Geological disaster real-time monitoring and early warning method and system based on big data and environmental feature fusion

Abstract

The present application relates to mine safety hidden danger monitoring technical field, especially in kind based on big data and environmental characteristic fusion geological disaster real-time monitoring early warning method and system, include: obtain real-time monitoring time, then utilize the preset excavation collection time period and with the excavation collection unit receive excavation vibration signal, based on the excavation vibration signal obtains real-time mine image and mine monitoring signal set, otherwise, utilize the microseismic monitoring unit obtains real-time mine image and mine monitoring signal set, the real-time mine image is summarized, obtains mine image sequence, constructs the identification safety hidden danger point data set based on mine image sequence, analysis identification safety hidden danger point data set, completes a kind of based on big data and environmental characteristic fusion geological disaster real-time monitoring early warning.The present application utilizes with the excavation collection signal and microseismic monitoring technology real-time monitoring the environmental characteristic of excavation roadway, and utilizes big data to carry out hidden danger detection to environmental characteristic, to realize the real-time monitoring and early warning of the safety hidden danger under the mine.

Description

Technical Field

[0001] This invention relates to the field of mine safety hazard monitoring technology, and in particular to a method and system for real-time monitoring and early warning of geological disasters based on the fusion of big data and environmental characteristics. Background Technology

[0002] In coal mining, excavation is typically carried out by tunneling machines. Traditional geological hazard monitoring technologies mainly rely on manual inspections and static sensors, resulting in long data update cycles. Existing on-the-spot monitoring systems often employ single-signal acquisition modes, such as monitoring only the cutting vibration of the tunneling machine. Microseismic monitoring technology can also monitor coal mine roadways.

[0003] While the above methods can achieve real-time monitoring of geological hazards in coal mine roadways, microseismic monitoring technology is only used for stability assessment and is subject to interference from the tunneling machine during tunneling. Without combining it with the real-time operation data of the tunneling machine, it is also subject to noise interference from the tunneling machine. Therefore, a method is needed that can combine the real-time operation data of the tunneling machine with microseismic monitoring technology and use big data to detect hidden dangers in environmental characteristics, so as to achieve real-time monitoring and early warning of safety hazards in mines. Summary of the Invention

[0004] This invention provides a method for real-time monitoring and early warning of geological disasters based on the fusion of big data and environmental characteristics, and a computer-readable storage medium. Its main purpose is to use the signal acquisition during excavation and microseismic monitoring technology to monitor the environmental characteristics of the tunnel in real time, and to use big data to detect potential hazards in the environmental characteristics, so as to realize real-time monitoring and early warning of safety hazards in the mine.

[0005] To achieve the above objectives, this invention provides a real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental characteristics, comprising:

[0006] The system receives real-time environmental monitoring commands from the mine and confirms the coal mine monitoring environment based on these commands. The coal mine monitoring environment includes: a tunneling machine, tunneling roadways, and a data acquisition unit. The data acquisition unit includes: a tunneling acquisition unit and a shutdown acquisition unit. The tunneling acquisition unit includes multiple tunneling geophones, and the shutdown acquisition unit includes multiple static mine environment geophones and a microseismic monitoring unit.

[0007] The real-time monitoring time is obtained. If the real-time monitoring time is within a preset tunneling period, the tunneling vibration signal is received using the preset tunneling acquisition period and the tunneling acquisition unit. Based on the tunneling vibration signal, a real-time mine image and a mine monitoring signal set are obtained. The tunneling vibration signal comes from the vibration of the tunneling machine.

[0008] If the real-time monitoring time is not during the tunneling period, the real-time mine image and mine monitoring signal set are obtained by the microseismic monitoring unit, wherein the mine monitoring signal set is obtained by the microseismic monitoring unit.

[0009] The real-time mine images are summarized according to the preset monitoring time sequence to obtain a mine image sequence. All safety hazard point markers in the mine image sequence are identified to obtain a safety hazard point marker sequence. Based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence, a safety hazard point marker dataset is constructed. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence.

[0010] The system parses and identifies a dataset of potential safety hazards. If a safety warning instruction is generated based on the parsed dataset, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, the system returns to the step of obtaining the real-time monitoring time, thus completing a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental characteristics.

[0011] Optionally, the acquisition of real-time mine images and mine monitoring signal sets based on tunneling vibration signals includes:

[0012] The tunneling vibration signal is fitted using a pre-constructed seismic interferometry technique to obtain a set of virtual shot signals. A low-pass filtering operation is performed on the set of virtual shot signals to obtain a denoised signal. The denoised signal is analyzed using a pre-constructed scattering imaging method, and the analyzed denoised signal is visualized to obtain a mine hazard detection image corresponding to the real-time monitoring time.

[0013] The mine hazard detection images were analyzed to obtain real-time mine images containing safety hazard point markers, and the tunneling vibration signals were identified as the mine monitoring signal set.

[0014] Optionally, the acquisition of real-time mine images and mine monitoring signal sets using the microseismic monitoring unit includes:

[0015] Determine whether there is a pre-constructed mine static monitoring time sequence that is the same as the real-time monitoring time;

[0016] If there is a mine static monitoring time that is the same as the real-time monitoring time in the mine static monitoring time series, then multiple static mine environmental detectors are used to receive microseismic monitoring signals, analyze the microseismic monitoring signals, and obtain real-time mine images based on the microseismic monitoring signals, and confirm the microseismic monitoring signals as the mine monitoring signal set.

[0017] Optionally, determining whether there exists a pre-constructed mine static monitoring time series that is the same as the real-time monitoring time includes:

[0018] Based on the real-time monitoring time, the adjacent historical hidden danger monitoring time period is obtained. The latest historical hidden danger monitoring time that precedes the real-time monitoring time is extracted from the adjacent historical hidden danger monitoring time period. The hidden danger monitoring frequency control parameter is calculated based on the latest historical hidden danger monitoring time and the pre-constructed hidden danger monitoring frequency control parameter calculation formula. The target monitoring time is obtained using the hidden danger monitoring frequency control parameter. The target monitoring time is imported into the pre-constructed initial mine static monitoring time sequence to obtain the mine static monitoring time sequence. The mine static monitoring time sequence includes multiple mine static monitoring times.

[0019] Compare the target monitoring time with the real-time monitoring time. If the target monitoring time lags behind the real-time monitoring time, it is confirmed that there is no mine static monitoring time in the mine static monitoring time series that is the same as the real-time monitoring time. Otherwise, it is confirmed that there is a mine static monitoring time in the mine static monitoring time series that is the same as the real-time monitoring time.

[0020] Optionally, the step of constructing a dataset for identifying safety hazard points based on a set of mine monitoring signals, a sequence of mine images, and a sequence of safety hazard point identifiers includes:

[0021] A classification operation is performed on the safety hazard point identification sequence to obtain a hazard classification identification sequence. The hazard classification identification sequence includes multiple hazard classification identifiers, and each hazard classification identifier corresponds to one or more classified safety hazard point identifiers.

[0022] Extract the hazard classification identifiers sequentially from the hazard classification identifier sequence to obtain the target safety hazard points. Perform the following operations on each target safety hazard point:

[0023] Extract all mine monitoring signal sets under the monitoring time sequence from the data acquisition unit, and identify the characteristic signals of the target safety hazard point in all mine monitoring signal sets under the monitoring time sequence to obtain the initial hazard signal sequence;

[0024] Real-time mine images are extracted sequentially from the mine image sequence. The safety hazard image regions corresponding to the target safety hazard points are identified from the extracted real-time mine images. The safety hazard image regions are summarized to obtain the initial safety hazard point evolution map sequence corresponding to the target safety hazard point.

[0025] The initial safety hazard point evolution diagram sequence and the initial hazard signal sequence are marked by one or more classified safety hazard point identifiers respectively, to obtain the identified hazard point evolution diagram sequence and the identified hazard signal sequence. The identified hazard point evolution diagram sequence and the identified hazard signal sequence are then summarized to obtain safety hazard point data.

[0026] By summarizing the data on safety hazard points, a dataset of identified safety hazard points corresponding to the hazard classification and identification sequence is obtained.

[0027] Optionally, if a safety warning instruction is generated based on the parsed dataset of identified safety hazards, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine; otherwise, the process returns to the step of obtaining the real-time monitoring time, including:

[0028] The following operations are performed on all safety hazard point data in the dataset identifying safety hazard points:

[0029] Analyze the data on safety hazard points to obtain the analysis results;

[0030] If the analysis result of the hidden danger point is the preset hidden danger result, a safety warning command will be issued to the initiator of the real-time environmental monitoring command of the mine.

[0031] Otherwise, based on the safety hazard data, historical hazard data for a preset historical period is extracted from the pre-constructed historical evolution database, the historical hazard data and the safety hazard data are integrated to obtain historical-real-time hazard integrated data, and the historical-real-time hazard integrated data is parsed to obtain historical-real-time hazard parsing results;

[0032] If the historical-real-time hazard point analysis results are the preset historical-real-time hazard results, a safety warning command will be issued to the initiator of the mine real-time environmental monitoring command.

[0033] Otherwise, the results of the analysis of potential hazards are summarized to obtain the comprehensive data corresponding to the dataset of identified safety hazard points under the monitoring time series. If the comprehensive data is the preset comprehensive safety warning data, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction of the mine.

[0034] If the comprehensive data is not the preset comprehensive data for safety warnings, then extract the historical comprehensive data for the preset historical period from the historical evolution database. If the historical comprehensive data is not the preset historical comprehensive data for safety warnings, then return to the step of obtaining the real-time monitoring time.

