Coal field fire area gas and particulate matter monitoring method, device and equipment and storage medium

Abstract

The embodiment of the invention provides a coal field fire area gas and particulate matter monitoring method, device and equipment and a storage medium, and the method comprises the steps: carrying out the temperature inversion of thermal infrared remote sensing image data based on a preset temperature inversion algorithm, and obtaining a temperature abnormal region; based on a pre-constructed hyperspectral detection model, performing inversion on the soil moisture data, the chlorophyll content data and the vegetation coverage data to obtain a surface vegetation growth condition; based on the temperature abnormal area and the earth surface vegetation generation condition, a preset judgment model is adopted for judgment, and the fire combustion stage is determined; based on the acquired atmospheric monitoring data, determining a two-dimensional distribution condition of gas and particulate matters in each fire combustion stage; and based on the two-dimensional distribution condition, performing overlay analysis on the atmospheric monitoring data at different heights in the temperature abnormal region to obtain the three-dimensional distribution condition of the gas and the particulate matters. Therefore, the accurate judgment of the combustion condition of the fire area is improved, and the accuracy and reliability of gas and particulate matter monitoring are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of coal mine fire prevention, and relate to but are not limited to a method, device, equipment and storage medium for monitoring gas and particulate matter in a coal mine fire zone. Background Art

[0002] Coalfield fires are caused by spontaneous combustion of coal seams on the surface or underground due to human or natural reasons, which causes an exothermic oxidation reaction and causes serious waste of resources and environmental pollution. The combustion process produces a large amount of particulate matter and toxic and harmful gases, such as CO2, CO, H2S, SO2, etc. These gases and particulate matter have a serious impact on the environment and human health. Accurately obtaining the combustion conditions of coalfield fire areas and mastering the dynamic changes in fire areas are important prerequisites for clarifying the time limit and order of fire area governance and efficient monitoring and supervision. Therefore, accurately and quickly determining the emission of particulate matter and gas in coalfield fire areas is of great significance to coalfield fire area governance and improving the ecological environment of fire areas.

[0003] At present, there are relatively many studies on the detection of spontaneous combustion gases in coalfield fire areas, mainly involving CO, CH4, CO2, H2S, SO, NO x , C x H y At present, there are three main research methods for coalfield fire gas emissions: traditional field surveys, partial remote sensing data estimation, and indoor and outdoor combined methods. There are two main methods for fire area surveys: 1) Use field surveys, such as data collection, mapping, surface geophysical exploration, surface temperature measurement, etc. to obtain the burned area of ​​coalfield fire areas. 2) Use infrared imaging, remote sensing and other technologies to obtain the burned area of ​​coalfield fire areas. However, the current research mainly focuses on extracting the scope of the fire zone. There is a lack of systematic classification of the combustion conditions in the fire zone, and it is impossible to determine the emissions of particulate matter and gas. There are still several problems: (1) Research on particulate matter emissions in coalfield fire zones has not received widespread attention. It mainly involves setting up separate sampling points on the surface and conducting single sampling. It is impossible to obtain comprehensive data results within the combustion zone, and it is difficult to effectively compare the emissions of particulate matter and gas at different combustion stages; (2) Traditional field surveys have high accuracy, but they consume a lot of manpower and material resources, and the results are instantaneous and the measurement range is limited. It is difficult to obtain data from hidden fire zones and large-scale fire zones, and it is impossible to comprehensively detect particulate matter emissions from the low-temperature oxidation stage to the burnout stage in coalfield fire zones; (3) Although the combination of indoor and outdoor methods can effectively obtain data, the combustion conditions in the laboratory are not completely consistent with those outdoors, and the data results need further verification. Summary of the invention

[0004] In order to solve the above-mentioned technical problems existing in the prior art, the present application provides a method, device, equipment and storage medium for monitoring gas and particulate matter in a coalfield fire area.

