Mine mining area micro-seismic monitoring method and system based on multi-parameter fusion

Abstract

The invention relates to the technical field of earthquake monitoring, and discloses a mining area micro-earthquake monitoring method and system based on multi-parameter fusion, and the method comprises the steps: collecting time-space synchronization data of a mining area through employing a preset sensor array, and the time-space synchronization data comprise gravity data and seismic wave data; performing mixed noise reduction on the gravity data to obtain noise reduction data of the gravity data; extracting a gravity anomaly index of the mining area based on the noise reduction data; extracting a seismic wave energy ratio of the seismic wave data; based on the gravity anomaly index and the seismic wave energy ratio, establishing a joint criterion model of the mining area; and performing micro-seismic early warning on the mining area based on the combined criterion value of the combined criterion model and a preset early warning threshold value. The micro-seismic monitoring efficiency of the mining area can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of earthquake monitoring, and in particular to a method and system for microseismic monitoring in a mining area based on multi-parameter fusion. Background Art

[0002] In mining activities, microseismic monitoring is a key link to ensure safe production in mines. Mining operations will cause changes in the stress state of underground rock masses, triggering microseismic phenomena. If microseismic activities cannot be monitored in a timely and accurate manner, it may lead to serious accidents such as mine collapse and rock bursts, posing a huge threat to the safety of personnel and mining facilities.

[0003] At present, microseismic monitoring in mines faces many challenges. On the one hand, most of the existing monitoring methods rely on only a single type of data, such as monitoring seismic wave data only. However, microseismic activity is a complex geological process, and a single data cannot fully reflect its characteristics and laws. On the other hand, the mining environment is extremely complex and there are many interference factors. The continuous operation of mechanical equipment will generate mechanical vibration noise, and the geological tectonic activity itself will also bring interference signals. These noises and interferences will seriously pollute the monitoring data. Traditional noise reduction methods are difficult to effectively remove these complex interference components, which greatly reduces the accuracy of monitoring data. In addition, with the continuous expansion of mining scale and the gradual increase in mining depth, geological conditions have become more complex, and the laws of microseismic activity have become more elusive. In the face of such complex and changeable situations, the existing monitoring methods cannot meet the growing demand for microseismic monitoring in mine safety production. Summary of the invention

[0004] The present invention provides a method and system for microseismic monitoring in a mining area based on multi-parameter fusion, the main purpose of which is to solve the problem of low efficiency in microseismic monitoring in a mining area.

[0005] To achieve the above object, the present invention provides a method for microseismic monitoring of mining areas based on multi-parameter fusion, comprising: Using a preset sensor array to collect time-space synchronous data of a mining area, wherein the time-space synchronous data includes: gravity data and seismic wave data; Performing mixed denoising on the gravity data to obtain denoised data of the gravity data; Extracting a gravity anomaly index of the mining area based on the noise reduction data; extracting a seismic wave energy ratio of the seismic wave data; Based on the gravity anomaly index and the seismic wave energy ratio, a joint criterion model of the mining area is established; Based on the joint criterion value of the joint criterion model and a preset warning threshold, microseismic early warning is performed on the mining area.

[0006] Optionally, the hybrid noise reduction of the gravity data to obtain the noise-reduced data of the gravity data includes: Dynamically adjusting the wavelet basis function and threshold parameter according to the noise characteristics of the gravity data, and performing primary noise reduction processing on the signal corresponding to the gravity data at different scales based on the wavelet basis function and the threshold parameter to obtain the primary noise-reduced signal of the gravity data; Decomposing the primary noise-reduced signal into a plurality of intrinsic mode functions in descending order of frequency, and retaining the intrinsic mode function components related to microseismicity; Separating and removing the interference components unrelated to microseismicity in the intrinsic mode function components based on the independent component analysis algorithm.

[0007] Optionally, the separating and removing the interference components unrelated to microseismicity in the intrinsic mode function components based on the independent component analysis algorithm includes: Inputting the intrinsic mode function components into a pre-constructed independent component analysis model to construct a multi-dimensional signal space composed of the intrinsic mode function components; Performing whitening processing on the multi-dimensional signal space to remove the second-order correlation between the intrinsic mode function components in the multi-dimensional signal space; Based on the whitened multi-dimensional signal space and a pre-constructed objective function, separating independent source signal components, where the objective function is:

[0008] where, is the objective function for optimizing signal separation, is the hyperbolic tangent function, is the whitened signal determined based on the whitened multi-dimensional signal space, is the weight vector to be optimized, is the transpose of the weight vector to be optimized , is the standard Gaussian variable, is the sparse constraint coefficient, is the L1 norm, represents the expectation; Performing a correlation comparison according to the time-frequency domain characteristics of the source signal components and the prior knowledge of microseismic signals, and identifying the independent components representing the interference components in the source signal components.

[0009] Optionally, the calculation formula of the gravity anomaly index is as follows:

[0010] where, is the gravity anomaly index of the th sensor within a time window centered at is the th sensor's noise-reduced data within a time window centered at time; is the th sensor's mean value of the noise-reduced data within a time window centered at time; is the th sensor's standard deviation of the noise-reduced data within a time window centered at time; is the sensor identifier in the preset sensor array, is the time identifier.

[0011] Optionally, the calculation formula of the seismic wave energy ratio is as follows:

[0012] where is the seismic wave energy of the th sensor within a time window centered at time in a certain specific frequency band (denoted as frequency band 1), is the seismic wave energy of the th sensor within a time window centered at time in another specific frequency band (denoted as frequency band 2), is the frequency band identifier, is the length of the time window for calculating energy, is the integral variable representing time, is the th sensor's seismic wave energy ratio within a time window centered at time, is the sensor identifier in the preset sensor array, is the time identifier, is the component of the frequency band.