[0035] Optionally, the step of calculating the hazard monitoring frequency control parameters based on the latest historical hazard monitoring time and a pre-constructed hazard monitoring frequency control parameter calculation formula, and obtaining the target monitoring time using the hazard monitoring frequency control parameters, includes:

[0036] The data sequence of adjacent hidden danger points is obtained based on the adjacent historical hidden danger monitoring period. The adjacent hidden danger point data sequence includes multiple adjacent hidden danger point data, and the multiple adjacent hidden danger point data are sorted in chronological order. Each adjacent hidden danger point data includes: hidden danger point location characteristics, hidden danger point displacement characteristics, hidden danger point deformation characteristics, and hidden danger point frequency domain characteristics.

[0037] Extract the target data sequence from the data sequence of adjacent potential hazard points, and quantify the target data sequence to obtain the target numerical sequence. The target data sequence includes multiple target data of the same type, and the type of the target data is one of the following: potential hazard point location features, potential hazard point displacement features, potential hazard point deformation features, and potential hazard point frequency domain features.

[0038] Extract target values ​​sequentially from the target value sequence, and perform the following operations on each extracted target value:

[0039] Extract the next target value that lags behind the target value from the target value sequence to obtain the neighboring value. Calculate the difference between the target value and the neighboring value to obtain the first-order difference. Summarize the first-order differences to obtain the first difference sequence.

[0040] The first difference sequence is truncate using a preset sliding window and a preset sliding step size to obtain a sliding difference data sequence. Sliding difference data is extracted sequentially from the sliding difference data sequence, and the variance of the extracted sliding difference data is calculated to obtain the data change corresponding to the sliding difference data.

[0041] Summarize the data changes to obtain a data change sequence;

[0042] Based on the data change sequence and the first difference sequence, the first adjustment coefficient and the second adjustment coefficient are determined, and the calculation formula for the control parameter of the hidden danger monitoring frequency is constructed according to the first adjustment coefficient and the second adjustment coefficient.

[0043] The hazard monitoring frequency control parameters are calculated using the formula for calculating hazard monitoring frequency control parameters and the latest historical hazard monitoring times.

[0044] The target monitoring time is obtained based on the pre-constructed variable frequency control sequence and the frequency control parameters for hazard monitoring.

[0045] Optionally, the step of obtaining the target monitoring time based on the pre-constructed variable frequency control sequence and the hidden danger monitoring frequency control parameters includes:

[0046] Acquire multi-source monitoring data, extract a reference mine roadway set from the multi-source monitoring data, and sequentially extract reference mine roadways from the reference mine roadway set. Perform the following operations on each extracted reference mine roadway:

[0047] Obtain the reference frequency parameter set corresponding to the reference mine roadway, and calculate the reference control factor set corresponding to the reference frequency parameter set based on the calculation formula of the hidden danger monitoring frequency control parameter, wherein the reference frequency parameter and the reference control factor correspond one-to-one.

[0048] A mapping relationship is constructed between the reference frequency parameter set and the reference control factor set to obtain an initial frequency-factor dataset. The initial frequency-factor dataset is simplified to obtain a frequency-factor set, wherein the frequency-factor set includes multiple frequency-factors, and each frequency-factor includes: a reference frequency parameter and a reference control factor.

[0049] By summarizing the frequency-factor sets, we obtain the frequency-factor set group corresponding to the reference mine roadway set;

[0050] A reference frequency-based clustering operation is performed on the frequency-factor set using a pre-constructed clustering method and a preset number of cluster centers to obtain multiple frequency-factor clusters. The reference frequency-based clustering operation is a K-means clustering operation with the reference frequency as the variable, and the number of multiple frequency-factor clusters is equal to the number of cluster centers.

[0051] Frequency-factor clusters are extracted sequentially from multiple frequency-factor clusters, and the clustering frequency intervals and clustering factor intervals of the extracted frequency-factor clusters are identified. The mean of the clustering frequency intervals is calculated to obtain the fuzzy reference frequency.

[0052] By associating the clustering factor interval with the fuzzy reference frequency, a unit frequency conversion adjustment interval is obtained. The unit frequency conversion adjustment intervals are then summarized and integrated to obtain the frequency conversion adjustment sequence.

[0053] The target interval is obtained by finding the unit frequency conversion adjustment range where the hidden danger monitoring frequency control parameter is located in the frequency conversion adjustment sequence, and the fuzzy reference frequency corresponding to the target interval is confirmed as the target frequency. The target monitoring time is calculated based on the target frequency and the latest historical hidden danger monitoring time.

[0054] To achieve the above objectives, the present invention also provides a real-time monitoring and early warning system for geological disasters based on the fusion of big data and environmental characteristics, comprising:

[0055] The coal mine monitoring environment module is used to receive real-time environmental monitoring commands from the mine and confirm the coal mine monitoring environment based on the real-time environmental monitoring commands. The coal mine monitoring environment includes: a tunneling machine, tunneling roadways, and a data acquisition unit. The data acquisition unit includes: a tunneling acquisition unit and a shutdown acquisition unit. The tunneling acquisition unit includes multiple tunneling geophones, and the shutdown acquisition unit includes multiple static mine environment geophones and a microseismic monitoring unit.

[0056] The tunneling acquisition module is used to acquire real-time monitoring time. If the real-time monitoring time is within a preset tunneling period, the tunneling vibration signal is received using the preset tunneling acquisition period and the tunneling acquisition unit. Based on the tunneling vibration signal, a real-time mine image and a mine monitoring signal set are acquired. The tunneling vibration signal comes from the vibration of the tunneling machine.

[0057] The static acquisition module is used to acquire real-time mine images and mine monitoring signal sets using the microseismic monitoring unit if the real-time monitoring time is not during the tunneling period. The mine monitoring signal set is obtained by the microseismic monitoring unit.

[0058] The monitoring feedback module is used to summarize the real-time mine images according to a preset monitoring time sequence to obtain a mine image sequence, identify all safety hazard point markers in the mine image sequence to obtain a safety hazard point marker sequence, and construct a safety hazard point marker dataset based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence. The safety hazard point dataset is parsed. If a safety warning instruction is generated based on the parsed marker hazard dataset, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, the process returns to the step of obtaining the real-time monitoring time, thus completing a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental features.

[0059] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0060] A memory for storing at least one instruction; and a processor for executing the instructions stored in the memory to implement the above-described method for real-time monitoring and early warning of geological disasters based on the fusion of big data and environmental characteristics.

[0061] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned method for real-time monitoring and early warning of geological disasters based on the fusion of big data and environmental features.

[0062] To address the problems described in the background art, this invention acquires real-time monitoring data. If the real-time monitoring time falls within a preset tunneling period, the invention utilizes the preset tunneling acquisition period and an on-the-spot acquisition unit to receive tunneling vibration signals. Based on these vibration signals, real-time mine images and a mine monitoring signal set are acquired. The tunneling vibration signals originate from the vibration of the tunneling machine. During tunneling, this invention employs an on-the-spot acquisition unit to detect the geological structure beneath the rock strata, approximating the vibration source as equivalently generated by the tunneling machine, thereby reducing dependence on active vibration sources. If the real-time monitoring time does not fall within the tunneling period, a microseismic monitoring unit is used to acquire real-time mine images and a mine monitoring signal set. The mine monitoring signal set is obtained by the microseismic monitoring unit. This invention uses microseismic monitoring technology to monitor the tunneling roadway outside of the tunneling period to detect the geological structure beneath the rock strata, achieving real-time monitoring of the tunneling roadway. The real-time mine images are aggregated according to a preset monitoring time sequence to obtain a mine image sequence. All safety hazard point markers in the mine image sequence are identified to obtain a safety hazard point marker sequence. This invention categorizes safety hazard point markers for the same hazard point in the tunneling roadway, thereby facilitating the observation of the evolution of the same hazard point. A safety hazard point marker dataset is constructed based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence. In this embodiment, the monitoring frequency is adjusted; the monitoring frequency is increased when the collected data fluctuates significantly, and decreased when the data is relatively stable, thereby accurately monitoring multi-dimensional data of the coal mine roadway while conserving detection energy. The invention parses and identifies a dataset of safety hazard points. If a safety warning instruction is generated based on the parsed dataset, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, it returns to the step of obtaining the real-time monitoring time, thus completing a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental features. This invention, based on the premise that the hazard point parsing result is not a preset hazard result, combines historical hazard point data to analyze the characteristic hazard points corresponding to the hazard classification labels in the safety hazard point data. It adopts a progressively dimensional data analysis method and also combines historical comprehensive data to analyze all characteristic hazard points corresponding to multiple hazard classification labels in the identified safety hazard point dataset. This achieves both grasping point-to-point data hazard fluctuations and analyzing the environmental characteristics of the current coal mine roadway based on big data and historical data, thereby monitoring the morphology and other characteristics of the tunneling roadway and providing data support for determining whether a geological disaster has occurred. Therefore, this invention can utilize on-the-spot acquisition signals and microseismic monitoring technology to monitor the environmental characteristics of the tunneling roadway in real time, and use big data to detect hazard in the environmental characteristics, thereby achieving real-time monitoring and warning of underground mine safety hazards. Attached Figure Description

[0063] Figure 1 This is a flowchart illustrating a method for real-time monitoring and early warning of geological disasters based on the fusion of big data and environmental features, provided in an embodiment of the present invention.

[0064] Figure 2 A functional module diagram of a real-time monitoring and early warning system for geological disasters based on the fusion of big data and environmental features, provided in an embodiment of the present invention;

[0065] Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental features, as provided in an embodiment of the present invention.

[0066] Explanation of reference numerals in the attached figures:

[0067] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0068] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0069] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0070] This application provides a method for real-time monitoring and early warning of geological disasters based on the fusion of big data and environmental features. The executing entity of this method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or server device, and the software may be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0071] Reference Figure 1 The diagram shown is a flowchart illustrating a real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental features, according to an embodiment of the present invention. In this embodiment, the real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental features includes:

[0072] S1. Receive real-time environmental monitoring instructions for the mine, and confirm the coal mine monitoring environment based on the real-time environmental monitoring instructions. The coal mine monitoring environment includes: tunneling machine, tunneling roadway, and data acquisition unit. The data acquisition unit includes: tunneling acquisition unit and shutdown acquisition unit. The tunneling acquisition unit includes multiple tunneling detectors, and the shutdown acquisition unit includes multiple static mine environment detectors and microseismic monitoring unit.

