A data processing method and system for abnormal monitoring in coal mines

Abstract

The present invention relates to the technical field of data processing, and provides a data processing method and system for abnormal monitoring in a coal mine. The method comprises: obtaining coal mine spatial characteristics and setting a multi-level monitoring scheme, obtaining gas concentration data, and if the gas concentration threshold is met, identifying the first coal mine space as a quasi-gas abnormal area; particle density evaluation generates particle concentration, and if the concentration threshold is met, identifying the first coal mine space as a gas abnormal area; obtaining coal mine activity characteristics for abnormal analysis, generating gas abnormality factors, and combining gas concentration data to construct a multi-level early warning scheme to solve the technical problem that the monitoring data in the coal mine lacks effective verification means and the obtained monitoring data cannot be applied to actual abnormal early warning, realizes reliability analysis of gas data, effectively verifies the monitoring data in the coal mine, improves the accuracy and effectiveness of the monitoring data, and applies it to actual detection and early warning, thereby ensuring the technical effect of the reliability of abnormal monitoring and early warning in the coal mine.

Description

Technical Field

[0001] The present invention relates to the technical field related to data processing, and in particular to a data processing method and system for abnormal monitoring in a coal mine. Background Art

[0002] Abnormal monitoring data processing in coal mines refers to the processing and analysis of internal data collected from various monitoring equipment in coal mines in order to better understand the operating status of coal mines, predict possible abnormal situations, and take timely measures to avoid accidents and ensure coal mine production safety.

[0003] Coal mine accidents have occurred frequently in recent years. Data show that due to the lack of effective verification of monitoring data and the fact that monitoring data are often uploaded directly, the data has less content and is less referenceable. The accuracy and reliability of the monitoring data cannot be guaranteed, and the quality of the monitoring data is poor.

[0004] In summary, the existing technology has the technical problem that there is a lack of effective verification means for monitoring data in coal mines, and the obtained monitoring data cannot be applied to actual abnormal warning. Summary of the invention

[0005] The present application provides a data processing method and system for abnormal monitoring in coal mines, aiming to solve the technical problem that the monitoring data in coal mines in the prior art lacks effective verification means and the obtained monitoring data cannot be applied to actual abnormal warning.

[0006] In view of the above problems, the present application provides a data processing method and system for abnormal monitoring in a coal mine.

[0007] The first aspect disclosed in the present application provides a data processing method for abnormal monitoring in a coal mine, wherein the method comprises: obtaining spatial characteristics of the coal mine, wherein the spatial characteristics of the coal mine include spatial volume coefficient and heat source distribution information; setting a multi-level monitoring scheme according to the spatial volume coefficient and the heat source distribution information; obtaining gas concentration data of a first coal mine space according to the multi-level monitoring scheme; when the gas concentration data meets the gas concentration threshold of the first coal mine space, identifying the first coal mine space as a quasi-gas abnormal area; performing particle density evaluation on the gas concentration data to generate particle concentration; when the particle concentration meets the concentration threshold, identifying the first coal mine space as a gas abnormal area; obtaining coal mine activity characteristics for abnormal analysis to generate a gas abnormality factor; and constructing a multi-level early warning scheme according to the gas abnormality factor and the gas concentration data.

[0008] Another aspect disclosed in the present application provides a data processing system for monitoring abnormalities in a coal mine, wherein the system includes: a feature acquisition module for acquiring spatial features of a coal mine, wherein the spatial features of the coal mine include spatial volume coefficients and heat source distribution information; a monitoring scheme setting module for setting a multi-level monitoring scheme based on the spatial volume coefficients and the heat source distribution information; a concentration data acquisition module for acquiring gas concentration data of a first coal mine space according to the multi-level monitoring scheme; a first judgment module for identifying the first coal mine space as a quasi-gas abnormal area when the gas concentration data meets a gas concentration threshold of the first coal mine space; a concentration generation module for performing particle density evaluation on the gas concentration data to generate a particle concentration; a second judgment module for identifying the first coal mine space as a gas abnormal area when the particle concentration meets a concentration threshold; an abnormal factor generation module for acquiring coal mine activity features for abnormal analysis to generate a gas abnormal factor; and an early warning scheme construction module for constructing a multi-level early warning scheme based on the gas abnormal factor and the gas concentration data.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] By acquiring the spatial characteristics of coal mines and setting a multi-level monitoring plan, gas concentration data is obtained; when the gas concentration data meets the gas concentration threshold, the first coal mine space is identified as a quasi-gas anomaly area; the particle density of the gas concentration data is evaluated to generate particle concentration; when the particle concentration meets the concentration threshold, the first coal mine space is identified as a gas anomaly area; the coal mine activity characteristics are obtained for anomaly analysis, and a gas anomaly factor is generated. Combined with the gas concentration data, a multi-level early warning plan is constructed, which realizes the reliability analysis of gas data, effectively verifies the monitoring data in the coal mine, improves the accuracy and effectiveness of the monitoring data, and applies it to actual detection and early warning, ensuring the technical effect of the reliability of abnormal monitoring and early warning in coal mines.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A possible flow chart of a data processing method for abnormal monitoring in a coal mine is provided for the embodiment of the present application;