[0005] The technical method of the embodiment of the present application is implemented as follows:

[0006] The present application embodiment provides a method for monitoring gas and particulate matter in a coalfield fire area, the method comprising:

[0007] Acquire thermal infrared remote sensing image data and hyperspectral image data of the target coalfield; the hyperspectral image data at least includes soil moisture, chlorophyll content and vegetation coverage data;

[0008] Based on a preset temperature inversion algorithm, temperature inversion is performed on the thermal infrared remote sensing image data to obtain temperature anomaly areas;

[0009] Based on a pre-built hyperspectral detection model, the soil moisture data, the chlorophyll content data and the vegetation coverage data are inverted to obtain the surface vegetation growth conditions of the target coalfield;

[0010] Based on the abnormal temperature area and the growth of the surface vegetation, a preset judgment model is used to determine the fire combustion stage;

[0011] Determine the two-dimensional distribution of the gas and particulate matter at each of the fire combustion stages based on the acquired atmospheric monitoring data;

[0012] Based on the two-dimensional distribution, the atmospheric monitoring data at different heights in the temperature anomaly area are superimposed and analyzed to obtain the three-dimensional distribution of the gas and particulate matter.

[0013] The present application embodiment provides a coalfield fire zone gas and particulate matter monitoring device, the device comprising:

[0014] An acquisition module is used to acquire thermal infrared remote sensing image data and hyperspectral image data of a target coalfield; the hyperspectral image data at least includes soil moisture data, chlorophyll content data and vegetation coverage data;

[0015] An inversion module is used to perform temperature inversion on the thermal infrared remote sensing image data based on a preset temperature inversion algorithm to obtain an abnormal temperature area;

[0016] The inversion module is further used to invert the soil moisture data, the chlorophyll content data and the vegetation coverage data based on a pre-built hyperspectral detection model to obtain the surface vegetation growth conditions of the target coalfield;

[0017] A determination module, used to determine the fire combustion stage by using a preset determination model based on the abnormal temperature area and the growth of the surface vegetation;

[0018] The determination module is further used to determine the two-dimensional distribution of the gas and particulate matter at each of the fire combustion stages based on the acquired atmospheric monitoring data;

[0019] The analysis module is used to perform superposition analysis on the atmospheric monitoring data at different heights in the temperature anomaly area based on the two-dimensional distribution to obtain the three-dimensional distribution of the gas and particulate matter.

[0020] In some embodiments, the acquisition module is also used to acquire the coalfield geological information of the target coalfield; based on the coalfield geological information, determine the coalfield boundary information and flight route of the target coalfield; based on the coalfield boundary information and the flight route, determine the infrared image map; pre-process the infrared image map to obtain the thermal infrared remote sensing image data.

[0021] In some embodiments, the inversion module is also used to obtain the maximum value, minimum value, surface temperature change value between each flight strip tested synchronously, and thermal infrared image brightness value of the thermal infrared remote sensing image data; the temperature change value is determined by the surface temperature tested in real time; the maximum value, the minimum value, the surface temperature change value and the thermal infrared image brightness value are substituted into the preset temperature inversion algorithm for calculation to obtain the temperature anomaly area.

[0022] In some embodiments, the preset temperature inversion algorithm is expressed as:

[0023]

[0024] Among them, Δt represents the surface temperature change value between each flight strip of the synchronous test; G represents the brightness value of the thermal infrared image; and T represents the inversion value of the ground temperature.

[0025] In some embodiments, the chlorophyll content data is determined by: based on the preset physiological characteristic differences of vegetation at different burning stages, a correlation analysis method is used to select characteristic bands of the high light spectrum image data to obtain target characteristic sensitive bands; the target characteristic sensitive bands can best indicate the chlorophyll content data;

[0026] The method for selecting characteristic bands through correlation analysis is as follows:

[0027]

[0028] Among them, r xy is the Pearson correlation coefficient of variables x and y, indicating the degree of linear association between two sets of continuous variables; n represents the number of observations; x i represents the i-th observation value of x; y i is the i-th observation value of y.

[0029] In some embodiments, the determination module is further used to perform preprocessing, normalization processing and interpolation processing on the atmospheric monitoring data in sequence to obtain the two-dimensional distribution of the gas and particulate matter within the combustion area.