[0013] Optionally, the combined criterion value generated by the combined criterion model has the following calculation formula:

[0014] where is the combined criterion value, is the total number of sensors in the preset sensor array, is the sensor identifier in the preset sensor array, is the weight coefficient of the gravity anomaly index of the th sensor in the preset sensor array, is the th sensor's gravity anomaly index within a time window centered at time, is the weight coefficient of the seismic wave energy ratio of the th sensor in the preset sensor array, is the th sensor's seismic wave energy ratio within a time window centered at time, is the error term, is the time identifier.

[0015] Optionally, the weight coefficients in the joint criterion model are determined by the following method: Collect the characteristic data sets of the gravity anomaly index and the seismic wave energy ratio in the historical microseismic events in the mine mining area; Determine the contribution degree coefficients of the parameters of each sensor in the preset sensor array through principal component analysis; Determine the optimal weight combination of the joint criterion model based on the contribution degree coefficients; Perform spatial differential configuration on the weight coefficients in the joint criterion model based on the rock stratum stress distribution characteristics in the mine mining area and the optimal weight combination.

[0016] Optionally, the microseismic early warning of the mine mining area based on the joint criterion value of the joint criterion model and a preset early warning threshold includes: Real-time monitor the joint criterion value of the joint criterion model; When the joint criterion value exceeds the preset first early warning threshold, trigger a primary early warning signal; When the joint criterion value continuously exceeds the preset second early warning threshold for a set duration, trigger a high-level early warning signal; Send early warning information and location distribution maps to the monitoring terminals in the mine mining area through a visualization interface and a communication module.

[0017] Optionally, the method for setting the early warning threshold includes: Statistically analyze the distribution characteristics of the joint criterion benchmark values during the historical safe period in the mine mining area; Determine the joint criterion critical values corresponding to different danger levels in the mine mining area according to the distribution characteristics; Adaptive threshold adjustment is performed on the combined criterion threshold value according to the mining evolution stage of the mine mining area to obtain the early warning threshold of the mine mining area.

[0018] To solve the above problems, the present invention also provides a microseismic monitoring system for a mine mining area based on multi-parameter fusion, and the system includes: A data synchronous acquisition module, configured to acquire spatio-temporal synchronous data of a mine mining area by using a preset sensor array, wherein the spatio-temporal synchronous data includes: gravity data and seismic wave data; A hybrid noise reduction module, configured to perform hybrid noise reduction on the gravity data to obtain the noise-reduced data of the gravity data; An abnormal index extraction module, configured to extract the gravity abnormal index of the mine mining area based on the noise-reduced data; A seismic wave energy ratio generation module, configured to extract the seismic wave energy ratio of the seismic wave data; A combined criterion model establishment module, configured to establish a combined criterion model of the mine mining area based on the gravity abnormal index and the seismic wave energy ratio; A microseismic early warning module, configured to perform microseismic early warning on the mine mining area based on the combined criterion value of the combined criterion model and a preset early warning threshold.

[0019] The present invention constructs a monitoring framework for a mine mining area, fuses gravity data and seismic wave data, realizes multi-parameter collaborative monitoring, and lays a foundation for accurately evaluating microseismic risks. Among them, for the noise reduction processing of gravity data, adaptive wavelet threshold noise reduction, empirical mode decomposition, and independent component analysis algorithms are adopted to effectively remove complex noise interference, extract weak microseismic signals, improve data quality and feature extraction accuracy. The gravity abnormal index and the seismic wave energy ratio are quantified through specific calculation formulas, enhancing the characterization ability of microseismic activity characteristics. The weight coefficients of the combined criterion model are determined based on historical data and principal component analysis to achieve spatial differential configuration, making the model more suitable for the actual geology and mining conditions of the mine, improving the accuracy and scientificity of microseismic risk assessment. At the same time, grading early warning thresholds and duration conditions are set according to the combined criterion value, and combined with adaptive threshold adjustment, microseismic early warning can be issued in a timely and accurate manner, providing sufficient response time for mine safety production. Therefore, the present invention proposes a microseismic monitoring method and system for a mine mining area based on multi-parameter fusion, which can solve the problem of low microseismic monitoring efficiency in the mine mining area. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic flowchart of a microseismic monitoring method for a mine mining area based on multi-parameter fusion provided by an embodiment of the present invention; Figure 2Functional module diagram of the microseismic monitoring system for mine mining areas based on multi-parameter fusion provided by an embodiment of the present invention; The realization of the purpose of the present invention, functional features and advantages will be further described in conjunction with embodiments with reference to the accompanying drawings. Specific embodiments

[0021] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0022] The embodiments of the present application provide a microseismic monitoring method for mine mining areas based on multi-parameter fusion. The execution subject of the microseismic monitoring method for mine mining areas based on multi-parameter fusion includes but is not limited to at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiments of the present application. In other words, the microseismic monitoring method for mine mining areas based on multi-parameter fusion can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, Content Delivery Network (CDN), and big data and artificial intelligence platforms.

[0023] Refer to Figure 1 As shown, it is a schematic flowchart of the microseismic monitoring method for mine mining areas based on multi-parameter fusion provided by an embodiment of the present invention. In this embodiment, the microseismic monitoring method for mine mining areas based on multi-parameter fusion includes: S1. Use a preset sensor array to collect spatio-temporal synchronization data of the mine mining area, where the spatio-temporal synchronization data includes: gravity data and seismic wave data.