[0013] Figure 2 A possible flow chart of setting a multi-level monitoring scheme in a data processing method for abnormal monitoring in a coal mine is provided for an embodiment of the present application;

[0014] Figure 3 A possible flow chart of determining gas anomaly factors in a data processing method for abnormal monitoring in a coal mine is provided for an embodiment of the present application;

[0015] Figure 4 A possible structural schematic diagram of a data processing system for abnormality monitoring in a coal mine is provided for the embodiment of the present application.

[0016] Explanation of the accompanying drawings: feature acquisition module 100, monitoring scheme setting module 200, concentration data acquisition module 300, first judgment module 400, concentration generation module 500, second judgment module 600, abnormal factor generation module 700, early warning scheme construction module 800. DETAILED DESCRIPTION

[0017] The embodiment of the present application provides a data processing method and system for abnormal monitoring in coal mines, which solves the technical problem that there is a lack of effective verification means for monitoring data in coal mines and the obtained monitoring data cannot be applied to actual abnormal warnings. It realizes the reliability analysis of gas data, effectively verifies the monitoring data in coal mines, improves the accuracy and effectiveness of monitoring data, and applies it to actual detection and warning, thereby ensuring the technical effect of the reliability of abnormal monitoring and early warning in coal mines.

[0018] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0019] Embodiment 1:

[0020] like Figure 1 As shown, an embodiment of the present application provides a data processing method for abnormal monitoring in a coal mine, wherein the method comprises:

[0021] S10: Acquire spatial characteristics of the coal mine, wherein the spatial characteristics of the coal mine include spatial volume coefficient and heat source distribution information;

[0022] S20: setting a multi-level monitoring scheme according to the space volume coefficient and the heat source distribution information;

[0023] like Figure 2 As shown, step S20 includes the steps of:

[0024] S21: setting a first monitoring frequency according to the space volume coefficient;

[0025] S22: adjusting the first monitoring frequency by partition according to the heat source distribution information to obtain a second monitoring frequency;

[0026] S23: adjusting the gas concentration limit according to the heat source distribution information to generate the gas concentration threshold;

[0027] S24: Setting the multi-level monitoring scheme according to the second monitoring frequency and the gas concentration threshold.

[0028] Specifically, the data quality is improved by analyzing the reliability of gas data and enriching the data content according to the logic of processing data quality. Based on this, the spatial characteristics of the coal mine include spatial volume coefficient and heat source distribution information. The spatial volume and heat source distribution are used as search contents, and search symbols are set. Data is searched in the data storage unit of the data processing system for abnormal monitoring in the coal mine to obtain the spatial characteristics of the coal mine. The spatial volume coefficient can be a first-level (volume size is (0, 5], unit: m 3 ), Level 2 (volume size is (5, 10], unit: m 3 ), level 3 (volume size is (10, 15], unit: m 3 ) and other levels and sizes, the heat source distribution information includes the heat source location point, heat source calorific value and other related indicator characteristics;

[0029] According to the space volume coefficient and the heat source distribution information, a multi-level monitoring scheme is set, including that the first monitoring frequency can be set to 3 minutes / time. According to the space volume coefficient, the first monitoring frequency is set, for example: if the space volume is less than 5 cubic meters, the monitoring frequency is 3 minutes / time; if the space volume is 5-10 cubic meters, the monitoring frequency is 5 minutes / time; there are a large number of combustible gases and coal dust in coal mines, and high temperature can easily cause safety accidents such as explosions and fires. A relatively low temperature should be maintained, and ventilation and fire prevention management should be strengthened to reduce the risk of accidents. Based on this, the first monitoring frequency is partitioned and adjusted according to the heat source distribution information to obtain the second monitoring frequency. For example, a radius range is defined, for example, 20 meters: 20 meters around each heat source is defined, and the monitoring frequency within 20 meters is originally 10 minutes / time. At this time, according to the temperature information of the heat source, the higher the temperature, the higher the monitoring frequency within the adjusted 20-meter range;