[0030] In some embodiments, the device further comprises: a construction module for performing adaptive learning rate and category feature processing using a CatBoost gradient boosting tree algorithm; the adaptive learning rate processing is calculated as follows: Where t is the number of iterations, η t is the learning rate of the tth iteration, α t is the average learning rate of the previous t rounds of iterations; the objective function of the CatBoost gradient boosting tree algorithm is: Among them, y i is the true label of sample i, is the predicted label of the sample, and θ is the model parameter; a symmetric decision tree model is constructed, and the combustion stage of the fire zone is determined in the split results of each part; based on the determined loss function, the decision tree structure is optimized to obtain the target symmetric decision tree model.

[0031] An embodiment of the present application provides an electronic device, comprising: a memory for storing executable instructions; and a processor for implementing the above-mentioned coalfield fire zone gas and particulate matter monitoring method when executing the executable instructions stored in the memory.

[0032] An embodiment of the present application provides a computer-readable storage medium that stores executable instructions for causing a processor to execute the executable instructions to implement the above-mentioned coalfield fire zone gas and particulate matter monitoring method.

[0033] The coalfield fire zone gas and particulate matter monitoring method provided in the embodiment of the present application, on the one hand, can determine the combustion state of the coalfield fire zone through the surface temperature and the distribution of surface vegetation generation, significantly improving the accurate judgment of the combustion situation of the fire zone, and the understanding of the fire zone situation is more comprehensive, no longer limited to temperature changes. Compared with traditional field geological surveys, this method saves time and effort, and can save a lot of computing resources and human resources; at the same time, it can also obtain the basic situation of high-temperature fire zones that cannot be reached on the spot, providing more comprehensive data for research; on the other hand, combined with CatBoost, a coalfield fire zone combustion stage determination method based on machine learning is proposed, which significantly improves the accuracy and reliability of gas and particulate matter monitoring and supervision; this method is based on the gas detection system, and more accurately understands the particulate matter and gas emissions of the coalfield fire zone at different combustion stages, providing a basic basis for subsequent environmental assessments of particulate matter and gas emission characteristics; it not only improves the level of judgment on the combustion situation of the fire zone, but also provides more reliable technical support for the monitoring and management of coalfield fire zones. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a structural schematic diagram of a coalfield fire zone gas and particulate matter monitoring system provided in an embodiment of the present application;

[0035] Figure 2 It is a flow chart of a method for monitoring gas and particulate matter in a coalfield fire zone provided in an embodiment of the present application;

[0036] Figure 3 It is a schematic diagram of the composition structure of the coalfield fire area gas and particulate matter monitoring device provided in the embodiment of the present application;

[0037] Figure 4 It is a schematic diagram of the composition structure of the coalfield fire area gas and particulate matter monitoring equipment provided in the embodiment of the present application. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0039] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict. Unless otherwise defined, all technical and scientific terms used in the embodiments of the present application have the same meaning as those commonly understood by those skilled in the art of the technical field of the embodiments of the present application. The terms used in the embodiments of the present application are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0040] Next, an exemplary application of the coalfield fire zone gas and particulate matter monitoring device when implemented as a server will be described.

[0041] See also Figure 1 , Figure 1 It is a structural schematic diagram of the coalfield fire zone gas and particle monitoring system 10 provided in the embodiment of the present application. In order to realize the monitoring of coalfield fire zone gas and particle, the embodiment of the present application can provide a coalfield fire zone gas and particle monitoring platform, and the coalfield fire zone gas and particle monitoring platform can be implemented as a coalfield fire zone gas and particle monitoring application. The coalfield fire zone gas and particle monitoring system 10 provided in the embodiment of the present application includes a terminal 110, a network 120 and a server 130, wherein the server 130 is a server of the coalfield fire zone gas and particle monitoring application. The server 130 can constitute the gas and particle monitoring equipment of the embodiment of the present application. The terminal 110 is connected to the server 130 via the network 120, and the network 120 can be a wide area network or a local area network, or a combination of the two.