[0024] In the embodiment of the present invention, the use of a preset sensor array to collect spatio-temporal synchronization data of the mine mining area includes: Arrange the sensor array in a three-dimensional grid distribution on the surface and underground roadways of the mine mining area; Dynamically adjust the sensor spacing according to the mining intensity, rock formation depth and lithological structure; Use a GPS synchronization module to ensure that the time synchronization error of all sensors is less than 1 microsecond; Transmit the spatio-temporal synchronization data to the central processing unit in real time, where the spatio-temporal synchronization data includes: gravity data and seismic wave data.

[0025] Specifically, the sensor array is arranged on the surface of the mine exploitation area and the underground roadway in a three-dimensional grid distribution. The sensor array is a collection composed of multiple sensors, and these sensors work together to collect physical data at different positions in the mine exploitation area, which is the basic equipment for obtaining monitoring data. The three-dimensional grid distribution can comprehensively cover the spatial range of the mine exploitation area, enabling data collection at every position from the surface to the underground roadway, ensuring the comprehensiveness of data collection, so as to more accurately monitor the distribution of microseismic activities in the entire mine area.

[0026] Specifically, the sensor spacing is dynamically adjusted according to the mining intensity, rock stratum depth, and lithological structure. The mining intensity reflects the frequency and scale of mine exploitation activities. In areas with high mining intensity, microseismic activities may be more frequent, and denser sensor arrangements are required. Different rock stratum depths result in different microseismic propagation characteristics. Microseismic signals may attenuate more in deeper rock strata, and the sensor spacing needs to be reasonably adjusted to ensure effective signal reception. Different lithological structures, such as different hardness and elasticity of rocks, also lead to different propagation speeds and attenuation degrees of microseismic signals in them. Dynamically adjusting the sensor spacing according to these factors can make the sensor layout more adaptable to the actual geology and mining conditions of the mine, ensuring that the collected data can accurately reflect the characteristics of microseismic activities.

[0027] Specifically, a GPS synchronization module is used to ensure that the time synchronization error of all sensors is less than 1 microsecond. The GPS synchronization module utilizes the precise timekeeping function of the Global Positioning System to provide a unified time reference for each sensor. In microseismic monitoring, time synchronization is crucial because the occurrence time and propagation time of microseismic events are key information for analyzing the laws of microseismic activities and locating microseismic sources. A time synchronization error less than 1 microsecond can ensure high-precision consistency in time for the data collected by different sensors.

[0028] Specifically, spatio-temporal synchronized data refers to data that contains both spatial position information (reflected by the three-dimensional grid distribution of sensors) and precise time information (achieved through the GPS synchronization module). Here, it mainly refers to gravity data and seismic wave data. Among them, the gravity data reflects the changes in the gravity field in the mine exploitation area. Microseismic activities may cause slight changes in the local gravity field, and monitoring the changes in gravity data can provide important clues for microseismic monitoring. Seismic wave data is the seismic wave signal generated when a microseismic event occurs, directly carrying key characteristic information such as the energy and frequency of the microseismic event.

[0029] S2. Perform hybrid noise reduction on the gravity data to obtain the noise-reduced data of the gravity data.

[0030] In the embodiment of the present invention, the performing hybrid noise reduction on the gravity data to obtain the noise-reduced data of the gravity data includes: Dynamically adjust the wavelet basis function and threshold parameter according to the noise characteristics of the gravity data, and perform primary noise reduction processing on the signal corresponding to the gravity data at different scales based on the wavelet basis function and the threshold parameter to obtain the primary noise-reduced signal of the gravity data; Decompose the primary noise-reduced signal into multiple intrinsic mode functions in order of decreasing frequency, and retain the intrinsic mode function components related to microseismicity; Based on the independent component analysis algorithm, separate and remove the interference components unrelated to microseismicity in the intrinsic mode function components.

[0031] Specifically, the wavelet basis function is the core of wavelet transform. Different wavelet basis functions have different abilities to capture signal characteristics. By dynamically adjusting it, it can better adapt to the characteristics of complex noise in gravity data. For example, for noise in different frequency ranges, selecting a suitable wavelet basis function can more effectively match and process. The threshold parameter is used to control the degree of retention of signal details in the wavelet transform process. Dynamically adjusting the threshold according to the noise intensity can retain the useful microseismic signal to the greatest extent while removing noise. Processing the signal at different scales is like screening data with filters of different precisions, initially filtering strong noise at a large scale and finely processing weak noise at a small scale to achieve comprehensive noise reduction.

[0032] Furthermore, use quantum computing technology to accelerate the calculation process of the ICA algorithm. Among them, using quantum computing technology to accelerate the calculation process of the ICA algorithm includes: using quantum parallel computing to optimize the decomposition step of the covariance matrix in ICA and reducing the iteration time of the ICA algorithm.

[0033] Specifically, the intrinsic mode function (IMF) is a concept in empirical mode decomposition (EMD). Each IMF represents the characteristic fluctuation component of the signal at different time scales. After the primary noise reduction of gravity data, it contains multiple frequency components, some of which are related to microseismicity and some are residual interference. After decomposition by frequency, the IMF components related to microseismicity can be selected according to the frequency characteristics of microseismic signals. For example, mine microseismic signals are mostly concentrated in a specific frequency range. By comparing the frequencies of each IMF component, the key parts can be retained, and irrelevant high-frequency or low-frequency interference can be removed to further purify the information related to microseismicity.