[0030] Generally speaking, the gas concentration threshold standard specified in GB 41022-2021 "Basic Indicators for Coal Mine Gas Extraction" and other relevant standards is 1.5%. When the gas concentration exceeds the threshold, it is easy to cause a gas explosion accident. Verification shows that the relationship between gas concentration and ignition temperature is: the ignition temperature corresponding to a gas concentration of 2% is 810°C, the ignition temperature corresponding to a gas concentration of 3.4% is 665°C, and the ignition temperature corresponding to a gas concentration of 6.5% is 512°C. Based on this, the gas concentration limit is adjusted according to the heat source distribution information. In areas with higher heat source temperatures, the lower the gas concentration threshold, and the corresponding gas concentration threshold is generated; the gas concentration threshold is used as the detection content, the detection task is performed at the second monitoring frequency, and the multi-level monitoring scheme is set to provide data support for subsequent analysis.

[0031] S30: acquiring gas concentration data of the first coal mine space according to the multi-level monitoring scheme;

[0032] S40: When the gas concentration data meets the gas concentration threshold of the first coal mine space, marking the first coal mine space as a quasi-gas abnormality area;

[0033] Specifically, the multi-level monitoring scheme is executed, data retrieval is performed in the first coal mine space, and gas concentration data of the first coal mine space is obtained, wherein the gas concentration data also includes time series information corresponding to the collected gas concentration; the gas concentration data is compared to meet the gas concentration threshold of the first coal mine space. When the gas concentration data meets the gas concentration threshold of the first coal mine space, the quasi-gas abnormality area is an area with greater safety hazards, and the first coal mine space is identified as a quasi-gas abnormality area, providing a basis for subsequent analysis.

[0034] S50: performing particle density evaluation on the gas concentration data to generate particle concentration;

[0035] Step S50 includes the steps of:

[0036] S51: Acquire first sensor distribution information of the gas concentration data;

[0037] S52: collecting reference gas concentration data of adjacent sensors according to the first sensor distribution information, wherein the distribution distance between the adjacent sensors and the first sensor is less than or equal to an adjacent distance threshold;

[0038] S53: Perform particle density evaluation on the gas concentration data according to the reference gas concentration data to obtain the particle concentration.

[0039] Specifically, if the sensor for collecting gas concentration data is abnormal, it may cause the gas concentration data to be abnormal. Based on this, it is possible to use the method of mutual verification of adjacent sensors to determine whether this situation exists, perform particle density evaluation on the gas concentration data, and generate particle concentration, including setting a neighboring distance threshold, the neighboring distance threshold generally does not exceed 1m, preferably, in the embodiment of the present application, the neighboring distance threshold can be set to 0.7m, randomly determine a sensor in the first coal mine space and record it as the first sensor, take the first sensor as the center, take the neighboring distance threshold as the radius, and obtain the first sensor distribution information of the gas concentration data, the first sensor distribution information includes all sensor spatial position information in the space with the first sensor as the center and the neighboring distance threshold as the radius; the distribution distance between the adjacent sensor and the first sensor is less than or equal to the neighboring distance threshold, and according to the first sensor distribution information, collect reference gas concentration data of the adjacent sensor, at the same time point, the reference gas concentration data corresponds one-to-one with the first sensor distribution information; according to the reference gas concentration data, perform particle density evaluation on the gas concentration data, obtain the particle concentration, and use the method of mutual verification of adjacent sensors to verify the possibility of abnormality of the sensor itself, so as to improve data quality.

[0040] Step S53 includes the steps of:

[0041] S531: constructing a first coordinate axis based on the distribution distance, and constructing a second coordinate axis based on the gas concentration deviation;

[0042] S532: constructing a particle density evaluation coordinate system based on the first coordinate axis and the second coordinate axis;

[0043] S533: determining distribution coordinate information of the proximity sensor in the particle density evaluation coordinate system according to the reference gas concentration data and the gas concentration data;

[0044] S534: constructing a k-distance core neighborhood, wherein the k-distance core neighborhood includes k distribution coordinate information from near to far;

[0045] S535: Calculate multiple reachable distances based on the k-distance core neighborhood and the distribution coordinate information;

[0046] S536: Perform mean analysis on the multiple reachable distances and calculate the inverse to generate the particle concentration.