[0073] Understandably, the real-time environmental monitoring command for the mine is issued by the monitoring personnel of the tunneling roadway to monitor the environment of the tunneling roadway. According to the real-time environmental monitoring command for the mine, this embodiment of the invention monitors the coal mine environment in real time. If the monitoring results indicate that there is a safety hazard in the tunneling roadway, such as a displacement exceeding the standard displacement occurring at a certain point inside the rock strata during tunneling, or a large amount of debris falling or vibrations that do not meet the human-specified standards occurring at the tunneling location, then an early warning is issued to the initiator of the real-time environmental monitoring command for the mine. Therefore, this invention is a method for monitoring the safety status of underground mine roadways and can be used in monitoring devices to implement the method described in this invention.

[0074] Furthermore, the on-the-go data acquisition unit is used to acquire data from the tunneling roadway during tunneling machine operation, and the on-the-go geophone is used to receive signals emitted by the tunneling machine's vibrations. The off-the-go data acquisition unit is used to acquire geological conditions from the tunneling roadway when the tunneling machine is not operating. Specifically, the off-the-go data acquisition unit uses a microseismic monitoring unit to acquire geological conditions, and specifically, it uses multiple static mine environment geophones to receive signals.

[0075] S2, Obtain real-time monitoring time.

[0076] Understandably, the real-time monitoring time is the current time.

[0077] S3. If the real-time monitoring time is within a preset tunneling period, the preset tunneling acquisition period and the accompanying acquisition unit are used to receive the tunneling vibration signal, and real-time mine images and mine monitoring signal sets are obtained based on the tunneling vibration signal, wherein the tunneling vibration signal comes from the vibration of the tunneling machine.

[0078] It should be noted that the tunneling period is a manually set time during which a tunneling machine excavates the tunnel. The tunneling acquisition period is the time during which the on-site acquisition unit receives tunneling vibration signals within the tunneling period. For example, if the tunneling work is scheduled for 8:00-11:00, the tunneling machine will not operate continuously during this period. For instance, when cutterhead replacement, maintenance, or encountering different rock strata are required, the tunneling machine may pause for, for example, 30 minutes. The time during which the tunneling machine does not pause is the tunneling acquisition period. The on-site geophone receives tunneling vibration signals with the tunneling machine's vibration as the source during the tunneling acquisition period. The tunneling vibration signal is the signal received by the on-site geophone during the tunneling acquisition period.

[0079] Furthermore, the acquisition of real-time mine images and mine monitoring signal sets based on tunneling vibration signals includes:

[0080] The tunneling vibration signal is fitted using a pre-constructed seismic interferometry technique to obtain a set of virtual shot signals. A low-pass filtering operation is performed on the set of virtual shot signals to obtain a denoised signal. The denoised signal is analyzed using a pre-constructed scattering imaging method, and the analyzed denoised signal is visualized to obtain a mine hazard detection image corresponding to the real-time monitoring time.

[0081] The mine hazard detection images were analyzed to obtain real-time mine images containing safety hazard point markers, and the tunneling vibration signals were identified as the mine monitoring signal set.

[0082] It should be noted that seismic interferometry is an existing technology, and its function is to generate data for imaging underground geological structures using tunneling vibration signals. The virtual shot signal set is the equivalent signal set generated when the tunneling location is fitted as an artificially excited seismic source, after cross-correlating the tunneling vibration signals received by multiple tunneling detectors. For example, after the vibration of the tunneling machine is equivalently generated as an artificially excited seismic source, tunneling detectors A, B, C, and D all receive the tunneling vibration signals. Using tunneling detector A as a virtual source and tunneling detectors B, C, and D as receiving points, three virtual shot signals are generated. In this case, one virtual shot signal describes the cross-correlation result of a "virtual source-receiving point". Therefore, with the support of seismic interferometry, in this embodiment of the invention, the vibration generated by the tunneling machine during tunneling can be equivalently represented as a seismic source, thereby reducing the dependence on active seismic sources. The low-pass filtering operation is an operation to reduce noise in the virtual shot signal set. Optionally, in this embodiment of the invention, a Gaussian filter is used for filtering.

[0083] Specifically, the method of analyzing the denoised signal using seismic scattering wave imaging is an existing technology, so the process of obtaining the mine hazard detection image corresponding to the real-time monitoring time will not be described in detail here.

[0084] For example, assuming the tunneling vibration signal is received with a time window of 10 seconds, and there are 3 on-the-go detectors, a set of 6 virtual shot signals can be generated. The set of virtual shot signals is then analyzed using a scattered wave imaging method (e.g., the reverse time migration method in the prior art), ultimately generating a mine hazard detection image reflecting the underground structure. A portion of the mine hazard detection image corresponding to each virtual shot signal is used in the overall mine hazard detection image imaging process. Therefore, by superimposing the mine hazard detection image sets corresponding to each virtual shot signal, a single mine hazard detection image reflecting the underground geological structure can be obtained.

[0085] Furthermore, the purpose of analyzing the mine hazard detection images to obtain real-time mine images containing safety hazard point markers is to identify material components in the mine hazard detection images that differ from those in conventional tunneling roadways, such as collapse pillars and rock crushing zones. It should also be noted that the mine hazard detection images in this embodiment are the results of a single detection. Therefore, when analyzing the mine hazard detection images, all mine hazard detection images collected during the tunneling acquisition period prior to the real-time monitoring time can be used as references to analyze whether a safety hazard has occurred at the real-time monitoring time.

[0086] Specifically, the safety hazard point identifier refers to the identifier used to identify hazard points. The information contained in the safety hazard point identifier can be artificially defined, such as: region, type, roadway code, tunneling machine code, predicted spatial location, etc., thereby generating the safety hazard point identifier: [N City - Collapse Column - Roadway N223 - Tunneling Machine No. 1 - (x, y, z)].

[0087] S4. If the real-time monitoring time is not within the tunneling period, the real-time mine image and mine monitoring signal set are obtained by the microseismic monitoring unit, wherein the mine monitoring signal set is obtained by the microseismic monitoring unit.

[0088] Furthermore, the acquisition of real-time mine images and mine monitoring signal sets using the microseismic monitoring unit includes:

[0089] Determine whether there is a pre-constructed mine static monitoring time sequence that is the same as the real-time monitoring time;

[0090] If there is a mine static monitoring time that is the same as the real-time monitoring time in the mine static monitoring time series, then multiple static mine environmental detectors are used to receive microseismic monitoring signals, analyze the microseismic monitoring signals, and obtain real-time mine images based on the microseismic monitoring signals, and confirm the microseismic monitoring signals as the mine monitoring signal set.

[0091] It should be explained that a microseismic monitoring unit is a unit that uses micro-vibration monitoring technology to monitor tunnels that are not in the tunneling phase. The microseismic monitoring signal is the signal received by multiple static mine environment detectors corresponding to the microseismic monitoring unit when monitoring tunnels being excavated.

[0092] Furthermore, determining whether there exists a mine static monitoring time in the pre-constructed mine static monitoring time series that is the same as the real-time monitoring time includes:

[0093] Based on the real-time monitoring time, the adjacent historical hidden danger monitoring time period is obtained. The latest historical hidden danger monitoring time that precedes the real-time monitoring time is extracted from the adjacent historical hidden danger monitoring time period. The hidden danger monitoring frequency control parameter is calculated based on the latest historical hidden danger monitoring time and the pre-constructed hidden danger monitoring frequency control parameter calculation formula. The target monitoring time is obtained using the hidden danger monitoring frequency control parameter. The target monitoring time is imported into the pre-constructed initial mine static monitoring time sequence to obtain the mine static monitoring time sequence. The mine static monitoring time sequence includes multiple mine static monitoring times.

[0094] Compare the target monitoring time with the real-time monitoring time. If the target monitoring time lags behind the real-time monitoring time, it is confirmed that there is no mine static monitoring time in the mine static monitoring time series that is the same as the real-time monitoring time. Otherwise, it is confirmed that there is a mine static monitoring time in the mine static monitoring time series that is the same as the real-time monitoring time.

[0095] It should be noted that the adjacent historical hazard monitoring period is a period prior to the real-time monitoring time, and the latest historical hazard monitoring time is the last time that monitoring was carried out within the adjacent historical hazard monitoring period. For example, if the real-time monitoring time is 8:00, then the adjacent historical hazard monitoring period is 7:30-8:00. Within 7:30-8:00, monitoring was carried out at 7:30, 7:40, and 7:50. Among them, 7:50 is the last time that monitoring was carried out in terms of time within the adjacent historical hazard monitoring period. Therefore, the latest historical hazard monitoring time is 7:50.

[0096] It should be noted that the frequency of the microseismic monitoring unit can be adjusted according to the hidden danger monitoring frequency control parameters, thereby calculating the next time the tunnel needs to be inspected, i.e., the target monitoring time. The initial mine static monitoring sequence is a sequence of all monitoring times before the real-time monitoring time, arranged in chronological order. Therefore, the mine static monitoring sequence is the sequence obtained by importing the target monitoring time into the initial mine static monitoring sequence, and multiple mine static monitoring times are also arranged in chronological order. At this time, by comparing the target monitoring time with the real-time monitoring time, it can be determined whether there is a mine static monitoring time in the mine static monitoring sequence that is the same as the real-time monitoring time.

[0097] For example, if the target monitoring time is 8:20, the real-time monitoring time is 8:18, and the latest historical hidden danger monitoring time in the initial mine static monitoring time sequence is 8:10, the mine static monitoring time sequence includes: [8:10, 8:20]. Then the target monitoring time lags behind the real-time monitoring time, indicating that it is not yet time to use the microseismic monitoring unit for monitoring. Therefore, it is confirmed that there is no mine static monitoring time in the mine static monitoring time sequence that is the same as the real-time monitoring time.