[0042] In some embodiments, please refer to Figure 1When monitoring the gas and particulate matter in the coalfield fire area, the terminal 110 sends the gas and particulate matter monitoring task to the server 130 through the network 120. The server 130 responds to the gas and particulate matter monitoring task initiated by the terminal 110 and obtains the thermal infrared remote sensing image data and hyperspectral image data of the target coalfield; the hyperspectral image data at least includes soil moisture, chlorophyll content and vegetation coverage data; based on the preset temperature inversion algorithm, the thermal infrared remote sensing image data is temperature inverted to obtain the temperature abnormal area; based on the pre-built hyperspectral detection model, the soil moisture is detected. The data, the chlorophyll content data and the vegetation coverage data are inverted to obtain the surface vegetation growth of the target coalfield; based on the temperature anomaly area and the surface vegetation growth, a preset judgment model is used to determine the fire combustion stage; based on the acquired atmospheric monitoring data, the two-dimensional distribution of the gas and particulate matter in each fire combustion stage is determined; based on the two-dimensional distribution, the atmospheric monitoring data at different heights in the temperature anomaly area are superimposed and analyzed to obtain the three-dimensional distribution of the gas and particulate matter, that is, the gas and particulate matter monitoring results. After obtaining the gas and particulate matter monitoring results, the server 130 sends the gas and particulate matter monitoring results to the terminal 110 through the network 120.

[0043] The present application embodiment provides a method for monitoring gas and particulate matter in a coalfield fire area. Figure 2 , Figure 2 This is a flow chart of a method for monitoring gas and particulate matter in a coalfield fire area provided in an embodiment of the present application, which will be combined with Figure 2 The steps shown are explained.

[0044] Step S210, obtaining thermal infrared remote sensing image data and hyperspectral image data of the target coalfield; the hyperspectral image data at least includes soil moisture, chlorophyll content and vegetation coverage data.

[0045] Here, the hyperspectral image data at least includes temperature, soil moisture, chlorophyll content, vegetation coverage data, etc. Of course, the hyperspectral impact data may also include other types of data, which are not specifically limited in the embodiments of the present application.

[0046] In some embodiments, thermal infrared remote sensing image data is mainly used to capture thermal radiation information of coal fields. Since different materials have different absorption and emission characteristics of thermal radiation, thermal infrared images can show the temperature difference between the coal field and its surrounding environment. This has important application value in thermal anomaly detection, geothermal resource assessment, and environmental monitoring of coal fields. For example, by analyzing thermal infrared image data, the heat distribution and changes inside the coal field can be detected, and then the distribution and thickness of the coal seam can be inferred.

[0047] In some embodiments, the hyperspectral image data provides detailed spectral information of coalfield features. Hyperspectral data has the characteristics of image-spectrum integration, which contains both spatial information of features and rich spectral information. This enables hyperspectral images to more accurately distinguish different substances in coalfields, such as vegetation, water bodies, soil, and coal seams. In addition, the high spectral resolution and spectral continuity of hyperspectral data enable it to capture subtle changes and differences in coalfields, which is of great significance for the identification, classification, and resource assessment of coal seams.

[0048] In practical applications, thermal infrared remote sensing image data and hyperspectral image data can complement each other and provide comprehensive and accurate information for resource assessment, environmental monitoring and development and utilization of coal fields. By combining these two types of data, we can have a deeper understanding of the geological characteristics, environmental conditions and resource distribution of coal fields, and provide a scientific basis for gas and particulate matter monitoring in coal fields.

[0049] Step S220, based on a preset temperature inversion algorithm, temperature inversion is performed on the thermal infrared remote sensing image data to obtain a temperature anomaly area.

[0050] In some embodiments, the preset temperature inversion algorithm is usually an algorithm that uses remote sensing image data to infer the surface temperature. This algorithm is based on the principle of thermal infrared remote sensing, that is, all objects above absolute zero will emit thermal radiation, and there is a certain relationship between the temperature of the object and the intensity of the thermal radiation it emits.

[0051] The basic steps of the preset temperature inversion algorithm usually include: Radiation calibration: convert the original DN (digital number) value observed by the remote sensor into a radiation brightness value or a radiation temperature value. Atmospheric correction: Due to the absorption and scattering of thermal radiation by the atmosphere, the observed thermal infrared radiation does not come entirely from the surface. Therefore, the data needs to be atmospherically corrected to eliminate the influence of the atmosphere and obtain the true thermal radiation information of the surface. Surface emissivity estimation: Surface emissivity is the ratio of surface thermal radiation to blackbody thermal radiation, which depends on factors such as the type and state of the surface material and the observation angle. Temperature inversion: Based on Planck's law and Stefan-Boltzmann's law, the surface temperature can be inverted using the radiation brightness value and the surface emissivity.