[0047] Specifically, the air circulation speed in the mine is limited, and the deviation of the gas concentration data monitored by the adjacent sensors meets certain limitations. Based on this, the particle density of the gas concentration data is evaluated according to the reference gas concentration data to obtain the particle concentration, including taking the position where the distribution distance and the concentration deviation of the first sensor are both 0 as the origin coordinate of the coordinate axis, taking the distribution distance as the first coordinate axis, and taking the gas concentration deviation as the second coordinate axis, and constructing a particle density evaluation coordinate system based on the first coordinate axis, the second coordinate axis and the origin coordinate. It should be noted that the particle density evaluation coordinate system only has the area of ​​the first quadrant and the area of ​​the second quadrant. The area of ​​the four quadrants; the gas concentration data is taken as 0, and the concentration deviation between the reference gas concentration data and the gas concentration data is taken as the coordinate value of the second coordinate axis, and the distribution coordinate information of the adjacent sensor in the particle density evaluation coordinate system is determined; k is used to characterize the number of adjacent sensors, such as k=3, and the coordinates corresponding to the third sensor from the origin coordinate from near to far are taken as the radius, and a fan is made in the particle density evaluation coordinate system (the fan-shaped area of ​​k=3 includes the fan-shaped area of ​​k=1 and the fan-shaped area of ​​k=2), and a k-distance core neighborhood is constructed, wherein k≥2 and k is a positive integer, and the k-distance core neighborhood includes k distribution coordinate information from near to far;

[0048] Based on the k-distance core neighborhood and the distribution coordinate information, multiple reachable distances are calculated, wherein the reachable distance of any coordinate within the k-distance core neighborhood is equal to the farthest distance between the k coordinates and the origin; the reachable distance of any coordinate not within the k-distance core neighborhood is equal to its actual distance from the origin; the mean of the multiple reachable distances is calculated, the inverse of the result obtained by the mean analysis is calculated, and the result obtained by the inverse is recorded as the particle concentration, so as to provide support for mutual verification of adjacent sensors.

[0049] S60: When the particle concentration meets the concentration threshold, marking the first coal mine space as a gas abnormality area;

[0050] S70: Obtain the coal mine activity characteristics for abnormal analysis and generate gas abnormality factors;

[0051] like Figure 3 As shown, step S70 includes the steps of:

[0052] S71: The coal mine activity characteristics include activity type characteristics, activity location characteristics and activity duration characteristics, wherein the coal mine activity characteristics belong to a preset time zone;

[0053] S72: Inputting the activity type feature, the activity location feature and the activity duration feature into a gas emission analysis model to obtain gas emission prediction data;

[0054] S73: When the gas emission prediction data is greater than or equal to the gas emission threshold, the coal mine activity characteristics are added to the gas anomaly factor.

[0055] Specifically, a concentration threshold is set, and the concentration threshold is generally not less than 90%. If the particle concentration is greater than or equal to the concentration threshold, it indicates that the adjacent sensors have verified each other, that is, the first sensor itself has no abnormality; if the particle concentration is less than the concentration threshold, it indicates that the adjacent sensors have failed to verify each other, that is, the first sensor itself has an abnormality; when the particle concentration meets the concentration threshold, that is, the particle concentration is less than the concentration threshold, the first coal mine space is identified as a gas abnormality area; the gas abnormality factor includes the activity type and activity location that cause the gas monitoring data to be abnormal, obtains the coal mine activity characteristics for abnormality analysis, generates the gas abnormality factor, and improves the richness of the gas abnormality warning information;

[0056] Set the gas emission threshold: For mines with an annual output of 0.4 Mt or less, the gas emission threshold can be set to 15 m 2 / min; For mines with an annual output of 0.4-0.6 Mt or less, the gas emission threshold can be set to 20m 2 / min; For mines with an annual output of 0.6~1.0Mt, the gas emission threshold can be set to 25m 2 / min; For mines with an annual output of 1.0-1.5 Mt, the gas emission threshold can be set to 30m 2 / min;

[0057] The coal mine activities include but are not limited to ventilation and production (blasting or coal cutting). The coal mine activity characteristics belong to a preset time zone. The coal mine activity characteristics include activity type characteristics, activity location characteristics and activity duration characteristics. The activity location characteristics are used to characterize the unique activity location characteristics of the excavation working face, air inlet tunnel, air outlet tunnel or other coal mines; a gas emission volume analysis model is constructed, and the activity type characteristics, activity location characteristics and activity duration characteristics are used as input data, input into the gas emission volume analysis model, and output to obtain gas emission volume prediction data; the gas emission volume prediction data and the gas emission volume threshold value are both absolute gas emission volumes, and the absolute gas emission volume is used to characterize the amount of gas (in units of m2) emitted from the coal seam and the mined coal (rock) per unit time. 3 / min), set a gas emission threshold, compare the gas emission prediction data with the gas emission threshold, and when the gas emission prediction data is greater than or equal to the gas emission threshold, add the coal mine activity characteristics into the gas anomaly factor to provide support for timely safety warning of coal mine activities.