[0098] Furthermore, the step of calculating the hazard monitoring frequency control parameters based on the latest historical hazard monitoring time and the pre-constructed hazard monitoring frequency control parameter calculation formula, and obtaining the target monitoring time using the hazard monitoring frequency control parameters, includes:

[0099] The data sequence of adjacent hidden danger points is obtained based on the adjacent historical hidden danger monitoring period. The adjacent hidden danger point data sequence includes multiple adjacent hidden danger point data, and the multiple adjacent hidden danger point data are sorted in chronological order. Each adjacent hidden danger point data includes: hidden danger point location characteristics, hidden danger point displacement characteristics, hidden danger point deformation characteristics, and hidden danger point frequency domain characteristics.

[0100] Extract the target data sequence from the data sequence of adjacent potential hazard points, and quantify the target data sequence to obtain the target numerical sequence. The target data sequence includes multiple target data of the same type, and the type of the target data is one of the following: potential hazard point location features, potential hazard point displacement features, potential hazard point deformation features, and potential hazard point frequency domain features.

[0101] Extract target values ​​sequentially from the target value sequence, and perform the following operations on each extracted target value:

[0102] Extract the next target value that lags behind the target value from the target value sequence to obtain the neighboring value. Calculate the difference between the target value and the neighboring value to obtain the first-order difference. Summarize the first-order differences to obtain the first difference sequence.

[0103] The first difference sequence is truncate using a preset sliding window and a preset sliding step size to obtain a sliding difference data sequence. Sliding difference data is extracted sequentially from the sliding difference data sequence, and the variance of the extracted sliding difference data is calculated to obtain the data change corresponding to the sliding difference data.

[0104] Summarize the data changes to obtain a data change sequence;

[0105] Based on the data change sequence and the first difference sequence, the first adjustment coefficient and the second adjustment coefficient are determined, and the calculation formula for the control parameter of the hidden danger monitoring frequency is constructed according to the first adjustment coefficient and the second adjustment coefficient.

[0106] The hazard monitoring frequency control parameters are calculated using the formula for calculating hazard monitoring frequency control parameters and the latest historical hazard monitoring times.

[0107] The target monitoring time is obtained based on the pre-constructed variable frequency control sequence and the frequency control parameters for hazard monitoring.

[0108] It should be noted that the adjacent hazard point data sequence is a sequence composed of data obtained by the data acquisition unit during the adjacent historical hazard monitoring period.

[0109] For example, the location characteristics of the potential hazard point can be represented as a set of monitored location data. For instance, when using a microseismic monitoring unit, if the expansion of cracks on the inner wall surface of a tunnel is detected, recording the location of the crack expansion yields the location characteristics of the potential hazard point. Another example is using a tunneling data acquisition unit to collect the three-dimensional coordinates of a collapse column within the rock in geodetic coordinates. The displacement characteristics of the potential hazard point can be the maximum movement distance of the expanded edge of the crack on the inner wall surface of the tunnel. The deformation characteristics of the potential hazard point can be the area of ​​the crack after its expansion, or the volume of the collapse column. The frequency domain characteristics of the potential hazard point can be the vibration signal detected by the microseismic monitoring unit. Therefore, through the above examples, it can be understood that the location characteristics, displacement characteristics, deformation characteristics, and frequency domain characteristics of the potential hazard point are all artificially defined sets of data, unified within the data of adjacent potential hazard points. The purpose is to monitor the morphology and other characteristics of the tunneling roadway, thereby providing data support for determining whether a geological disaster has occurred. When selecting the location features, displacement features, deformation features, and frequency domain features of potential hazards, selection can be based on expert experience, big data, or methods such as academic papers and books. The features included in the data of neighboring potential hazards include, but are not limited to, the location features, displacement features, deformation features, and frequency domain features of potential hazards exemplified in this embodiment of the invention.

[0110] Specifically, depending on the different target data sequences, the numericalization of the target data sequences can be achieved through different existing technologies. For example, if the deformation characteristics of the hidden danger point are the crack area after the cracks on the inner wall surface of the tunnel are expanded when monitored using a microseismic monitoring unit, the value of the crack area can be used as the target value to form a target numerical sequence, thereby simplifying the subsequent calculation of the hidden danger monitoring frequency control parameters and reducing computational energy consumption.

[0111] Furthermore, since each neighboring hazard point data includes hazard point location features, hazard point displacement features, hazard point deformation features, and hazard point frequency domain features, and since the purpose of extracting the target data sequence in this embodiment of the invention is to study the change of a certain feature in the neighboring hazard point data, for example, if the goal is to study the change of hazard point displacement features, then all target data in the target data sequence is a sequence composed of hazard point displacement features extracted sequentially from each neighboring hazard point data of multiple neighboring hazard point data. Therefore, the target data sequence includes multiple target data of the same type, and the type of the target data is one of hazard point location features, hazard point displacement features, hazard point deformation features, and hazard point frequency domain features. It should also be noted that this embodiment of the invention performs the same operation on each feature in the neighboring hazard point data. Since the target data sequence is arranged in chronological order, the numerical target data sequence, i.e., the target numerical sequence, is also arranged instantaneously in chronological order.

[0112] For example, if the target value sequence is [j, k, l], when the target value is j, the neighboring value is k. The first difference is the target value j minus the neighboring value k. And so on, the expression for the first difference sequence is: [jk, kl].

[0113] Furthermore, since the sliding window and the sliding step size are existing technologies, their specific meanings will not be explained here. The sliding truncation is an operation that extracts values ​​from a difference sequence based on the sliding window and the sliding step size. For example, suppose the first-order difference sequence is [n, m, o, p], the sliding step size is 1, and the size of the sliding window is 2. Then, the sliding difference data sequence obtained by sliding truncation of the first-order difference sequence is [n, m], [m, o], and [o, p]. At this time, [n, m], [m, o], and [o, p] are all sliding difference data. Therefore, the meaning of sliding difference data is the data in the sliding difference data sequence. The data change corresponding to the sliding difference data is the variance of all data in the sliding difference data. The method for calculating the variance is existing technology and will not be elaborated here.

[0114] It should also be noted that the data change sequence is a sequence obtained by arranging all data changes in chronological order of acquisition. The first and second adjustment coefficients are correction coefficients in the calculation formula for the frequency control parameters of hazard monitoring, and can be obtained using various existing technologies. For example, they can be obtained through empirical formulas, or by designing orthogonal experiments, utilizing data from a historical evolution database, and combining the least squares method to fit the correction coefficients. This embodiment of the invention does not limit these methods.

[0115] For example, the first-order difference sequence is [n, m, o, p], the sliding step size is 1, and the sliding window size is 2. The sliding difference data sequence is [n, m], [m, o], and [o, p]. The difference corresponding to the sliding difference data sequence is calculated to obtain [n1, m1, o1]. At this time, various existing technologies can be used to complete the data change sequence. This embodiment of the invention does not limit this. In this example, the first-order difference p is used to complete [n1, m1, o1] to obtain the data change sequence [n1, m1, o1, p]. Therefore, if the extracted first-order difference is n, the corresponding data change in the data change sequence can be obtained as n1. At this time, the data change function is the function obtained by fitting [n, m] to the data, and the data change function is a linear function. The slope of the data change function is used as the parameter in the calculation formula of the initial hidden danger monitoring frequency control parameter.

[0116] Understandably, since each neighboring hazard point data includes hazard point location features, hazard point displacement features, hazard point deformation features, and hazard point frequency domain features, and one target data sequence corresponds to one feature, the total number of target data is equal to the number of environmental features (e.g., hazard point location features, hazard point displacement features, hazard point deformation features, and hazard point frequency domain features) contained in the neighboring hazard point data. Furthermore, since each target data sequence can be used to calculate an initial hazard monitoring frequency control parameter calculation formula, the initial hazard monitoring frequency control parameter calculation formula is equal to the number of environmental features contained in the neighboring hazard point data. The calculation formulas for all initial hazard monitoring frequency control parameters corresponding to the data sequences of adjacent hazard points are merged. The resulting hazard monitoring frequency control parameter calculation formulas integrate the environmental characteristics of the tunneling roadway during the adjacent historical hazard monitoring period. Based on the fluctuations of environmental characteristics (e.g., first-order difference, data change, slope of the data change function), the target monitoring time is calculated. This allows for increasing the monitoring frequency when the collected data has large fluctuations and decreasing the monitoring frequency when the data is relatively stable, thereby enabling accurate monitoring of multidimensional data of coal mine roadways while saving detection energy.

[0117] Furthermore, the acquisition of the target monitoring time based on the pre-constructed variable frequency control sequence and the hidden danger monitoring frequency control parameters includes:

[0118] Acquire multi-source monitoring data, extract a reference mine roadway set from the multi-source monitoring data, and sequentially extract reference mine roadways from the reference mine roadway set. Perform the following operations on each extracted reference mine roadway:

[0119] Obtain the reference frequency parameter set corresponding to the reference mine roadway, and calculate the reference control factor set corresponding to the reference frequency parameter set based on the calculation formula of the hidden danger monitoring frequency control parameter, wherein the reference frequency parameter and the reference control factor correspond one-to-one.

[0120] A mapping relationship is constructed between the reference frequency parameter set and the reference control factor set to obtain an initial frequency-factor dataset. The initial frequency-factor dataset is simplified to obtain a frequency-factor set, wherein the frequency-factor set includes multiple frequency-factors, and each frequency-factor includes: a reference frequency parameter and a reference control factor.

[0121] By summarizing the frequency-factor sets, we obtain the frequency-factor set group corresponding to the reference mine roadway set;

[0122] A reference frequency-based clustering operation is performed on the frequency-factor set using a pre-constructed clustering method and a preset number of cluster centers to obtain multiple frequency-factor clusters. The reference frequency-based clustering operation is a K-means clustering operation with the reference frequency as the variable, and the number of multiple frequency-factor clusters is equal to the number of cluster centers.