[0052] In the preset temperature inversion algorithm, some preset parameters may need to be considered, such as preset surface emissivity, atmospheric transmittance, etc. The selection of these parameters is usually based on prior knowledge or experimental data, and may vary depending on different application scenarios and regions.

[0053] It should be noted that the embodiment of the present application does not specifically limit the type of temperature inversion algorithm.

[0054] Step S230: Based on the pre-constructed hyperspectral detection model, the soil moisture data, the chlorophyll content data and the vegetation coverage data are inverted to obtain the surface vegetation growth conditions of the target coalfield.

[0055] In some embodiments, the hyperspectral detection model is a model for detecting and analyzing target objects based on hyperspectral imaging technology. Hyperspectral imaging technology can continuously image target objects in multiple spectral bands to obtain rich spectral and spatial information. The hyperspectral detection model uses this information, combined with specific algorithms and models, to perform qualitative and quantitative detection and analysis of target objects.

[0056] In an embodiment of the present application, the pre-constructed hyperspectral detection model is used to invert soil moisture data, chlorophyll content data and vegetation coverage data, thereby determining the vegetation formation status and using it as a determination result of the combustion stage of a coalfield fire area.

[0057] When building a hyperspectral detection model, the following steps are usually required: Data collection: First, hyperspectral imaging data needs to be collected, which usually includes image information of the target object in different spectral bands. This data can be obtained through devices such as hyperspectral imagers or remote sensing satellites. Preprocessing: The collected raw data needs to undergo a series of preprocessing operations, such as denoising, calibration, and normalization, to improve the quality and reliability of the data. Feature extraction: Extract feature information related to the target object from the preprocessed data. These features may include spectral reflectance, absorptivity, emissivity, soil moisture, chlorophyll and vegetation coverage data, as well as spatial morphology, texture and other features. Model establishment: Based on the extracted feature information, hyperspectral detection models are established using algorithms such as machine learning and deep learning. These models can classify, identify or quantitatively analyze target objects based on known data. Model verification and optimization: Use an independent verification data set to verify and evaluate the established model, and optimize and adjust the model based on the evaluation results to improve its detection accuracy and generalization ability.

[0058] Step S240, based on the temperature abnormality area and the surface vegetation growth conditions, a preset determination model is used to determine the fire combustion stage.

[0059] In some embodiments, the abnormal temperature area in the coalfield fire zone refers to a local high temperature area inside the coalfield caused by spontaneous combustion or human factors. The temperature of these areas is usually significantly higher than the temperature of the surrounding environment, so they can be identified and monitored by remote sensing technology.

[0060] In some embodiments, the surface vegetation distribution information of the coalfield fire area is an environmental indicator. In practical applications, it is usually judged by soil moisture data, chlorophyll content data and vegetation coverage data.

[0061] In some embodiments, the fire combustion stages include: low temperature oxidation stage (latent period), self-heating stage (self-heating period), spontaneous combustion stage (combustion period), and burnout stage (extinguishing period).

[0062] Step S250, based on the acquired atmospheric monitoring data, determining the two-dimensional distribution of the gas and particulate matter in each of the fire combustion stages.

[0063] In the embodiment of the present application, a DJI drone equipped with the Lingxiu Mini2 atmospheric detection system is selected to obtain accurate particulate matter (PM2.5 and PM10) and gas spatial distribution information in real time. The gas types are common indicator gases of coal spontaneous combustion, CO, CO2, SO2, O3, NO2, VOCs, and CxHy.

[0064] Step S260: Based on the two-dimensional distribution, the atmospheric monitoring data at different heights in the temperature anomaly area are superimposed and analyzed to obtain the three-dimensional distribution of the gas and particulate matter.

