Further, in step S2, the risk of coal and gas outbursts is set to four levels: safe, mild, moderate and severe according to the outburst risk level division value f. The scores F for each level of high risk, moderate risk and severe risk are 0≤F<35, 35≤F<70, 70≤F<85 and F≥85 respectively.

Further, before performing Kriging interpolation in step S3, it also includes performing Log conversion on the quantified microseismic monitoring data to determine whether the quantified data obeys a normal distribution.

Further, in step S3, the weight coefficient λ _i is calculated by theoretical variograms including spherical models, Gaussian function models, exponential models and power function models.

Further, the weight coefficient λ _i is calculated through a Gaussian function model, and the Gaussian function model is as follows:

Substituting the calculated r value into the following formula to obtain the weight coefficient λ _i ,

In the formula, h is the interpolation distance, a is the variable range, c ₀ and c are the initial nugget value and nugget value respectively.

Description of the drawings

Figure 1 is a schematic flowchart of the method for quickly dividing the outburst dangerous area of a mining working face based on Kriging interpolation provided by the present invention.

Figure 2 is a schematic diagram of the coordinates of the microseismic signal and its relationship with the tunnel provided by the embodiment of the present invention.

Figure 3 is a normal distribution QQ normal diagram after Log transformation of quantified microseismic monitoring data provided by the embodiment of the present invention.

Figure 4 is a trend analysis diagram after Log transformation of quantified microseismic monitoring data provided by the embodiment of the present invention.

Figure 5 is a visual cloud diagram of the classification of dangerous areas of coal and gas outbursts provided by the embodiment of the present invention.

Detailed ways

In order to make it easy to understand the technical means, creative features, objectives and effects of the present invention, the present invention will be further explained below in conjunction with specific illustrations.

Please refer to Figures 1 to 5. The method for quickly dividing the outburst dangerous area of a mining working face based on Kriging interpolation provided by the present invention includes the following steps:

S1. Microseismic data monitoring and preprocessing: Microseismic monitoring data from the mining working face are collected through the data monitoring system, and the Holt index smoothing method is used to complete the data processing of the microseismic monitoring data to accurately eliminate the data generated by the mining construction in the mining working face. The interference anomaly data is used to obtain mine safety monitoring data with time series attributes. The microseismic data monitoring includes the spatial distribution characteristics of microseismic events and the time series change characteristics of microseismic indicators.

As a specific embodiment, the data monitoring system includes a microseismic probe, a monitoring substation and a ground center station. The microseismic probe is installed in the tunnel to transmit monitoring data to the monitoring substation. The monitoring substation communicates through a network switch. Monitoring data is transmitted to the ground center station. Those skilled in the art can also use other methods to implement the above-mentioned data monitoring system.

As a specific example, affected by complex underground factors such as temperature, dust, water vapor, and electromagnetic interference, microseismic monitoring data may cause data anomalies and missing data during the collection, transmission, and processing processes. For example, in addition to relatively strong random noise in microseismic monitoring records, there will also be strong energy low-frequency background noise, strong energy disturbance signals, wellbore waves, guided waves, etc. Therefore, in order to obtain monitoring data information that reflects the true situation of the mine, it is necessary to Preprocess mine monitoring data. Abnormal data is eliminated and processed, and based on the structural characteristics of monitoring data such as gas concentration and wind speed, the Holt exponential smoothing method is used for data completion processing. The specific technical principle formula is as follows:

In the formula, α _{and γ are} smoothing _parameters , , b _t is the modified trend value at time t obtained by combining the two adjacent smoothing values with the previous period's trend value b _t-1 , and T is the number of prediction periods.

S2. Quantification of microseismic monitoring data: Based on the research results on the disaster mechanism of coal and gas outbursts, determine the early warning indicators of dynamic coal and gas outburst continuous monitoring technology based on microseismic monitoring data. When mining outburst coal seams underground, the actual outburst area accounts for the entire 5% to 10% of the coal seam is highlighted, and the distribution of microseismic parameters is similar to the normal distribution function. According to the characteristics of the normal distribution, when the normal distribution interval is (μ-2σ, μ+2σ), the variable falls into this The probability of the interval is 95.45%. Based on this, the present invention establishes a normal distribution multi-level early warning outburst model based on the probability characteristics of the normal distribution function, performs quantitative processing on the microseismic monitoring data, calculates the outburst risk level division value f of all microseismic monitoring data, and uses the analysis Values are used to quantitatively divide each protrusion risk grade division value f. The discrimination formula of the protrusion risk grade division value f is as follows:

in,

In the formula, x represents the random error risk discrimination coefficient, m _i represents the microseismic parameter, represents the overall sample mean of microseismic parameters, and S _m is the standard deviation of microseismic parameters.