[0123] Frequency-factor clusters are extracted sequentially from multiple frequency-factor clusters, and the clustering frequency intervals and clustering factor intervals of the extracted frequency-factor clusters are identified. The mean of the clustering frequency intervals is calculated to obtain the fuzzy reference frequency.

[0124] By associating the clustering factor interval with the fuzzy reference frequency, a unit frequency conversion adjustment interval is obtained. The unit frequency conversion adjustment intervals are then summarized and integrated to obtain the frequency conversion adjustment sequence.

[0125] The target interval is obtained by finding the unit frequency conversion adjustment range where the hidden danger monitoring frequency control parameter is located in the frequency conversion adjustment sequence, and the fuzzy reference frequency corresponding to the target interval is confirmed as the target frequency. The target monitoring time is calculated based on the target frequency and the latest historical hidden danger monitoring time.

[0126] It is understood that multi-source monitoring data is data obtained from monitoring equipment in tunneling roadways in different regions. During the extraction of a reference mine roadway set from the multi-source monitoring data, experts or technicians can screen out tunneling roadways that can serve as references for the tunneling roadways in this embodiment of the invention, thus obtaining a reference mine roadway set. It should also be noted that the number of reference mine roadways in the set is greater than or equal to two. In the process of screening reference mine roadways, a comprehensive consideration can be made from various aspects such as geological conditions (e.g., rock hardness, compressive strength, groundwater conditions), tunneling machine parameters (e.g., cutterhead thrust, torque, speed), and tunneling roadways in the same mining area or adjacent strata, thereby selecting reference mine roadways that have reference value for the tunneling roadways in this embodiment of the invention. This embodiment of the invention does not limit this.

[0127] Furthermore, the reference frequency parameter set comprises the frequency monitored by the microseismic monitoring unit in the reference mine roadway, and the set of environmental characteristics in the reference mine roadway at the monitored frequency, including but not limited to: hazard point location characteristics, hazard point displacement characteristics, hazard point deformation characteristics, and hazard point frequency domain characteristics. The reference frequency is the frequency monitored by the microseismic monitoring unit in the reference mine roadway. The calculation of the reference control factor set corresponding to the reference frequency parameter set based on the hazard monitoring frequency control parameter calculation formula allows for the calculation of the corresponding reference control factor based on a reference frequency parameter. This is consistent with the method in this embodiment of the invention for calculating the hazard monitoring frequency control parameter using the hazard monitoring frequency control parameter calculation formula, the latest historical hazard monitoring time, and the pre-constructed hazard monitoring frequency control parameter calculation formula.

[0128] Furthermore, the method for constructing the mapping relationship between the reference frequency parameter set and the reference control factor set to obtain the initial frequency-factor dataset is as follows: store the reference frequency parameters and the corresponding reference control factors in the same data, and the data obtained at this time is the initial frequency-factor data.

[0129] Specifically, this invention sequentially extracts initial frequency-factor data from the initial frequency-factor dataset, and extracts the reference frequency and reference control factor from the initial frequency-factor data. It then correlates the reference frequency and reference control factor to obtain frequency-factors, and summarizes the frequency-factors to obtain a frequency-factor set. Since the preceding embodiments of this invention have established the relationship between the reference frequency parameter and the reference control factor using the calculation formula for the hidden danger monitoring frequency control parameter, in subsequent analysis, the frequency monitored by the microseismic monitoring unit in the reference frequency parameter is used as the reference frequency and stored together with the reference control factor. This provides data support for subsequently confirming the target monitoring time based on the hidden danger monitoring frequency control parameter corresponding to the tunneling roadway.

[0130] It should be noted that the clustering method is K-means clustering, and the number of cluster centers can be set manually, for example, 5 cluster centers. The reference frequency-based clustering operation is a K-means clustering operation with the reference frequency as the variable, and the K-means clustering operation with the reference frequency as the variable is existing technology, which will not be described in detail here.

[0131] Furthermore, a frequency-factor cluster is a cluster obtained after clustering operations, and a frequency-factor cluster includes multiple frequency-factors. The clustering frequency interval is a continuous one-dimensional interval formed by the largest and smallest reference frequencies in the frequency-factor cluster, and the clustering factor interval is a continuous one-dimensional interval formed by the largest and smallest reference control factors in the frequency-factor cluster. The calculation of the mean of the clustering frequency interval is based on existing technology; the average of the largest and smallest reference frequencies can be calculated to obtain the fuzzy reference frequency. The method for associating the clustering factor interval with the fuzzy reference frequency to obtain the unit frequency conversion adjustment interval is as follows: the fuzzy reference frequency and the clustering factor interval are stored in a data store, which is the unit frequency conversion adjustment interval.

[0132] It should be noted that in the process of summarizing and integrating the unit frequency conversion control intervals to obtain the frequency conversion control sequence, it is possible that directly putting all the clustering factor intervals in the unit frequency conversion control interval together cannot form a continuous interval. For example, if the first clustering factor interval is [0, a] and the second clustering factor interval is [b, c], and a is greater than b, then there is an intersection [b, a] between the first and second clustering factor intervals. The intersection part is corresponding to the higher frequency fuzzy reference frequency. That is, the [a, b] part uses the higher frequency clustering factor interval between the first and second clustering factor intervals as the fuzzy reference frequency of [a, b]. For example, if the fuzzy reference frequency corresponding to the second clustering factor interval is greater than the fuzzy reference frequency of the first clustering factor interval, then the fuzzy reference frequency corresponding to the second clustering factor interval is used as the fuzzy reference frequency of the intersection of the first and second clustering factor intervals.

[0133] For example, suppose that a is less than b but not equal to b. Then, there exists a reference control factor in the interval [a, b] between the first and second clustering factor intervals, but no corresponding fuzzy reference frequency can be found. In this case, the higher fuzzy reference frequency between the two adjacent clustering factor intervals is used. Therefore, it is necessary to summarize and integrate the unit frequency conversion control intervals to obtain the frequency conversion control sequence.

[0134] Understandably, each reference control factor in the reference control factor set can be found to have a fuzzy reference frequency in the frequency conversion control sequence, and values ​​other than those in the reference control factor set can also be found to have a fuzzy reference frequency in the frequency conversion control sequence. Therefore, this embodiment of the invention selects reference tunneling roadways from multi-source monitoring data, and calculates the reference control factor in reverse using the same formula for calculating the hidden danger monitoring frequency control parameter, given the known monitoring frequency of the reference mine roadway. This provides a basis for calculating the target monitoring time by setting the frequency of the microseismic monitoring unit using the hidden danger monitoring frequency control parameter for the tunneling roadway currently being excavated.

[0135] Furthermore, once the target frequency and the latest historical hazard monitoring time are known, the specific time of the next monitoring can be calculated. The specific calculation method is based on existing technology and will not be elaborated here.

[0136] S5. Summarize the real-time mine images according to the preset monitoring time sequence to obtain a mine image sequence.

[0137] It should be explained that the monitoring sequence is a sequence obtained by sorting the data in chronological order within a preset time period. For example, the preset time period is 24 hours.

[0138] S6. Identify all safety hazard point markers in the mine image sequence to obtain a safety hazard point marker sequence. Construct a safety hazard point marker dataset based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence.

[0139] Furthermore, the step of constructing a dataset of identified safety hazard points based on the mine monitoring signal set, mine image sequence, and safety hazard point identification sequence includes:

[0140] A classification operation is performed on the safety hazard point identification sequence to obtain a hazard classification identification sequence. The hazard classification identification sequence includes multiple hazard classification identifiers, and each hazard classification identifier corresponds to one or more classified safety hazard point identifiers.

[0141] Extract the hazard classification identifiers sequentially from the hazard classification identifier sequence to obtain the target safety hazard points. Perform the following operations on each target safety hazard point:

[0142] Extract all mine monitoring signal sets under the monitoring time sequence from the data acquisition unit, and identify the characteristic signals of the target safety hazard point in all mine monitoring signal sets under the monitoring time sequence to obtain the initial hazard signal sequence;

[0143] Real-time mine images are extracted sequentially from the mine image sequence. The safety hazard image regions corresponding to the target safety hazard points are identified from the extracted real-time mine images. The safety hazard image regions are summarized to obtain the initial safety hazard point evolution map sequence corresponding to the target safety hazard point.

[0144] The initial safety hazard point evolution diagram sequence and the initial hazard signal sequence are marked by one or more classified safety hazard point identifiers respectively, to obtain the identified hazard point evolution diagram sequence and the identified hazard signal sequence. The identified hazard point evolution diagram sequence and the identified hazard signal sequence are then summarized to obtain safety hazard point data.

[0145] By summarizing the data on safety hazard points, a dataset of identified safety hazard points corresponding to the hazard classification and identification sequence is obtained.

[0146] It should be noted that a single real-time mine image may contain one, zero, or multiple safety hazard point markers. However, for a particular hazard point, if one safety hazard point marker exists in one real-time mine image, then the corresponding safety hazard point marker will also exist in one or more other real-time mine images within the mine image sequence. Therefore, it is necessary to classify the safety hazard point markers for the same hazard point in the tunneling roadway to facilitate observation of the evolution of the same hazard point. Therefore, the embodiments of the invention perform a classification operation on the safety hazard point marker sequence to obtain a hazard classification marker sequence. The hazard classification marker represents the total marker for the same hazard point, while one or more classified safety hazard point markers corresponding to one hazard classification marker represent the markers of all hazard points for the same hazard point in the mine image sequence.