[0058] Step S72 also includes the steps of:

[0059] S721: Obtain gas management record parameters, wherein the gas management record parameters include activity type records, activity location records, activity duration records, and gas outflow volume records;

[0060] S722: training a first base processor based on a BP neural network according to the activity type record, the activity location record, the activity duration record, and the gas outflow volume record;

[0061] S723: Acquire the activity type record, the activity location record, the activity duration record, and the gas outflow volume record whose output accuracy of the first base processor does not meet the preset accuracy, and set them as a first error data set;

[0062] S724: when the first error data set is larger than the converged data amount, increase the weight of the first error data set in the original data set, and train the second base processor based on the BP neural network;

[0063] S725: Repeat the training. When the Nth error data set is less than or equal to the converged data volume, fuse the first base processor, the second base processor, and finally the Nth base processor to obtain the gas emission analysis model.

[0064] Specifically, the activity type feature, the activity location feature and the activity duration feature are input into the gas emission analysis model to obtain the gas emission prediction data, and before that, the method further includes extracting empirical data in the data processing system for abnormal monitoring in the coal mine, obtaining the activity type record, activity location record, activity duration record and gas emission record in the past year, and using the activity type record, activity location record, activity duration record and gas emission record in the past year as gas management record parameters;

[0065] Taking the BP neural network as the model basis, using the activity type record, the activity location record, the activity duration record and the gas outflow volume record as the construction data, constructing a new combination feature based on the time series information in the gas management record parameter, and transmitting it into the BP neural network for model convergence learning, and training the first base processor; in the process of training the first base processor, synchronously calculating the output accuracy rate, and obtaining the output accuracy rate of the first base processor; obtaining the activity type record, the activity location record, the activity duration record and the gas outflow volume record whose output accuracy rate of the first base processor does not meet the preset accuracy rate; setting the activity type record, the activity location record, the activity duration record and the gas outflow volume record whose output accuracy rate of the first base processor does not meet the preset accuracy rate as the first error data set;

[0066] The converged data volume, i.e., the data volume of the activity type record, the activity location record, the activity duration record and the gas outflow volume record whose output accuracy of the first base processor meets the preset accuracy, calculates the data volume of the first error data set, compares the data volume of the first error data set with the converged data volume, and when the data volume of the first error data set is greater than the converged data volume, gains the weight of the first error data set in the original data set, and trains the second base processor based on the BP neural network;

[0067] Fuse N (N is a positive integer greater than or equal to 3) base processors: average the node weights of multiple base processors, the model nodes with fewer error data sets have larger fusion weights, and the model nodes with more error data sets have smaller fusion weights; repeat the training, and when the Nth error data set is less than or equal to the converged data volume, fuse the first base processor, the second base processor until the Nth base processor, fuse N base processors, obtain the gas emission analysis model, and provide model support for gas emission analysis.

[0068] S80: Construct a multi-level early warning plan based on the gas abnormality factor and the gas concentration data.

[0069] Step S80 includes the steps of:

[0070] S81: matching a gas elimination plan according to the gas abnormality factor and the gas concentration data, wherein the gas elimination plan includes emission management personnel and emission staff;

[0071] S82: generating a personnel evacuation plan according to the gas abnormality area, wherein the personnel evacuation plan includes evacuation range information;

[0072] S83: Construct the multi-level warning plan according to the emission management personnel, the emission staff and the evacuation range information.