As a specific embodiment, according to the outburst hazard level division value f, the hazards of coal and gas outbursts are set as safe (I), mild hazard (II), moderate hazard (III) and severe hazard (IV). ) four levels. In order to be able to judge and quantitatively express the risk level, we choose to use scores to quantitatively divide each level of risk. Each level represents a domain value, such as safety, mild risk, and moderate risk. The scores F for each level of serious risk are 0≤F<35, 35≤F<70, 70≤F<85, and F≥85 respectively, as shown in Table 1 below.

Table 1 Quantitative classification of outstanding risks

As a specific example, according to the comprehensive daily report of microseismic monitoring of a certain coal mine working face, three days of microseismic monitoring data were selected for quantitative processing. A total of 571 microseismic signals were generated within three days. The coordinates of the microseismic signals and their relationship with the tunnel are as shown in the figure. 2 shown. To determine the outstanding comprehensive evaluation level of the energy in the microseismic monitoring data, first calculate the α value of all data, and then calculate the f value of all microseismic monitoring data based on the range of the α value.

S3. Perform Kriging interpolation on the quantified microseismic monitoring data: Since the microseismic signals are not spatially continuous, the quantified microseismic dynamic indicators need to be interpolated to form a continuous indicator coverage area. The kriging algorithm is based on the theory of variograms and structural analysis, and uses the original data of regionalized variables and the structural characteristics of variograms to perform linear unbiased optimal estimation of the values of regionalized variables at unsampled points. Interpolation method, the final result of the kriging algorithm is the kriging estimate. That is, the estimated value of any point to be estimated is obtained by the linear combination of n valid sample values within the influence range of the point to be estimated. This invention uses the built-in geostatistical module of ArcGIS to perform kriging interpolation on the quantified microseismic monitoring data to obtain the kriging estimated value of any point to be estimated/> The specific calculation formula is as follows:

In _{the formula ,} contribution.

As a specific embodiment, the weight coefficient λ _i of equation (4) can be calculated by different forms of theoretical variograms. Since the distance between the point to be estimated and the known point within the influence range is different, the corresponding theoretical variogram models are also different. . Commonly used variogram theoretical models include spherical models, Gaussian function models, exponential models, and power function models.

As a specific embodiment, the weight coefficient λ _i in this method is calculated through a Gaussian function model, and the Gaussian function model is as follows:

Substituting the calculated r value into the following formula to obtain the weight coefficient λ _i ,

In the formula, h is the interpolation distance, a is the variable range, c ₀ and c are the initial nugget value and nugget value respectively.

As a specific embodiment, before performing kriging interpolation in step S3, it also includes performing Log conversion on the quantified microseismic monitoring data to determine whether the quantified data obeys the normal distribution, that is, using kriging The premise of interpolation is that the data serves a normal distribution. Log transformation is performed on the quantitative data, and the transformed normal distribution QQ normal chart as shown in Figure 3 and the trend analysis chart as shown in Figure 4 can be obtained. After transformation The fitting parameter R ² between the data and the standard normal deviation is 0.998, so the transformed data can be considered to obey the normal distribution.

S4. Division of outburst danger areas: Input the spatial distribution characteristics of microseismic monitoring data obtained after Kriging interpolation into ArcGIS, a powerful geographical data processing system, to form a visualized coal and gas outburst on the mining face as shown in Figure 5. The hazard area classification level cloud chart realizes the visualization of the hazard level of the outburst hazard area on the mining working surface, and sets the hazards of coal and gas outbursts into four levels: safe, mild hazard, moderate hazard, and severe hazard.

Compared with the existing technology, the method for quickly dividing the mining face outburst danger zone based on Kriging interpolation provided by the present invention has the following beneficial effects: 1). The coal and gas outburst microseismic monitoring system included in the present invention has the ability to automatically eliminate noise function, the Holt exponential smoothing method is used to complete the microseismic monitoring data, and the interference information generated by the mining construction in the mining working face is accurately eliminated, thereby greatly improving the monitoring accuracy of coal and rock fracture information; 2), This invention establishes dynamic coal and gas outburst continuous monitoring technology early warning indicators based on microseismic monitoring data, and realizes the function of non-contact real-time monitoring of the danger of coal and gas outbursts in mining working faces. The normal distribution multi-level early warning outburst model is adopted. The energy and time series change characteristics of microseismic events are mainly used as judgment indicators to ensure the accuracy and efficiency of early warning; 3). The method for quickly dividing the mining working face outburst dangerous area based on Kriging interpolation provided by the present invention solves the existing problem. The amount of technical drilling projects is large, the interference from human factors is large, and it affects production to a certain extent. It is possible to make a dynamic analysis of the stress field distribution characteristics near the geological structures before and during the mining process of the coal mining face, and highlight dangerous locations and The moment of occurrence facilitates risk analysis; 4), this method applies microseismic monitoring technology to coal and gas outburst mines, based on the theoretical basis of coal and gas outburst mechanism and geostatistics, and cooperates with GIS technology and Kriging algorithm to establish a The dynamic evaluation of microseismic data can realize the visualization of the danger level of the prominent danger zone in the mining working face.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified. Modifications or equivalent substitutions without departing from the spirit and scope of the technical solution of the present invention shall be included in the scope of the claims of the present invention.