[0065] The coalfield fire zone gas and particulate matter monitoring method, device, equipment and storage medium provided in the embodiment of the present application can, on the one hand, determine the combustion state of the coalfield fire zone through the surface temperature and the distribution of surface vegetation generation, significantly improve the accurate judgment of the combustion condition of the fire zone and the more comprehensive understanding of the fire zone condition, no longer limited to temperature changes; compared with traditional field geological surveys, this method saves time and effort, and can save a lot of computing resources and human resources; at the same time, it can also obtain the basic conditions of high-temperature fire zones that cannot be reached on the spot, providing more comprehensive data for research; on the other hand, combined with CatBoost, a coalfield fire zone combustion stage determination method based on machine learning is proposed, which significantly improves the accuracy and reliability of the determination of the combustion stage of the fire zone; this method is based on the gas detection system, and more accurately understands the particulate matter and gas emissions of the coalfield fire zone at different combustion stages, providing a basic basis for the subsequent monitoring and supervision of the particulate matter and gas emission characteristics; it not only improves the judgment level of the combustion condition of the fire zone, but also provides more reliable technical support for the monitoring and management of the coalfield fire zone.

[0066] In some embodiments, the acquisition of thermal infrared remote sensing image data of the target coalfield in the above step S210 can be achieved by the following steps S211 to S214:

[0067] Step S211, obtaining coalfield geological information of the target coalfield.

[0068] In some embodiments, coalfield geological information refers to comprehensive data and information about the coalfield and its internal geological conditions, coal resource characteristics, etc.

[0069] In some embodiments, coalfield geological information may include: geological background of coalfield formation, geological structural characteristics of the coalfield area, coal seam occurrence characteristics, coal quality characteristics, physical properties of coal, chemical properties of coal, hydrogeological conditions, geological structure, coal resource reserves and distribution, etc. Of course, coalfield geological information may also include other types of information, which is not specifically limited in the embodiments of the present application.

[0070] Step S212: determining the coalfield boundary information and flight route of the target coalfield based on the coalfield geological information.

[0071] In some embodiments, the coalfield boundary information mainly includes two parts: the natural boundary of the coalfield and the artificial boundary. The natural boundary is usually formed by natural factors such as geological structure, coal seam outcrop, fault, fold axis, etc.; while the artificial boundary is demarcated by technical means and administrative means according to the needs of exploration, development, management, etc.

[0072] Step S213, determining an infrared image map based on the coalfield boundary information and the flight route.

[0073] Step S214, preprocessing the infrared image to obtain the thermal infrared remote sensing image data.

[0074] In some embodiments, preprocessing refers to a series of processing on the original infrared image to remove noise, correct distortion, enhance image features, etc., so as to obtain thermal infrared remote sensing image data that is more suitable for subsequent analysis.

[0075] In some embodiments, preprocessing may include sensor calibration, geometric correction, radiation correction, noise removal, contrast enhancement, cropping and mosaicking, etc. Of course, preprocessing may also include other processing methods, which are not specifically limited in the embodiments of the present application.

[0076] In some embodiments, the above step S220 can be implemented by following the steps S221 to S222:

[0077] Step S221, obtaining the maximum value, minimum value, surface temperature change value between each flight strip tested synchronously, and thermal infrared image brightness value of the thermal infrared remote sensing image data.

[0078] Here, the air temperature change value is determined by the surface temperature tested in real time.

[0079] Step S222, substituting the maximum value, the minimum value, the surface temperature change value and the thermal infrared image brightness value into the preset temperature inversion algorithm for calculation to obtain the temperature anomaly area.

[0080] In some embodiments, the preset temperature inversion algorithm is expressed as:

[0081]

[0082] Among them, Δt represents the surface temperature change value between each flight strip of the synchronous test; G represents the brightness value of the thermal infrared image; and T represents the inversion value of the ground temperature.

[0083] In some embodiments, the chlorophyll content data is determined by:

[0084] Based on the preset physiological characteristics of vegetation at different burning stages, the correlation analysis method is used to select characteristic bands of the high light spectrum image data to obtain target characteristic sensitive bands; the target characteristic sensitive bands can best indicate chlorophyll content data; the method of selecting characteristic bands through correlation analysis is:

[0085]

[0086] Among them, r xy is the Pearson correlation coefficient of variables x and y, indicating the degree of linear association between two sets of continuous variables; n represents the number of observations; x i represents the i-th observation value of x; y i is the i-th observation value of y.

[0087] In some embodiments, the above step S250 can be implemented in the following manner: preprocessing, normalizing and interpolating the atmospheric monitoring data in sequence to obtain the two-dimensional distribution of the gas and particulate matter within the combustion zone.