[0073] Specifically, according to the gas abnormality factor and the gas concentration data, a multi-level early warning scheme is constructed, including: the gas elimination scheme includes emission management personnel and emission staff; according to the gas abnormality factor and the gas concentration data, the gas elimination scheme is matched, that is, firstly, the private user end and the public notification end of each emission management personnel and emission staff are notified to eliminate the gas abnormality factor; according to different gas concentrations and gas activity types, different staff and different management personnel are matched. Generally speaking, the greater the gas concentration and the greater the gas outflow of the activity type, the higher the management personnel level;

[0074] The personnel evacuation plan includes evacuation range information. A personnel evacuation plan is generated based on the gas abnormality area, which is to send private messages to users in the evacuation range, and notify them by broadcasting notifications to evacuate to areas with lower gas concentrations or leave the mine; thereby generating multi-level early warning data. The multi-level early warning plan, that is, various levels of combined early warning plans constructed by the emission management personnel, the emission staff and the evacuation range information, conducts abnormal monitoring in the coal mine and is applied to the mine evacuation early warning process to avoid gas explosion disasters and ensure the lives of miners.

[0075] In summary, the data processing method and system for abnormal monitoring in a coal mine provided by the embodiments of the present application have the following technical effects:

[0076] 1. By adopting the method of obtaining the spatial characteristics of coal mines and setting a multi-level monitoring scheme, gas concentration data is obtained; when the gas concentration data meets the gas concentration threshold, the first coal mine space is identified as a quasi-gas abnormal area; the particle density of the gas concentration data is evaluated to generate the particle concentration; when the particle concentration meets the concentration threshold, the first coal mine space is identified as a gas abnormal area; the coal mine activity characteristics are obtained for abnormal analysis to generate gas abnormal factors, and a multi-level early warning scheme is constructed in combination with the gas concentration data. This application provides a data processing method and system for abnormal monitoring in coal mines, realizes the reliability analysis of gas data, effectively verifies the monitoring data in coal mines, improves the accuracy and effectiveness of the monitoring data, and applies it to actual detection and early warning, ensuring the technical effect of the reliability of abnormal monitoring and early warning in coal mines.

[0077] 2. As the coal mine activity characteristics including activity type characteristics, activity location characteristics and activity duration characteristics are adopted and input into the gas emission analysis model, the gas emission prediction data is obtained; when the gas emission prediction data is greater than or equal to the gas emission threshold, the coal mine activity characteristics are added into the gas anomaly factor, and when the gas emission prediction data is greater than or equal to the gas emission threshold, the coal mine activity characteristics are added into the gas anomaly factor, providing support for timely safety warning of coal mine activities.

[0078] Embodiment 2:

[0079] Based on the same inventive concept as the data processing method for abnormal monitoring in a coal mine in the aforementioned embodiment, Figure 4 As shown, an embodiment of the present application provides a data processing system for abnormal monitoring in a coal mine, wherein the system includes:

[0080] The feature acquisition module 100 is used to acquire the spatial features of the coal mine, wherein the spatial features of the coal mine include the spatial volume coefficient and the heat source distribution information;

[0081] A monitoring scheme setting module 200, for setting a multi-level monitoring scheme according to the space volume coefficient and the heat source distribution information;

[0082] The concentration data acquisition module 300 is used to acquire the gas concentration data of the first coal mine space according to the multi-level monitoring scheme;

[0083] A first judgment module 400 is used to identify the first coal mine space as a quasi-gas abnormal area when the gas concentration data meets the gas concentration threshold of the first coal mine space;

[0084] A concentration generation module 500 is used to evaluate the particle density of the gas concentration data and generate a particle concentration;

[0085] The second judgment module 600 is used to identify the first coal mine space as a gas abnormality area when the particle concentration meets the concentration threshold;

[0086] The abnormal factor generation module 700 is used to obtain the coal mine activity characteristics for abnormal analysis and generate gas abnormal factors;

[0087] The early warning scheme building module 800 is used to build a multi-level early warning scheme according to the gas abnormality factor and the gas concentration data.

[0088] Furthermore, the system comprises:

[0089] A first monitoring frequency setting module, used to set a first monitoring frequency according to the space volume coefficient;

[0090] A first monitoring frequency acquisition module, used to adjust the first monitoring frequency by partition according to the heat source distribution information to acquire a second monitoring frequency;

[0091] A gas concentration threshold generation module, used to adjust the gas concentration limit according to the heat source distribution information to generate the gas concentration threshold;

[0092] A monitoring scheme setting module is used to set the multi-level monitoring scheme according to the second monitoring frequency and the gas concentration threshold.

[0093] Furthermore, the system comprises:

[0094] A first sensor distribution information acquisition module, used to acquire first sensor distribution information of the gas concentration data;

[0095] A reference gas concentration data acquisition module, configured to acquire reference gas concentration data of adjacent sensors according to the first sensor distribution information, wherein the distribution distance between the adjacent sensors and the first sensor is less than or equal to an adjacent distance threshold;

[0096] The particle concentration acquisition module is used to evaluate the particle density of the gas concentration data according to the reference gas concentration data to obtain the particle concentration.