Claims (7)

1. A quick division method for outburst dangerous areas in mining working faces based on Kriging interpolation, which is characterized by including the following steps: S1. Microseismic data monitoring and preprocessing: Microseismic monitoring data from the mining working face are collected through the data monitoring system, and the Holt index smoothing method is used to complete the data processing of the microseismic monitoring data to accurately eliminate the data generated by the mining construction in the mining working face. The interference anomaly data is used to obtain mine safety monitoring data with time series attributes. The microseismic data monitoring includes the spatial distribution characteristics of microseismic events and the time series change characteristics of microseismic indicators; S2. Quantification of microseismic monitoring data: Based on the probability characteristics of the normal distribution function, establish a normal distribution multi-level early warning outburst model, perform quantitative processing on the microseismic monitoring data, and calculate the outburst risk grade division value f of all microseismic monitoring data. And use the scores to quantitatively divide each protrusion risk grade division value f. The discrimination formula of the protrusion risk grade division value f is as follows: in, In the formula, x represents the random error risk discrimination coefficient, m _i represents the microseismic parameter, represents the overall sample mean of microseismic parameters, and S _m is the standard deviation of microseismic parameters; S3. Perform kriging interpolation on the quantified microseismic monitoring data: Use the built-in geostatistics module of ArcGIS to perform kriging interpolation on the quantified microseismic monitoring data to obtain the kriging estimate of any point to be estimated. The specific calculation formula is as follows: In _{the formula ,} contribution; S4. Division of outburst dangerous areas: The spatial distribution characteristics of microseismic monitoring data obtained after Kriging interpolation are input into the ArcGIS geographical data processing system to form a visual cloud map of the classification of coal and gas outburst dangerous areas in the mining working face. 2. The method for quickly dividing the mining face outburst dangerous area based on Kriging interpolation according to claim 1, characterized in that the data monitoring system in step S1 includes a microseismic probe, a monitoring substation and a ground center station, so The microseismic probe is installed in the tunnel to transmit monitoring data to a monitoring substation, and the monitoring substation transmits the monitoring data to the ground center station through a network switch. 3. The method for quickly dividing the mining face protruding dangerous area based on Kriging interpolation according to claim 1, characterized in that in step S1, the Holt exponential smoothing method is used to perform data completion processing on the microseismic monitoring data. The specific principle formula is as follows: In the formula, α and γ are smoothing parameters, X _t is the monitoring data at time t, S _t is the smooth value at time t after correction using the smooth value S _t-1 of the previous period and the trend value b _t-1 , b _t is the modified trend value at time t obtained by combining the two adjacent smoothing values with the previous period's trend value b _t-1 , and T is the number of prediction periods. 4. The method for quickly dividing the outburst danger zone of the mining working face based on Kriging interpolation according to claim 1, characterized in that in the step S2, the danger of coal and gas outburst is divided into Safety is set into four levels: safety, mild risk, moderate risk and severe risk. The scores F of each level of safety, mild risk, moderate risk and severe risk are 0≤F respectively. <35, 35≤F<70, 70≤F<85, F≥85. 5. The method for quickly dividing the outburst dangerous area of the mining working face based on kriging interpolation according to claim 1, characterized in that, before performing the kriging interpolation in step S3, it also includes quantifying the microseismic The monitoring data is Log transformed to determine whether the quantified data obeys the normal distribution. 6. The method for quickly dividing the mining working face outburst dangerous area based on Kriging interpolation according to claim 1, characterized in that the weight coefficient λ _i in step S3 includes a spherical model, a Gaussian function model, an exponential model and Calculated using theoretical variograms including power function models. 7. The method for quickly dividing the mining working face outburst dangerous area based on Kriging interpolation according to claim 6, characterized in that the weight coefficient λ _i is calculated by a Gaussian function model, and the Gaussian function model is as follows: Substituting the calculated r value into the following formula to obtain the weight coefficient λ _i , In the formula, h is the interpolation distance, a is the variable range, c ₀ and c are the initial nugget value and nugget value respectively.