[0088] In some embodiments, the above method may further include: using CatBoost gradient boosting tree algorithm to perform adaptive learning rate and category feature processing; the calculation method of the adaptive learning rate processing is as follows: Where t is the number of iterations, η t is the learning rate of the tth iteration, α t is the average learning rate of the previous t rounds of iterations; the objective function of the CatBoost gradient boosting tree algorithm is: Among them, y i is the true label of sample i, is the predicted label of the sample, and θ is the model parameter; a symmetric decision tree model is constructed, and the combustion stage of the fire zone is determined in the split results of each part; based on the determined loss function, the decision tree structure is optimized to obtain the target symmetric decision tree model.

[0089] The following is an explanation of an exemplary application of the embodiments of the present application in a practical application scenario.

[0090] S1, obtain the coalfield geological map of the study area, and determine the coalfield boundary and UAV flight route based on the existing geological data, coalfield overview, and field geological survey.

[0091] S2, for the coalfield boundary and the UAV flight route, use the UAV equipped with an infrared thermal imager and RTK to synchronously record the attitude parameters. Fly according to the determined route to obtain infrared images. The flight conditions require the time to be between 10:00-14:00, and the weather must be clear, with light wind and few clouds, and the wind force must be below level 4, with high visibility. While performing thermal infrared remote sensing imaging, synchronous ground temperature measurement is performed, and several ground feature landmarks with large temperature differences and significant features are selected to accurately measure and record coordinates, temperature, time, wind and other parameters, providing basic data for subsequent temperature calibration and inversion.

[0092] S3, the infrared image obtained in S2 must be pre-processed, spliced ​​and mosaicked to form a complete thermal infrared remote sensing image data (DOM).

[0093] In some embodiments, the infrared image in step S3 is preprocessed using the OpenCV open source library to convert the image type (convert the RAW format to JPG or PNG), filter and denoise, perform geometric correction, radiation correction, atmospheric correction (FLAASH atmospheric correction) using GCPS, select bands, image fusion and image enhancement.

[0094] In some embodiments, the infrared remote sensing image is preprocessed including geometric correction, radiation correction, and atmospheric correction to remove noise and interference and improve data quality and accuracy. The atmospheric correction adopts the FLAASH atmospheric correction method.

[0095] In some embodiments, the image registration map in step S3 is subjected to coarse registration and fine registration in sequence to obtain an image registration map that matches the geographic coordinates. The image fusion adopts the Laplace pyramid method to eliminate the seams and then stitch multiple images into an image covering the entire coalfield area to be measured.

[0096] In some embodiments, a digital surface model, a digital terrain model and an orthophoto map are generated for the image result map in step S3.

[0097] S4, for the thermal infrared image data after splicing processing in S3, temperature inversion is performed to obtain the surface temperature. Specifically, according to the maximum and minimum values ​​of the temperature sensing of the thermal infrared sensor, the temperature change values ​​between each flight band obtained by the synchronously tested surface temperature are used to complete the ground temperature inversion of the thermal infrared image.

[0098] In some embodiments, a temperature inversion algorithm is used to perform temperature inversion on the preprocessed remote sensing image to obtain a surface temperature image. Specifically, the temperature change values ​​between each flight band are obtained based on the maximum and minimum temperature sensing values ​​of the thermal infrared sensor and the surface temperature tested synchronously to complete the ground temperature inversion of the thermal infrared image.

[0099] In some embodiments, the surface temperature inversion method in step S4 is:

[0100]

[0101] Where Tmax and Tmin represent the maximum and minimum values ​​of the temperature sensing of the thermal infrared sensor; Δt represents the change value of the surface temperature between each flight strip of the synchronous test; G represents the brightness value of the thermal infrared image; and T represents the inversion value of the ground temperature.

[0102] S5, based on the real temperature difference inverted from the surface and combined with the brightness difference, the temperature anomaly area is determined and the isotherm map of the coalfield is generated.

[0103] S6, the drone is equipped with both a thermal infrared instrument and a hyperspectral instrument, so that it can obtain hyperspectral images for research while obtaining infrared images.