[0147] For example, if there is an existing hazard point, a collapse column, identified as D1 in a real-time mine image, then this collapse column will also appear in other real-time mine images in the mine image sequence, identified as hazard points D2, D3, D4, D5, D6, D7, D8, D9, D1, D1, D1, D1, D2, D1, D2, D1, D1, D2, D1, D3, D1, D1, D2 ... 24 Therefore, in the classification operation of the safety hazard point identification sequence, D1, D2, D3, and D... 24 Stored in a hazard classification identifier, D1, D2, D3, D 24 All of these are identifiers for the aforementioned classified safety hazard points. When subsequently identifying the characteristic signals of the target safety hazard point within the set of all mine monitoring signals under the aforementioned monitoring time sequence, the identified signals are D1, D2, D3, and D... 24 The corresponding characteristic signal is the signal used to characterize the collapse column corresponding to the hidden danger classification mark, which is collected by the tunneling acquisition unit and the shutdown acquisition unit in the signal acquisition unit under the monitoring time sequence. The characteristic signals are arranged according to the monitoring time sequence to obtain the initial hidden danger signal sequence.

[0148] It should be noted that the safety hazard image area is the image area in the real-time mine image used to represent the target safety hazard point. For example, if the target safety hazard point is a collapse column, then the safety hazard image area is the image corresponding to the area of ​​the collapse column in the real-time mine image, which can be achieved by cropping the real-time mine image.

[0149] Furthermore, the initial safety hazard evolution sequence is a sequence obtained by arranging the safety hazard image regions according to the monitoring time sequence. As understood earlier in this example, the hazard classification markers are essentially extracted from real-time mine images, and the safety hazard image regions are also extracted from real-time mine images. Therefore, it is easy to find which real-time mine image corresponds to the classified safety hazard marker from the detection image sequence. Then, based on the real-time mine image corresponding to the classified safety hazard marker, the corresponding safety hazard image region in the real-time mine image can be found. At this point, by using the classified hazard markers as text and marking them in the safety hazard image region, it is possible to mark the initial safety hazard using one or more classified safety hazard markers. The sequence of hazard point evolution diagrams yields the sequence of hazard point evolution diagrams. Similarly, the initial hazard signal sequence is identified based on the mine monitoring signal set. In the example above, it is a mine image sequence constructed based on microseismic monitoring signals or tunneling vibration signals from the mine monitoring signal set. Therefore, similar to using the multiple classified hazard point identifiers to mark the initial hazard point evolution diagram sequence, one or more classified hazard point identifiers can also be used to mark the initial hazard signal sequence to obtain the hazard signal sequence. The implementation process only requires finding the correspondence between the multiple classified hazard points and the initial hazard signal sequence, so it will not be elaborated here.

[0150] S7. Parse the dataset of identified safety hazards. If a safety warning instruction is generated based on the parsed dataset of identified hazards, then send a safety warning instruction to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, return to the step of obtaining the real-time monitoring time to complete a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental features.

[0151] Furthermore, if a safety warning instruction is generated based on the parsed dataset of identified safety hazards, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine; otherwise, the process returns to the step of obtaining the real-time monitoring time, including:

[0152] The following operations are performed on all safety hazard point data in the dataset identifying safety hazard points:

[0153] Analyze the data on safety hazard points to obtain the analysis results;

[0154] If the analysis result of the hidden danger point is the preset hidden danger result, a safety warning command will be issued to the initiator of the real-time environmental monitoring command of the mine.

[0155] Otherwise, based on the safety hazard data, historical hazard data for a preset historical period is extracted from the pre-constructed historical evolution database, the historical hazard data and the safety hazard data are integrated to obtain historical-real-time hazard integrated data, and the historical-real-time hazard integrated data is parsed to obtain historical-real-time hazard parsing results;

[0156] If the historical-real-time hazard point analysis results are the preset historical-real-time hazard results, a safety warning command will be issued to the initiator of the mine real-time environmental monitoring command.

[0157] Otherwise, the results of the analysis of potential hazards are summarized to obtain the comprehensive data corresponding to the dataset of identified safety hazard points under the monitoring time series. If the comprehensive data is the preset comprehensive safety warning data, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction of the mine.

[0158] If the comprehensive data is not the preset comprehensive data for safety warnings, then extract the historical comprehensive data for the preset historical period from the historical evolution database. If the historical comprehensive data is not the preset historical comprehensive data for safety warnings, then return to the step of obtaining the real-time monitoring time.

[0159] It should be noted that the hazard point analysis results are the results after analyzing the safety hazard point data, and can be stored in the form of charts, videos, text, etc. The content and display format of the hazard point analysis results are arranged according to the needs of monitoring personnel. For example, if the safety hazard point data is an initial safety hazard point evolution diagram sequence and an initial hazard signal sequence for a certain collapse column, the analysis can be performed by using existing technology to fit the initial safety hazard point evolution diagram sequence into an evolution video, or by constructing a two-dimensional fluctuation curve from the data features in the initial safety hazard point evolution diagram sequence (e.g., hazard point location features, hazard point displacement features, hazard point deformation features, and hazard point frequency domain features, etc.). Examples will not be listed here. The hazard result is a result where a hazard may exist in the hazard point analysis results, which is artificially set. For example, if the area of ​​a certain crack exists in the safety hazard point data, and the change value of the area of ​​that crack exceeds a manually set change value threshold, then the hazard point analysis result can be confirmed as the hazard result. Therefore, the specific setting method and standard of the hazard result can be determined in various ways, such as based on expert opinions and empirical formulas.

[0160] Furthermore, the safety early warning instruction is an instruction used to issue an early warning to the initiating end of the real-time environmental monitoring instruction in the mine, including the hazard analysis results and the data of hazards existing in the hazard analysis results. The historical evolution database is a database composed of the previous monitoring results of the current tunneling roadway. The historical time period is a time period set by humans to limit the time range for extracting historical hazard point data from the historical evolution database. The historical hazard point data is the collection of all data of the same hazard point corresponding to the hazard classification identifier in the historical evolution database within the historical time period. The integration of the historical hazard point data and the safety hazard point data is as follows: the historical hazard point data is spliced ​​together according to the composition method of the safety hazard point data (the composition of the safety hazard point data includes: the initial safety hazard point evolution diagram sequence and the initial hazard signal sequence) in chronological order. Therefore, the historical-real-time hazard point integrated data is the data obtained after integrating the historical hazard point data and the safety hazard point data. Furthermore, the process of analyzing the integrated historical and real-time hazard point data to obtain the historical and real-time hazard point analysis results is similar to the process of analyzing the safety hazard point data to obtain the hazard point analysis results, and can achieve the same effect. On the basis that the hazard point analysis results are not the preset hazard results, the historical hazard point data is combined to analyze the hazard points represented by the hazard classification labels corresponding to the safety hazard point data.

[0161] Understandably, if the historical-real-time hazard point analysis results are not the preset historical-real-time hazard results, it means that even if historical hazard point data is combined, and only the hazard points represented by the hazard classification identifiers corresponding to the safety hazard point data are analyzed, no hazard exists. Therefore, in this embodiment of the invention, the hazard point analysis results of the dataset of identified safety hazard points under the monitoring time series are summarized to obtain comprehensive data, and it is determined whether the comprehensive data is safety early warning comprehensive data. The safety early warning comprehensive data is similar to the hazard results and is used to characterize the presence of hazards in the comprehensive data.

[0162] Furthermore, when the comprehensive data is not the preset comprehensive data for safety warnings, historical comprehensive data for a preset historical period is extracted from the historical evolution database. Combined with the historical comprehensive data, all representative hazard points corresponding to multiple hazard classification labels in the dataset of identified safety hazard points are analyzed. Therefore, the historical comprehensive data is similar to the historical hazard point data and can achieve the same effect, which will not be elaborated here.

[0163] To address the problems described in the background art, this invention acquires real-time monitoring data. If the real-time monitoring time falls within a preset tunneling period, the invention utilizes the preset tunneling acquisition period and an on-the-spot acquisition unit to receive tunneling vibration signals. Based on these vibration signals, real-time mine images and a mine monitoring signal set are acquired. The tunneling vibration signals originate from the vibration of the tunneling machine. During tunneling, this invention employs an on-the-spot acquisition unit to detect the geological structure beneath the rock strata, approximating the vibration source as equivalently generated by the tunneling machine, thereby reducing dependence on active vibration sources. If the real-time monitoring time does not fall within the tunneling period, a microseismic monitoring unit is used to acquire real-time mine images and a mine monitoring signal set. The mine monitoring signal set is obtained by the microseismic monitoring unit. This invention uses microseismic monitoring technology to monitor the tunneling roadway outside of the tunneling period to detect the geological structure beneath the rock strata, achieving real-time monitoring of the tunneling roadway. The real-time mine images are aggregated according to a preset monitoring time sequence to obtain a mine image sequence. All safety hazard point markers in the mine image sequence are identified to obtain a safety hazard point marker sequence. This invention categorizes safety hazard point markers for the same hazard point in the tunneling roadway, thereby facilitating the observation of the evolution of the same hazard point. A safety hazard point marker dataset is constructed based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence. In this embodiment, the monitoring frequency is adjusted; the monitoring frequency is increased when the collected data fluctuates significantly, and decreased when the data is relatively stable, thereby accurately monitoring multi-dimensional data of the coal mine roadway while conserving detection energy. The invention parses and identifies a dataset of safety hazard points. If a safety warning instruction is generated based on the parsed dataset, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, it returns to the step of obtaining the real-time monitoring time, thus completing a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental features. This invention, based on the premise that the hazard point parsing result is not a preset hazard result, combines historical hazard point data to analyze the characteristic hazard points corresponding to the hazard classification labels in the safety hazard point data. It adopts a progressively dimensional data analysis method and also combines historical comprehensive data to analyze all characteristic hazard points corresponding to multiple hazard classification labels in the identified safety hazard point dataset. This achieves both grasping point-to-point data hazard fluctuations and analyzing the environmental characteristics of the current coal mine roadway based on big data and historical data, thereby monitoring the morphology and other characteristics of the tunneling roadway and providing data support for determining whether a geological disaster has occurred. Therefore, this invention can utilize on-the-spot acquisition signals and microseismic monitoring technology to monitor the environmental characteristics of the tunneling roadway in real time, and use big data to detect hazard in the environmental characteristics, thereby achieving real-time monitoring and warning of underground mine safety hazards.