[0034] Specifically, the independent component analysis (ICA) algorithm is a technology that can separate independent source signals from mixed signals. In a mine environment, even after the first two steps of processing, there may still be various interferences in gravity data. For example, interference signals generated by mechanical vibrations and geological structure activities are mixed with microseismic signals in the IMF components. The ICA algorithm constructs a multi-dimensional signal space, performs whitening processing to remove the second-order correlation between components, and then separates the independent source signal components according to a specific objective function.

[0035] Specifically, collaborative processing is performed on adaptive wavelet threshold denoising, EMD decomposition, and the ICA algorithm. Among them, adaptive wavelet threshold denoising is used to remove high-frequency noise, EMD decomposition is used to extract local features of the signal, and the ICA algorithm is used to separate spatially independent components.

[0036] Furthermore, for mechanical vibration noise and geological structure noise, the number and weights of independent components in the ICA algorithm are adjusted.

[0037] Specifically, the denoising effect is verified in real time during the hybrid denoising process. When the improvement amplitude of the signal-to-noise ratio is lower than the set threshold, the acquisition parameter adjustment instruction of the sensor array is triggered.

[0038] Specifically, separating and removing the interference components unrelated to microseismic events in the intrinsic mode function components based on the independent component analysis algorithm includes: Inputting the intrinsic mode function components into a pre-constructed independent component analysis model to construct a multi-dimensional signal space composed of the intrinsic mode function components; Performing whitening processing on the multi-dimensional signal space to remove the second-order correlation between the intrinsic mode function components in the multi-dimensional signal space; Based on the whitened multi-dimensional signal space and a pre-constructed objective function, independent source signal components are separated, where the objective function is:

[0039] where, is the objective function for optimizing signal separation, is the hyperbolic tangent function, is the whitened signal determined based on the whitened multi-dimensional signal space, is the weight vector to be optimized, is the weight vector to be optimized transpose, is a standard Gaussian variable, is the sparse constraint coefficient, is the L1 norm, represents the expectation; Based on the time-frequency domain characteristics of the source signal components and the prior knowledge of microseismic signals, a correlation comparison is performed to identify the independent components representing interference components in the source signal components.

[0040] Specifically, the multi-dimensional signal space is centered to eliminate the mean shift, and the second-order correlation between the components in the signal space is removed through whitening processing.

[0041] Specifically, the ICA processing steps include: input signal, whitening processing, construction of the objective function, iterative solution, and component identification.

[0042] Specifically, the objective function adopts an adaptive non - linear negentropy criterion, enhances the capture of the non - Gaussian characteristics of microseismic signals through the tanh function, and introduces L1 regularization to suppress noise redundant components, improving the capture ability of weak microseismic signals and anti - noise robustness.

[0043] Furthermore, the interference components are filtered from the multi - dimensional signal space through inverse projection reconstruction to obtain the denoised gravity data.

[0044] Specifically, the objective function, used to measure the optimization objective of signal separation, contains a negentropy maximization term and a sparse constraint term. By minimizing , high non - Gaussian (microseismic signals) and sparse effective components are extracted from the noisy gravity data, suppressing complex interferences such as mechanical vibrations and geological activities in the mine environment, achieving the maximization of the non - Gaussianity (negentropy criterion) and enhancement of sparsity (L1 regularization) of the separated signals, thereby accurately extracting microseismic effective components.

[0045] Specifically, the hyperbolic tangent function, as a non - linear activation function, is used to approximate the negentropy of the signal. Its characteristics are: microseismic signals usually appear as short - time pulses or low - frequency vibrations, and their non - Gaussian characteristics are amplified by , being more sensitive to non - Gaussian signals (such as pulsed microseismic signals), enhancing the representation ability of weak signals, and facilitating the distinction from Gaussian noise (such as instrument thermal noise); the function output range is limited to , avoiding numerical overflow caused by strong interference signals in the mine (such as blasting vibrations), suppressing the interference of Gaussian noise, and improving the signal - to - noise ratio.

[0046] Specifically, the separation vector to be optimized , is used to extract independent source signals from the whitened signal ; by iteratively updating , it dynamically matches the main noise types in the current mining stage, and at the same time, different positions of in the sensor array can be differentially configured to adapt to the inhomogeneity of the rock formation stress distribution.

[0047] Specifically, refers to the linear projection of the separation vector and the whitened signal . By adjusting , independent components are separated from the whitened multi - dimensional signal space (including IMF components of EMD decomposition). Mine microseismic signals are mostly concentrated in frequency band, and the optimization direction of will strengthen the energy proportion of this frequency band.

[0048] Specifically, the whitening signal is generated by transformation with a whitening matrix. Since the sensors in the mine are limited by space and are vulnerable to electromagnetic coupling interference, the whitening process can eliminate such second-order correlations. Among them, the whitening signal , is the original signal in the multi-dimensional signal space after the whitening process, is the whitening matrix.

[0049] Specifically, based on the multi-dimensional signal space after the whitening process, a target function is constructed using the negative entropy maximization criterion, and the target function is solved by a fixed-point iteration algorithm to separate independent source signal components; by iteratively updating , maximizing the non-Gaussianity, thereby approximating the true independent components of the microseismic signal; the standard Gaussian variable is used as the benchmark for negative entropy calculation. By comparing the with the expected difference, the non-Gaussianity of the signal is quantified. The greater the difference, the stronger the non-Gaussianity of the separated signal (closer to the true microseismic signal).