[0097] Furthermore, the system comprises:

[0098] A coordinate axis construction module, used to construct a first coordinate axis based on the distribution distance and a second coordinate axis based on the gas concentration deviation;

[0099] A particle density evaluation coordinate system construction module, used to construct a particle density evaluation coordinate system based on the first coordinate axis and the second coordinate axis;

[0100] a distribution coordinate information determination module, configured to determine distribution coordinate information of the proximity sensor in the particle density evaluation coordinate system according to the reference gas concentration data and the gas concentration data;

[0101] A distance core neighborhood construction module is used to construct a k-distance core neighborhood, wherein the k-distance core neighborhood contains k distribution coordinate information from near to far;

[0102] A reachable distance calculation module, used for calculating a plurality of reachable distances based on the k-distance core neighborhood and the distribution coordinate information;

[0103] The particle concentration generation module is used to perform mean analysis on the multiple reachable distances and calculate the inverse thereof to generate the particle concentration.

[0104] Furthermore, the system comprises:

[0105] A coal mine activity feature determination module, for the coal mine activity features including activity type features, activity location features and activity duration features, wherein the coal mine activity features belong to a preset time zone;

[0106] A gas emission prediction data acquisition module, used for inputting the activity type feature, the activity location feature and the activity duration feature into a gas emission analysis model to acquire gas emission prediction data;

[0107] The gas abnormality factor determination module is used to add the coal mine activity characteristics into the gas abnormality factor when the gas emission prediction data is greater than or equal to the gas emission threshold.

[0108] Furthermore, the system comprises:

[0109] A gas management record parameter acquisition module, used to acquire gas management record parameters, wherein the gas management record parameters include activity type records, activity location records, activity duration records and gas outflow volume records;

[0110] A first base processor training module, used for training the first base processor based on a BP neural network according to the activity type record, the activity location record, the activity duration record and the gas outflow volume record;

[0111] A first error data set setting module, used for obtaining the activity type record, the activity location record, the activity duration record and the gas outflow volume record whose output accuracy of the first base processor does not meet the preset accuracy, and setting them as the first error data set;

[0112] A second base processor training module is used for, when the first error data set is larger than the converged data amount, increasing the weight of the first error data set in the original data set, and training the second base processor based on the BP neural network;

[0113] The base processor fusion module is used for repeated training. When the Nth error data set is less than or equal to the converged data volume, the first base processor, the second base processor and the Nth base processor are fused to obtain the gas emission analysis model.

[0114] Furthermore, the system comprises:

[0115] A gas elimination scheme matching module, used to match a gas elimination scheme according to the gas anomaly factor and the gas concentration data, wherein the gas elimination scheme includes emission management personnel and emission staff;

[0116] A personnel evacuation plan generating module, used to generate a personnel evacuation plan according to the gas abnormal area, wherein the personnel evacuation plan includes evacuation range information;

[0117] A multi-level warning scheme building module is used to build the multi-level warning scheme according to the emission management personnel, the emission staff and the evacuation range information.

[0118] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0119] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept. Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application intends to include these modifications and variations.

Claims

1. A data processing method for abnormal monitoring in a coal mine, characterized in that: include: Acquiring spatial characteristics of a coal mine, wherein the spatial characteristics of the coal mine include spatial volume coefficient and heat source distribution information; Setting a multi-level monitoring scheme according to the space volume coefficient and the heat source distribution information; According to the multi-level monitoring scheme, obtaining gas concentration data of the first coal mine space; When the gas concentration data meets the gas concentration threshold of the first coal mine space, marking the first coal mine space as a quasi-gas abnormal area; Performing particle density evaluation on the gas concentration data to generate particle concentration; When the particle concentration meets the concentration threshold, marking the first coal mine space as a gas abnormality area; Obtain coal mine activity characteristics for anomaly analysis and generate gas anomaly factors; Constructing a multi-level early warning scheme according to the gas abnormality factor and the gas concentration data; The gas concentration data is subjected to particle density evaluation to generate particle concentration, including: Acquiring first sensor distribution information of the gas concentration data; According to the first sensor distribution information, collecting reference gas concentration data of adjacent sensors, wherein the distribution distance between the adjacent sensors and the first sensor is less than or equal to an adjacent distance threshold; Performing particle density evaluation on the gas concentration data according to the reference gas concentration data to obtain the particle concentration; Performing particle density evaluation on the gas concentration data according to the reference gas concentration data to obtain the particle concentration includes: A first coordinate axis is constructed based on the distribution distance, and a second coordinate axis is constructed based on the gas concentration deviation; Based on the first coordinate axis and the second coordinate axis, construct a particle density evaluation coordinate system; Determining distribution coordinate information of the proximity sensor in the particle density evaluation coordinate system according to the reference gas concentration data and the gas concentration data; Constructing a k-distance core neighborhood, wherein the k-distance core neighborhood includes k distribution coordinate information from near to far; Calculating a plurality of reachable distances based on the k-distance core neighborhood and the distribution coordinate information; A mean analysis is performed on the multiple reachable distances and the inverse is calculated to generate the particle concentration.