[0164] like Figure 2The diagram shown is a functional block diagram of a real-time monitoring and early warning system for geological disasters based on the fusion of big data and environmental features, provided in an embodiment of the present invention.

[0165] The geological disaster real-time monitoring and early warning system 100 based on the fusion of big data and environmental features described in this invention can be installed in an electronic device. Depending on the functions implemented, the geological disaster real-time monitoring and early warning system 100 may include a coal mine environmental monitoring module 101, an in-drill data acquisition module 102, a static data acquisition module 103, and a monitoring feedback module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0166] The coal mine detection environment module 101 is used to receive real-time mine environmental monitoring instructions and confirm the coal mine detection environment based on the real-time mine environmental monitoring instructions. The coal mine detection environment includes: a tunneling machine, a tunneling roadway, and a data acquisition unit. The data acquisition unit includes: a tunneling acquisition unit and a shutdown acquisition unit. The tunneling acquisition unit includes multiple tunneling detectors, and the shutdown acquisition unit includes multiple static mine environment detectors and a microseismic monitoring unit.

[0167] The tunneling acquisition module 102 is used to acquire real-time monitoring time. If the real-time monitoring time is within a preset tunneling period, the tunneling vibration signal is received using the preset tunneling acquisition period and the tunneling acquisition unit. Based on the tunneling vibration signal, a real-time mine image and a mine monitoring signal set are acquired. The tunneling vibration signal comes from the vibration of the tunneling machine.

[0168] The static acquisition module 103 is used to acquire real-time mine images and mine monitoring signal sets using the microseismic monitoring unit if the real-time monitoring time is not within the tunneling period, wherein the mine monitoring signal set is obtained by the microseismic monitoring unit.

[0169] The monitoring feedback module 104 is used to summarize the real-time mine images according to a preset monitoring time sequence to obtain a mine image sequence, identify all safety hazard point markers in the mine image sequence to obtain a safety hazard point marker sequence, and construct a safety hazard point marker dataset based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence. The safety hazard point dataset is parsed. If a safety warning instruction is generated based on the parsed marker hazard dataset, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, the process returns to the step of obtaining the real-time monitoring time, thus completing a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental features.

[0170] In detail, the modules in the geological disaster real-time monitoring and early warning system 100 based on big data and environmental feature fusion described in this embodiment of the invention adopt the same approach as described above when in use. Figure 1 The method used here is the same as the real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental characteristics, and can produce the same technical effect. It will not be elaborated here.

[0171] like Figure 3 The diagram shown is a structural schematic of an electronic device for implementing a real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental features, according to an embodiment of the present invention.

[0172] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a geological disaster real-time monitoring and early warning method program based on the fusion of big data and environmental characteristics.

[0173] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental characteristics, but also to temporarily store data that has been output or will be output.

[0174] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a real-time monitoring and early warning method for geological disasters based on the fusion of big data and environmental characteristics), and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0175] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0176] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0177] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management system, thereby enabling functions such as charging management, discharging management, and power consumption management through the power management system. The power supply may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0178] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0179] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0180] The program for real-time monitoring and early warning of geological disasters based on the fusion of big data and environmental features, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0181] The system receives real-time environmental monitoring commands from the mine and confirms the coal mine monitoring environment based on these commands. The coal mine monitoring environment includes: a tunneling machine, tunneling roadways, and a data acquisition unit. The data acquisition unit includes: a tunneling acquisition unit and a shutdown acquisition unit. The tunneling acquisition unit includes multiple tunneling geophones, and the shutdown acquisition unit includes multiple static mine environment geophones and a microseismic monitoring unit.

[0182] The real-time monitoring time is obtained. If the real-time monitoring time is within a preset tunneling period, the tunneling vibration signal is received using the preset tunneling acquisition period and the tunneling acquisition unit. Based on the tunneling vibration signal, a real-time mine image and a mine monitoring signal set are obtained. The tunneling vibration signal comes from the vibration of the tunneling machine.

[0183] If the real-time monitoring time is not during the tunneling period, the real-time mine image and mine monitoring signal set are obtained by the microseismic monitoring unit, wherein the mine monitoring signal set is obtained by the microseismic monitoring unit.

[0184] The real-time mine images are summarized according to the preset monitoring time sequence to obtain a mine image sequence. All safety hazard point markers in the mine image sequence are identified to obtain a safety hazard point marker sequence. Based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence, a safety hazard point marker dataset is constructed. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence.

[0185] The system parses and identifies a dataset of potential safety hazards. If a safety warning instruction is generated based on the parsed dataset, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, the system returns to the step of obtaining the real-time monitoring time, thus completing a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental characteristics.

[0186] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0187] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or system capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0188] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0189] The system receives real-time environmental monitoring commands from the mine and confirms the coal mine monitoring environment based on these commands. The coal mine monitoring environment includes: a tunneling machine, tunneling roadways, and a data acquisition unit. The data acquisition unit includes: a tunneling acquisition unit and a shutdown acquisition unit. The tunneling acquisition unit includes multiple tunneling geophones, and the shutdown acquisition unit includes multiple static mine environment geophones and a microseismic monitoring unit.

[0190] The real-time monitoring time is obtained. If the real-time monitoring time is within a preset tunneling period, the tunneling vibration signal is received using the preset tunneling acquisition period and the tunneling acquisition unit. Based on the tunneling vibration signal, a real-time mine image and a mine monitoring signal set are obtained. The tunneling vibration signal comes from the vibration of the tunneling machine.

[0191] If the real-time monitoring time is not during the tunneling period, the real-time mine image and mine monitoring signal set are obtained by the microseismic monitoring unit, wherein the mine monitoring signal set is obtained by the microseismic monitoring unit.

[0192] The real-time mine images are summarized according to the preset monitoring time sequence to obtain a mine image sequence. All safety hazard point markers in the mine image sequence are identified to obtain a safety hazard point marker sequence. Based on the mine monitoring signal set, the mine image sequence, and the safety hazard point marker sequence, a safety hazard point marker dataset is constructed. The safety hazard point data includes a marker hazard evolution diagram sequence and a marker hazard evolution signal sequence.

[0193] The system parses and identifies a dataset of potential safety hazards. If a safety warning instruction is generated based on the parsed dataset, a safety warning instruction is sent to the initiator of the real-time environmental monitoring instruction in the mine. Otherwise, the system returns to the step of obtaining the real-time monitoring time, thus completing a real-time monitoring and warning system for geological disasters based on the fusion of big data and environmental characteristics.