[0050] Specifically, the non-Gaussian degree of the microseismic signal is measured by the difference from the to distinguish effective signals from background noise; the statistical characteristics of the are updated according to historical data to adapt to the microseismic feature differences of different lithological structures (such as hard rock and soft rock).

[0051] Specifically, the sparse constraint coefficient is used to control the weight of the L1 regularization term and adjust the sparsity of the separated signal; when , forcing some components of the to approach zero and suppressing redundant noise components; in the mine scenario, hierarchical regulation can be carried out using the . Among them, a higher is used in the deep mining area (high hydrostatic pressure environment) to suppress the high-frequency noise of rock bursts; while in the shallow area, the is reduced to retain low-frequency micro-fracture signals.

[0052] Specifically, the L1 norm refers to the sum of the absolute values of the elements of a vector. Introducing the L1 norm into the sparsity constraint forces the separated vector to concentrate on a few key sensor nodes, matching the local propagation characteristics of the mine microseismic signal; at the same time, in the complex mine noise environment, it enhances the robustness to occasional interference (such as mechanical vibration). For example, when some sensors are damaged by the underground environment, the sparse constraint can reduce the impact of faulty nodes on the noise reduction result.

[0053] Specifically, a sliding time window (such as 10 s) can be adopted to calculate the expected value to adapt to the time-varying characteristics of microseismic signals; when the mining activities change periodically (such as day and night alternation), the robustness of the algorithm can be enhanced through local expected value estimation.

[0054] Specifically, the dimension of the whitened signal matches the three-dimensional grid of sensor layout to ensure the integrity of spatial coverage; meanwhile, the separated microseismic signals are used to calculate the gravity anomaly index, and its signal-to-noise ratio directly affects the accuracy of the early warning threshold.

[0055] Specifically, on the basis of the combination of adaptive wavelet threshold denoising and empirical mode decomposition, the independent component analysis (ICA) algorithm is introduced. ICA can separate different independent components in the gravity data, further remove the interference components unrelated to microseismic events, and improve the denoising effect. The three algorithms complement each other's advantages and process the gravity data from different perspectives, significantly improving the denoising effect.

[0056] Specifically, the hybrid denoising algorithm is designed specifically for the processing of gravity data in mine microseismic monitoring. The mine environment is complex, and gravity data is easily interfered by various factors, such as the operation of mechanical equipment and geological structure changes. Traditional denoising methods may not be able to effectively remove these complex interference components, while the hybrid denoising algorithm can better adapt to the characteristics of the mine environment through the synergistic effect of multiple algorithms, accurately extract the information related to microseismic events, and provide a more reliable data basis for mine microseismic monitoring.

[0057] Specifically, in mine microseismic monitoring, accurate gravity data is crucial for judging microseismic activities. The existence of noise will interfere with the identification and analysis of microseismic signals, resulting in misjudgment or missed judgment. The hybrid denoising algorithm improves the quality of gravity data by effectively removing noise, making the microseismic signals clearer, so as to be able to monitor microseismic activities more accurately and provide more reliable early warning information for mine safety production.

[0058] S3. Extract the gravity anomaly index of the mine mining area based on the denoised data.

[0059] In the embodiment of the present invention, the calculation formula of the gravity anomaly index is as follows:

[0060] where is the gravity anomaly index of the th sensor within a time window centered at time, is the denoised data of the th sensor within a time window centered at time, is the The mean value of the noise-reduced data of a sensor within a time window centered at is the mean value of the noise-reduced data of the sensor within a time window centered at is the th sensor, and the standard deviation of the noise-reduced data of the th sensor within a time window centered at is the sensor identification in the preset sensor array, and is the time identification.

[0061] Specifically, calculating the gravity anomaly index through this formula can highlight the deviation degree of the gravity data from the average level. During the mining process, microseismic activities will cause slight changes in the local gravity field. When the value of [[VALUE]] is large, it indicates that the gravity data at this position is significantly different from the average level, and there may be geological activities related to microseismicity. For example, in a mining area, when a microseismic event occurs in a certain area, the gravity data in this area and its vicinity may show abnormal fluctuations, and the calculated gravity anomaly index will increase accordingly.

[0062] S4. Extract the seismic wave energy ratio of the seismic wave data.

[0063] In the embodiment of the present invention, the extraction of the seismic wave energy ratio of the seismic wave data includes: Performing wavelet decomposition on the seismic wave data to obtain components in different frequency bands; Generating the seismic wave energy ratio of the seismic wave data according to the components.

[0064] Specifically, wavelet decomposition is a signal processing technique that can decompose complex seismic wave signals according to frequency components. Just like decomposing a beam of mixed light into different colors of light through a prism, wavelet decomposition can decompose seismic waves into multiple signal components with different frequency ranges. This is because the seismic waves generated by microseismicity contain multiple frequency components, and seismic waves with different frequencies carry different information during propagation, and their relationships with the characteristics of microseismicity and geological structures are also different. Through wavelet decomposition, these components in different frequency bands can be separated.

[0065] Specifically, the calculation formula of the seismic wave energy ratio is as follows:

[0066] Wherein, is the seismic wave energy of the th sensor within a time window centered at in a certain specific frequency band (denoted as frequency band 1), and is the seismic wave energy of the th sensor within a time window centered at in another specific frequency band (denoted as frequency band 2). is the frequency band identifier, is the time window length for calculating energy, is the integral variable representing time, is the -th sensor's ratio of the seismic wave energy within a time window centered at time, is the sensor identifier in the preset sensor array, is the time identifier, is the frequency band component.