2. A data processing method for abnormal monitoring in a coal mine according to claim 1, characterized in that: According to the space volume coefficient and the heat source distribution information, a multi-level monitoring scheme is set, including: According to the space volume coefficient, setting a first monitoring frequency; According to the heat source distribution information, the first monitoring frequency is partitioned and adjusted to obtain a second monitoring frequency; adjusting the gas concentration limit according to the heat source distribution information to generate the gas concentration threshold; The multi-level monitoring scheme is set according to the second monitoring frequency and the gas concentration threshold.

3. The data processing method for abnormal monitoring in a coal mine according to claim 1, characterized in that: Obtain coal mine activity characteristics for anomaly analysis and generate gas anomaly factors, including: The coal mine activity characteristics include activity type characteristics, activity location characteristics and activity duration characteristics, wherein the coal mine activity characteristics belong to a preset time zone; Inputting the activity type feature, the activity location feature and the activity duration feature into a gas emission analysis model to obtain gas emission prediction data; When the gas emission prediction data is greater than or equal to the gas emission threshold, the coal mine activity characteristics are added to the gas anomaly factor.

4. A data processing method for abnormal monitoring in a coal mine as claimed in claim 3, characterized in that: Inputting the activity type feature, the activity location feature and the activity duration feature into a gas emission analysis model to obtain gas emission prediction data, the method comprising: Obtaining gas management record parameters, wherein the gas management record parameters include activity type records, activity location records, activity duration records, and gas outflow volume records; Training a first base processor based on a BP neural network according to the activity type record, the activity location record, the activity duration record and the gas outflow volume record; Obtaining the activity type record, the activity location record, the activity duration record, and the gas outflow volume record whose output accuracy of the first base processor does not meet a preset accuracy rate, and setting them as a first error data set; When the first error data set is larger than the converged data amount, the weight of the first error data set in the original data set is increased, and the second base processor is trained based on the BP neural network; Repeat the training, and when the Nth error data set is less than or equal to the converged data amount, fuse the first base processor, the second base processor, and finally the Nth base processor to obtain the gas emission analysis model.

5. The data processing method for abnormal monitoring in a coal mine according to claim 1, characterized in that: According to the gas abnormality factor and the gas concentration data, a multi-level early warning scheme is constructed, including: Matching a gas elimination plan according to the gas anomaly factor and the gas concentration data, wherein the gas elimination plan includes emission management personnel and emission staff; Generate a personnel evacuation plan according to the gas abnormality area, wherein the personnel evacuation plan includes evacuation range information; The multi-level early warning plan is constructed based on the emission management personnel, the emission staff and the evacuation range information.

6. A data processing system for abnormal monitoring in coal mines, characterized in that: A data processing method for implementing anomaly monitoring in a coal mine according to any one of claims 1 to 5, comprising: A feature acquisition module, used to acquire spatial features of a coal mine, wherein the spatial features of a coal mine include spatial volume coefficient and heat source distribution information; A monitoring scheme setting module, used to set a multi-level monitoring scheme according to the space volume coefficient and the heat source distribution information; A concentration data acquisition module, used to acquire gas concentration data of the first coal mine space according to the multi-level monitoring scheme; a first judgment module, configured to identify the first coal mine space as a quasi-gas abnormal area when the gas concentration data meets the gas concentration threshold of the first coal mine space; A concentration generation module, used to evaluate the particle density of the gas concentration data and generate a particle concentration; A second judgment module is used to identify the first coal mine space as a gas abnormality area when the particle concentration meets a concentration threshold; The abnormal factor generation module is used to obtain the characteristics of coal mine activities for abnormal analysis and generate gas abnormal factors; The early warning scheme construction module is used to construct a multi-level early warning scheme according to the gas abnormality factor and the gas concentration data.

Images