[0194] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0195] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0196] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0197] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0198] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A real-time monitoring and early warning method for geological disasters based on big data and environmental feature fusion, characterized in that, The method comprises: Receiving a coal mine real-time environment monitoring instruction, and confirming a coal mine detection environment based on the coal mine real-time environment monitoring instruction, wherein the coal mine detection environment comprises a heading machine, a heading tunnel, and a data acquisition unit, the data acquisition unit comprises a heading-accompanying acquisition unit and a shutdown acquisition unit, the heading-accompanying acquisition unit comprises a plurality of heading-accompanying detectors, and the shutdown acquisition unit comprises a plurality of static coal mine environment detectors and a microseismic monitoring unit; Obtaining a real-time monitoring time, and if the real-time monitoring time is in a preset heading period, receiving a heading vibration signal by using a preset heading acquisition period and the heading-accompanying acquisition unit, obtaining a real-time coal mine image and a coal mine monitoring signal set based on the heading vibration signal, wherein the heading vibration signal is from vibration of the heading machine; The obtaining of the real-time coal mine image and the coal mine monitoring signal set based on the heading vibration signal comprises: Fitting the heading vibration signal by using a pre-constructed seismic interference technology to obtain a virtual shot signal set, performing a low-pass filtering operation on the virtual shot signal set to obtain a noise-reduced signal, analyzing the noise-reduced signal by using a pre-constructed scattered wave imaging method, and visualizing the analyzed noise-reduced signal to obtain a coal mine hidden danger detection image corresponding to the real-time monitoring time; Analyzing the coal mine hidden danger detection image to obtain a real-time coal mine image containing a safety hidden danger point identifier, and confirming the heading vibration signal as the coal mine monitoring signal set; If the real-time monitoring time is not in the heading period, obtaining a real-time coal mine image and a coal mine monitoring signal set by using the microseismic monitoring unit, wherein the coal mine monitoring signal set is monitored by the microseismic monitoring unit; According to a preset monitoring time sequence, the real-time coal mine images are summarized to obtain a coal mine image sequence, all safety hidden danger point identifiers in the coal mine image sequence are identified to obtain a safety hidden danger point identifier sequence, and an identifier safety hidden danger point data set is constructed based on the coal mine monitoring signal set, the coal mine image sequence, and the safety hidden danger point identifier sequence, wherein the safety hidden danger point data comprises an identifier hidden danger evolution graph sequence and an identifier hidden danger evolution signal sequence; The identifier safety hidden danger point data set is analyzed, if a safety warning instruction is generated based on the analyzed identifier hidden danger data set, the safety warning instruction is sent to an initiation end of the coal mine real-time environment monitoring instruction, otherwise, the step of obtaining the real-time monitoring time is returned, and a geological disaster real-time monitoring and warning based on big data and environment feature fusion is completed. 2.The real-time monitoring and early warning method for geological disasters based on big data and environmental feature fusion according to claim 1, characterized in that, The obtaining of the real-time coal mine image and the coal mine monitoring signal set by using the microseismic monitoring unit comprises: Judging whether there is a same coal mine static monitoring time as the real-time monitoring time in a pre-constructed coal mine static monitoring time sequence; If there is a same coal mine static monitoring time as the real-time monitoring time in the coal mine static monitoring time sequence, a microseismic monitoring signal is received by using a plurality of static coal mine environment detectors, the microseismic monitoring signal is analyzed, a real-time coal mine image is obtained based on the microseismic monitoring signal, and the microseismic monitoring signal is confirmed as the coal mine monitoring signal set. 3.The method of claim 2, wherein, The judgment of whether there is a same coal mine static monitoring time as the real-time monitoring time in the pre-constructed coal mine static monitoring time sequence comprises: The method comprises the following steps: acquiring a neighboring historical hidden danger monitoring time period based on a real-time monitoring time point; extracting a latest historical hidden danger monitoring time point prior to the real-time monitoring time point from the neighboring historical hidden danger monitoring time period; calculating a hidden danger monitoring frequency control parameter based on the latest historical hidden danger monitoring time point and a pre-constructed hidden danger monitoring frequency control parameter calculation formula; acquiring a target monitoring time point by using the hidden danger monitoring frequency control parameter; and importing the target monitoring time point into a pre-constructed initial mine static monitoring time sequence to obtain a mine static monitoring time sequence, wherein the mine static monitoring time sequence comprises a plurality of mine static monitoring time points. The target monitoring time point is compared with the real-time monitoring time point. If the target monitoring time point lags behind the real-time monitoring time point, it is determined that there is no mine static monitoring time point identical to the real-time monitoring time point in the mine static monitoring time sequence; otherwise, it is determined that there is a mine static monitoring time point identical to the real-time monitoring time point in the mine static monitoring time sequence. 4.The method of claim 3, wherein, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: 5.The method of claim 4, wherein, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method Otherwise, historical hazard point data of a preset historical period is extracted from the pre-constructed historical evolution database based on the safety hazard point data, the historical hazard point data and the safety hazard point data are integrated to obtain historical-real-time hazard point integrated data, the historical-real-time hazard point integrated data is analyzed to obtain a historical-real-time hazard point analysis result; If the historical-real-time hazard point analysis result is a preset historical-real-time hazard result, a safety warning instruction is sent to the initiator of the mine real-time environment monitoring instruction; Otherwise, the hazard point analysis results are summarized to obtain comprehensive data corresponding to the identified safety hazard point data set under the monitoring time sequence, and if the comprehensive data is a preset safety warning comprehensive data, a safety warning instruction is sent to the initiator of the mine real-time environment monitoring instruction; If the comprehensive data is not the preset safety warning comprehensive data, historical comprehensive data of a preset historical period is extracted from the historical evolution database, and if the historical comprehensive data is not the preset historical safety warning comprehensive data, the step of obtaining the real-time monitoring time is returned. 6.The method of claim 5, wherein the method further comprises: The hazard monitoring frequency control parameter is calculated based on the latest historical hazard monitoring time and a pre-constructed hazard monitoring frequency control parameter calculation formula, and the target monitoring time is obtained by using the hazard monitoring frequency control parameter, including: Based on the adjacent historical hazard monitoring period, an adjacent hazard point data sequence is obtained, wherein the adjacent hazard point data sequence includes a plurality of adjacent hazard point data, and the plurality of adjacent hazard point data are sorted in order from early to late, and each adjacent hazard point data in the plurality of adjacent hazard point data includes hazard point position characteristics, hazard point displacement characteristics, hazard point deformation characteristics, and hazard point frequency domain characteristics; A target data sequence is extracted from the adjacent hazard point data sequence, and the target data sequence is numerically processed to obtain a target numerical sequence, wherein the target data sequence includes a plurality of target data of the same type, and the type of the target data is one of hazard point position characteristics, hazard point displacement characteristics, hazard point deformation characteristics, and hazard point frequency domain characteristics; The target numerical value is sequentially extracted from the target numerical sequence, and the following operations are performed on the extracted target numerical value: The next target numerical value lagging behind the target numerical value is extracted from the target numerical sequence to obtain a neighboring numerical value, the difference between the target numerical value and the neighboring numerical value is calculated to obtain a first-order difference value, and the first-order difference value is summarized to obtain a first-order difference sequence; The preset sliding window and the preset sliding step are used to slide and intercept the first-order difference sequence to obtain a sliding difference data sequence, the sliding difference data is sequentially extracted from the sliding difference data sequence, and the variance of the extracted sliding difference data is calculated to obtain a data change corresponding to the sliding difference data; The data change is summarized to obtain a data change sequence; The first adjustment coefficient and the second adjustment coefficient are confirmed based on the data change sequence and the first-order difference sequence, and the hazard monitoring frequency control parameter calculation formula is constructed according to the first adjustment coefficient and the second adjustment coefficient; The hazard monitoring frequency control parameter is calculated by using the hazard monitoring frequency control parameter calculation formula and the latest historical hazard monitoring time. The target monitoring time is obtained based on a pre-constructed variable frequency adjustment sequence and a hidden danger monitoring frequency control parameter. 7.The method of claim 6, wherein the method further comprises: The target monitoring time is obtained based on a pre-constructed variable frequency adjustment sequence and a hidden danger monitoring frequency control parameter. The multi-source monitoring data is obtained, reference mine roadway sets are extracted from the multi-source monitoring data, and reference mine roadways are sequentially extracted from the reference mine roadway sets. A reference frequency parameter set corresponding to the reference mine roadway is obtained, and a reference regulation factor set corresponding to the reference frequency parameter set is calculated based on the hidden danger monitoring frequency control parameter calculation formula, wherein the reference frequency parameter and the reference regulation factor are one-to-one corresponding. A mapping relationship between the reference frequency parameter set and the reference regulation factor set is constructed to obtain an initial frequency-factor data set, and the initial frequency-factor data set is simplified to obtain a frequency-factor set, wherein the frequency-factor set includes a plurality of frequency-factors, and each frequency-factor includes a reference frequency parameter and a reference regulation factor. The frequency-factor set group corresponding to the reference mine roadway set is obtained by aggregating the frequency-factor set. The plurality of frequency-factor clusters are obtained by performing a reference frequency-based clustering operation on the frequency-factor set group using a pre-constructed clustering method and a pre-set number of clustering centers, wherein the reference frequency-based clustering operation is a K-means clustering operation with reference frequency as a variable, and the number of the plurality of frequency-factor clusters is equal to the number of the clustering centers. The frequency-factor clusters are sequentially extracted from the plurality of frequency-factor clusters, the clustering frequency interval and the clustering factor interval of the extracted frequency-factor cluster are identified, the mean value of the clustering frequency interval is calculated to obtain a fuzzy reference frequency. The unit variable frequency adjustment interval is obtained by associating the clustering factor interval with the fuzzy reference frequency, the unit variable frequency adjustment intervals are aggregated and integrated to obtain the variable frequency adjustment sequence. The target interval is obtained by searching for the unit variable frequency adjustment interval in which the hidden danger monitoring frequency control parameter is located in the variable frequency adjustment sequence, the fuzzy reference frequency corresponding to the target interval is confirmed as the target frequency, and the target monitoring time is calculated based on the target frequency and the latest historical hidden danger monitoring time.

8. A geological disaster real-time monitoring and early warning system based on big data and environmental feature fusion, applied to the geological disaster real-time monitoring and early warning method based on big data and environmental feature fusion as claimed in claim 1, characterized in that, The system comprises: The coal mine detection environment module is configured to receive a mine real-time environment monitoring instruction and confirm a coal mine detection environment based on the mine real-time environment monitoring instruction, wherein the coal mine detection environment comprises a heading machine, a heading roadway, and a data acquisition unit, the data acquisition unit comprises a heading acquisition unit and a shutdown acquisition unit, the heading acquisition unit comprises a plurality of heading detectors, and the shutdown acquisition unit comprises a plurality of static mine environment detectors and a microseismic monitoring unit. The tunneling collection module is configured to obtain a real-time monitoring time. If the real-time monitoring time is in a preset tunneling period, a preset tunneling collection period and a tunneling collection unit are used to receive a tunneling vibration signal. Based on the tunneling vibration signal, a real-time mine image and a mine monitoring signal set are obtained. The tunneling vibration signal is from the vibration of the tunneling machine. The real-time mine image and the mine monitoring signal set based on the tunneling vibration signal include the following steps. A pre-constructed seismic interference technology is used to fit the tunneling vibration signal to obtain a virtual shot signal set. A low-pass filtering operation is performed on the virtual shot signal set to obtain a noise reduction signal. A pre-constructed scattered wave imaging method is used to analyze the noise reduction signal, and the analyzed noise reduction signal is visualized to obtain a mine hazard detection image corresponding to the real-time monitoring time. The mine hazard detection image is analyzed to obtain a real-time mine image containing a safety hazard point identifier. The tunneling vibration signal is confirmed as the mine monitoring signal set. The static collection module is configured to obtain a real-time mine image and a mine monitoring signal set using a microseismic monitoring unit if the real-time monitoring time is not in the tunneling period. The mine monitoring signal set is obtained by the microseismic monitoring unit. The monitoring feedback module is configured to aggregate the real-time mine image according to a preset monitoring time sequence to obtain a mine image sequence. All safety hazard point identifiers in the mine image sequence are identified to obtain a safety hazard point identifier sequence. Based on the mine monitoring signal set, the mine image sequence, and the safety hazard point identifier sequence, an identifier safety hazard point data set is constructed. The safety hazard point data includes an identifier hazard evolution graph sequence and an identifier hazard evolution signal sequence. The identifier safety hazard point data set is analyzed. If a safety warning instruction is generated based on the analyzed identifier hazard data set, the safety warning instruction is sent to the initiator of the mine real-time environment monitoring instruction. Otherwise, the step of obtaining the real-time monitoring time is returned. A geological disaster real-time monitoring and warning based on big data and environmental feature fusion is completed.

Images