[0067] Specifically, the ratio of seismic wave energy reflects the relative magnitude relationship of seismic wave energy in different frequency bands. In the scenario of mine microseismic monitoring, the energy distribution of seismic waves in different frequency bands will be different under normal conditions and when microseisms occur. For example, when a microseism occurs, the energy in certain frequency bands may suddenly increase or decrease, resulting in a change in the ratio of seismic wave energy. By monitoring the change in this energy ratio, signs of microseismic activity can be captured more sensitively.

[0068] S5. Based on the gravity anomaly index and the ratio of seismic wave energy, establish a joint criterion model for the mine mining area.

[0069] In the embodiment of the present invention, the joint criterion value generated by the joint criterion model is calculated as follows:

[0070] where, is the joint criterion value, is the total number of sensors in the preset sensor array, is the sensor identifier in the preset sensor array, is the weight coefficient of the gravity anomaly index of the -th sensor in the preset sensor array, is the -th sensor's gravity anomaly index within a time window centered at time, is the weight coefficient of the ratio of seismic wave energy of the -th sensor in the preset sensor array, is the -th sensor's ratio of seismic wave energy within a time window centered at time, is the error term, is the time identifier.

[0071] Specifically, the weight coefficient is used to measure the relative importance of the gravity anomaly index and the seismic wave energy ratio of each sensor in the calculation of the combined criterion value. By assigning appropriate weights to the parameters of different sensors, the contribution degree of each parameter to the final criterion value can be adjusted according to the actual situation. The error term is used to cover uncontrollable factors or minor effects that may occur during the model calculation process, making the model more reasonable and perfect.

[0072] Specifically, the weight coefficient in the combined criterion model is determined by the following method: Collect the characteristic data sets of the gravity anomaly index and the seismic wave energy ratio in the historical microseismic events in the mine mining area; Determine the contribution degree coefficients of the parameters of each sensor in the preset sensor array through principal component analysis; Determine the optimal weight combination of the combined criterion model based on the contribution degree coefficients; Perform spatial differential configuration of the weight coefficient in the combined criterion model based on the rock stratum stress distribution characteristics in the mine mining area and the optimal weight combination.

[0073] Specifically, first collect the characteristic data sets of the gravity anomaly index and the seismic wave energy ratio in the historical microseismic events in the mine mining area. These historical data contain the actual performance of various parameters during past microseismic events. By accumulating a large amount of historical data, the relationship between microseismic activities and these two parameters under different circumstances can be understood more comprehensively.

[0074] Specifically, the principal component analysis method is used to determine the contribution degree coefficients of the parameters of each sensor in the preset sensor array. Principal component analysis is a data dimensionality reduction technique that can transform multiple related variables into a few uncorrelated comprehensive variables (principal components) while retaining most of the information of the original data.

[0075] In the embodiment of the present invention, through principal component analysis, the relative importance of the gravity anomaly index and the seismic wave energy ratio of each sensor in reflecting microseismic activities, that is, the contribution degree coefficient, can be clarified. For example, due to the special location of some sensors, the parameters collected by them are more critical for judging microseismic activities, and principal component analysis can reflect this importance in the form of coefficients.

[0076] Specifically, after obtaining the contribution degree coefficients of the parameters of each sensor, through a certain algorithm or optimization method, a set of weight values is found, so that the combined criterion model can achieve the best effect when evaluating microseismic risks, such as maximizing the distinction between the occurrence and non-occurrence of microseisms and improving the accuracy and reliability of the model.

[0077] Specifically, according to the rock stratum stress distribution characteristics and the optimal weight combination in the mine exploitation area, the weight coefficients in the combined criterion model are configured with spatial differentiation. The rock stratum stress distributions in different areas of the mine are different, and the rules and characteristics of microseismic activities will also vary. For example, in areas with greater rock stratum stress, the possibility and impact degree of microseismic occurrences may be greater. Correspondingly, the weights of the parameters of the sensors in this area in the combined criterion model should also be higher. By this way of configuring the weight coefficients with spatial differentiation, the combined criterion model can better adapt to the actual situations in different areas of the mine, further improving the accuracy and pertinence of microseismic monitoring.

[0078] S6. Conduct microseismic early warning for the mine exploitation area based on the combined criterion value of the combined criterion model and a preset early warning threshold.

[0079] In the embodiment of the present invention, the conducting of microseismic early warning for the mine exploitation area based on the combined criterion value of the combined criterion model and a preset early warning threshold includes: Real-time monitor the combined criterion value of the combined criterion model; When the combined criterion value exceeds the preset first early warning threshold, trigger a primary early warning signal; When the combined criterion value continuously exceeds the preset second early warning threshold for a set duration, trigger a high-level early warning signal; Send early warning information and a location distribution map to the monitoring terminals in the mine exploitation area through a visualization interface and a communication module.

[0080] Specifically, the combined criterion value comprehensively considers key parameters such as the gravity anomaly index and the seismic wave energy ratio, and can reflect the comprehensive situation of microseismic activities in the mine exploitation area. It is the core basis for microseismic early warning.

[0081] Furthermore, when the combined criterion value exceeds the preset first early warning threshold, the monitoring platform immediately triggers a primary early warning signal. The first early warning threshold is a boundary value set according to the historical data and experience of the mine exploitation area. Exceeding this value means that abnormal situations possibly related to microseisms have occurred in the mine exploitation area. However, at this time, the microseismic risk is relatively low. The primary early warning signal can remind relevant personnel to start paying attention to the situation in this area and make preparations for further monitoring and response.

[0082] Further, if the combined criterion value not only exceeds the preset second warning threshold, but also continuously exceeds this threshold for a set duration, the monitoring platform will trigger a high-level warning signal. Among them, the second warning threshold is usually more stringent than the first warning threshold. The continuous exceeding of the combined criterion value indicates a high probability of microseismic occurrence or that the microseismic activity has been relatively intense, posing a greater threat to the safety of the mine. The high-level warning signal can prompt the mine to take more urgent and effective measures, such as organizing personnel evacuation, stopping operations in dangerous areas, etc., to ensure the safety of personnel lives and mine facilities.

[0083] Specifically, after either the primary or high-level warning signal is triggered, warning information and a location distribution map will be sent to the monitoring terminals in the mine exploitation area through the visualization interface and the communication module. The visualization interface can intuitively display warning-related information, such as the warning level, the possible microseismic area, etc., facilitating the staff to quickly understand the situation; the communication module ensures that the warning information can be transmitted to the monitoring terminals in a timely and accurate manner, enabling relevant personnel to obtain information in a timely manner regardless of their location. The location distribution map can help the staff quickly locate the area where microseismic may occur, facilitating the adoption of targeted measures, such as arranging personnel to inspect and rescue in this area.

[0084] Specifically, the method for setting the warning threshold includes: Statistically analyze the distribution characteristics of the combined criterion baseline value in the historical safe period of the mine exploitation area; Determine the combined criterion critical value corresponding to different danger levels in the mine exploitation area according to the distribution characteristics; Perform an adaptive threshold adjustment on the combined criterion critical value according to the exploitation evolution stage of the mine exploitation area to obtain the warning threshold of the mine exploitation area.

[0085] Specifically, the data in the historical safe period reflects the change range and law of the combined criterion value under the normal exploitation state of the mine.

[0086] Specifically, based on the above distribution characteristics, determine the combined criterion critical value corresponding to different danger levels in the mine exploitation area. Different danger levels are divided according to the possible harm degree caused by microseisms, and each level corresponds to a different combined criterion critical value. The higher the danger level, the more stringent the critical value, that is, it is more sensitive to the abnormal change of the combined criterion value. This can make the warning system more targeted and issue corresponding warnings in a timely manner according to different danger degrees.

[0087] Specifically, considering that the mining evolution stages in the mining area are different, factors such as geological conditions and mining activities will change. Adaptive threshold adjustment is performed on the critical value of the combined criterion to obtain the final warning threshold. For example, in the initial stage of mining, the geological conditions of the mine are relatively stable, and the warning threshold can be appropriately relaxed. As the mining depth increases and the mining scope expands, the geological conditions become complex and the microseismic risk increases. At this time, the warning threshold should be correspondingly reduced to improve the sensitivity of the warning system to adapt to the actual changes during the mining process of the mine and ensure the accuracy and effectiveness of microseismic warning.

[0088] As Figure 2 shown, it is a functional module diagram of a microseismic monitoring system for a mining area based on multi-parameter fusion provided by an embodiment of the present invention.

[0089] The microseismic monitoring system 100 for a mining area based on multi-parameter fusion described in the present invention can be installed in an electronic device. According to the functions to be realized, the microseismic monitoring system 100 for a mining area based on multi-parameter fusion may include a data synchronous acquisition module 101, a hybrid noise reduction module 102, an anomaly index extraction module 103, a seismic wave energy ratio generation module 104, a combined criterion model establishment module 105, and a microseismic warning module 106. The modules described in the present invention may also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0090] In this embodiment, the functions of each module / unit are as follows: The data synchronous acquisition module 101 is used to collect spatio-temporal synchronous data of the mining area by using a preset sensor array, where the spatio-temporal synchronous data includes: gravity data and seismic wave data; The hybrid noise reduction module 102 is used to perform hybrid noise reduction on the gravity data to obtain the noise-reduced data of the gravity data; The anomaly index extraction module 103 is used to extract the gravity anomaly index of the mining area based on the noise-reduced data; The seismic wave energy ratio generation module 104 is used to extract the seismic wave energy ratio of the seismic wave data; The combined criterion model establishment module 105 is used to establish a combined criterion model for the mining area based on the gravity anomaly index and the seismic wave energy ratio; The microseismic warning module 106 is used to perform microseismic warning on the mining area based on the combined criterion value of the combined criterion model and a preset warning threshold.

[0091] In several embodiments provided by the present invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0092] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0093] In addition, in each embodiment of the present invention, the functional modules can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a combination of hardware and software functional modules.

[0094] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0095] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence is a theory, method, technology and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. 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 or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for microseismic monitoring in mining areas based on multi-parameter fusion, characterized in that: The method comprises: Using a preset sensor array to collect time-space synchronous data of a mining area, wherein the time-space synchronous data includes: gravity data and seismic wave data; Performing mixed denoising on the gravity data to obtain denoised data of the gravity data; Extracting a gravity anomaly index of the mining area based on the noise reduction data; extracting a seismic wave energy ratio of the seismic wave data; Based on the gravity anomaly index and the seismic wave energy ratio, a joint criterion model of the mining area is established; Based on the joint criterion value of the joint criterion model and a preset warning threshold, microseismic early warning is performed on the mining area.

2. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 1, characterized in that: The performing mixed denoising on the gravity data to obtain denoised data of the gravity data includes: Dynamically adjusting the wavelet basis function and the threshold parameter according to the noise characteristics of the gravity data, and performing primary noise reduction processing on the signal corresponding to the gravity data at different scales based on the wavelet basis function and the threshold parameter to obtain the primary noise reduction signal of the gravity data; Decomposing the primary noise reduction signal into a plurality of eigenmode functions according to the frequency from high to low, and retaining the eigenmode function components related to the microseism; Based on the independent component analysis algorithm, interference components irrelevant to microseismicity in the intrinsic mode function components are separated and removed.

3. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 2, characterized in that: The method of separating and removing interference components irrelevant to microseisms in the intrinsic mode function components based on an independent component analysis algorithm includes: Inputting the intrinsic mode function components into a pre-constructed independent component analysis model to construct a multi-dimensional signal space composed of the intrinsic mode function components; Performing whitening processing on the multidimensional signal space to remove the second-order correlation between the intrinsic mode function components in the multidimensional signal space; Based on the multidimensional signal space after the whitening process and the pre-constructed objective function, the independent source signal components are separated, wherein the objective function for: , in, is the objective function used to optimize signal separation, is the hyperbolic tangent function, is a whitened signal determined based on the multidimensional signal space after the whitening process, is the weight vector to be optimized, is the weight vector to be optimized The transpose of is a standard Gaussian variable, is the sparse constraint coefficient, is the L1 norm, express expectations; A correlation comparison is performed based on the time-frequency domain characteristics of the source signal components and the prior knowledge of the microseismic signal, and an independent component representing the interference component in the source signal components is identified.

4. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 1, characterized in that: The calculation formula of the gravity anomaly index is as follows: ， in, It is The sensors are The gravity anomaly index within a time window centered at time , It is The sensors are The noise reduction data within a time window centered at time , It is The sensors are The mean of the noise reduction data in a time window centered at time , It is The sensors are The standard deviation of the denoised data within a time window centered at time , is a sensor identifier in the preset sensor array, Is a time stamp.

5. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 1, characterized in that: The calculation formula of the seismic wave energy ratio is as follows: ， in, It is The sensors are The seismic wave energy of a specific frequency band (referred to as frequency band 1) in a time window centered at the moment It is The sensors are The seismic wave energy of another specific frequency band (referred to as frequency band 2) in a time window centered at the moment is the frequency band identifier, is the time window length for calculating energy, is the integral variable representing time, It is The sensors are The seismic wave energy ratio within a time window centered at time, is a sensor identifier in the preset sensor array, is a time stamp, It is the frequency band The amount.

6. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 1, characterized in that: The joint criterion value generated by the joint criterion model The calculation formula is as follows: ， in, is the joint criterion value, is the total number of sensors in the preset sensor array, is a sensor identifier in the preset sensor array, is the first sensor in the preset sensor array The weight coefficient of the gravity anomaly index of each sensor, It is The sensors are The gravity anomaly index within a time window centered at time , is the first sensor in the preset sensor array The weight coefficient of the seismic wave energy ratio of each sensor, It is The sensors are The seismic wave energy ratio within a time window centered at time, is the error term, Is a time stamp.

7. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 6, characterized in that: The weight coefficient in the joint criterion model is determined by the following method: Collect characteristic data sets of gravity anomaly index and seismic wave energy ratio in historical microseismic events in the mining area; Determining the contribution coefficient of each sensor parameter in the preset sensor array by principal component analysis; Determining the optimal weight combination of the joint criterion model based on the contribution coefficient; The weight coefficients in the joint criterion model are spatially differentiated based on the rock stress distribution characteristics of the mining area and the optimal weight combination.

8. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 1, characterized in that: The method of providing microseismic early warning for the mining area based on the combined criterion value of the combined criterion model and a preset early warning threshold comprises: monitoring the joint criterion value of the joint criterion model in real time; When the combined criterion value exceeds a preset first warning threshold, a primary warning signal is triggered; When the combined criterion value exceeds the preset second warning threshold value for a set time period, a high-level warning signal is triggered; Early warning information and location distribution maps are sent to the monitoring terminal of the mining area through the visualization interface and communication module.

9. The method for microseismic monitoring of mining areas based on multi-parameter fusion according to claim 8, characterized in that: The method for setting the warning threshold includes: Statistically analyzing the distribution characteristics of the joint criterion benchmark values ​​of the mining area in the historical safety period; Determine the joint criterion critical value corresponding to the mining area at different danger levels according to the distribution characteristics; The threshold value of the joint criterion is adaptively adjusted according to the mining evolution stage of the mining area to obtain the early warning threshold value of the mining area.

10. A microseismic monitoring system for mining areas based on multi-parameter fusion, characterized in that: The system comprises: A data synchronization acquisition module, used to collect time-space synchronization data of a mining area using a preset sensor array, wherein the time-space synchronization data includes: gravity data and seismic wave data; A hybrid noise reduction module, used for performing hybrid noise reduction on the gravity data to obtain noise reduction data of the gravity data; An anomaly index extraction module, used to extract the gravity anomaly index of the mining area based on the noise reduction data; A seismic wave energy ratio generating module, used for extracting the seismic wave energy ratio of the seismic wave data; A joint criterion model building module, used to build a joint criterion model of the mining area based on the gravity anomaly index and the seismic wave energy ratio; The microseismic early warning module is used to provide microseismic early warning for the mining area based on the joint criterion value of the joint criterion model and a preset early warning threshold.

Images