[0104] S7, pre-process the acquired hyperspectral data, remove background noise and effectively extract the bands that are sensitive to soil moisture data, chlorophyll content data and vegetation coverage data. Then perform dark current correction, white plate correction (as a reference standard for calibration or measurement), radiation correction (eliminating image distortion caused by radiation error), and atmospheric correction.

[0105] In some implementations, the preprocessing of hyperspectral data based on remote sensing images of vegetation growth includes dark current correction, whiteboard correction, radiation correction and atmospheric correction; wherein the atmospheric correction adopts the FLAASH atmospheric correction method.

[0106] In some embodiments, correlation analysis is used to select characteristic bands, vegetation indices (renormalized difference vegetation index RDVI, red edge amplitude, red edge area) are used to establish a mathematical model, and soil moisture, chlorophyll content data and vegetation coverage data are inverted to construct a data set.

[0107] In some embodiments, soil moisture may be determined by:

[0108] Data preprocessing, denoising and correction: First, the hyperspectral data is denoised to eliminate sensor noise and atmospheric effects. Subsequently, radiation correction and geometric correction are performed to ensure the accuracy and consistency of the data. Band selection and standardization: According to the research purpose, bands that are sensitive to soil moisture and vegetation cover are selected. The selected bands are standardized for subsequent analysis.

[0109] Soil moisture and vegetation coverage sensitive bands: Analyze the bands in hyperspectral data that are closely related to soil moisture content. These bands are usually related to the physical and chemical properties of soil moisture. At the same time, identify bands related to vegetation green pigment content, vegetation structure, etc. These bands are crucial for estimating vegetation coverage.

[0110] Soil moisture and vegetation coverage model: Use the selected soil moisture sensitive bands, combined with physical models (such as moisture absorption models) or statistical models (such as partial least squares regression PLSR) to establish a soil moisture inversion model. Optimize model parameters through training sets and validation sets to ensure the accuracy and generalization of the model. Use vegetation indices (such as NDVI, SAVI, etc.) or methods based on mixed pixel decomposition, combined with vegetation coverage sensitive bands, to establish a vegetation coverage estimation model. Similarly, tune the model through training data and validation data.

[0111] The accuracy of the inversion results of soil moisture and vegetation coverage was evaluated, including statistical indicators (such as root mean square error RMSE, correlation coefficient R 2 The results of the determination were spatially expressed using GIS tools to produce soil moisture distribution maps and vegetation coverage distribution maps.

[0112] S8. According to the differences in physiological characteristics of vegetation at different burning stages (chlorophyll content, wilting, etc.), the extracted spectral reflectance is used to calculate the renormalized difference vegetation index (RDVI), red edge amplitude, red edge area, etc. Correlation analysis is used to select the characteristic sensitive band that can best indicate the chlorophyll content.

[0113] In some embodiments, based on the inversion modeling analysis of the CatBoost gradient boosting tree algorithm method, a training set is used to establish a hyperspectral vegetation growth recognition criterion and a hyperspectral detection model for withered vegetation (with a set chlorophyll content threshold), and the corresponding surface vegetation growth distribution is inverted.

[0114] In some embodiments, the method for selecting characteristic bands by correlation analysis in step S8 is:

[0115]

[0116] In the formula, r xyIt is expressed as the Pearson's correlation coefficient of variables x and y, indicating the degree of linear association between two sets of continuous variables; n is the number of observations; x i It is expressed as the i-th observation value of x; y i Denoted as is the i-th observation value of y.

[0117] In some embodiments, in step S8, the calculation method of the renormalized difference vegetation index in the chlorophyll content inversion is as follows:

[0118]

[0119] DVI=ρ NIR -ρ R

[0120]

[0121] Where, RDVI is the renormalized difference vegetation index; NDVI is the normalized difference vegetation index; ρ NIR Expressed as the spectral reflectance value of the near-infrared band of the remote sensing image; ρ R Expressed as the spectral reflectance value of the infrared band; Difference Vegetation Index (DVI).

[0122] S9, for the surface fire area, the CatBoost gradient boosting tree algorithm method is used to randomly divide the training set and the test set in a ratio of 7:3. Through steps such as feature traversal and data set partitioning, a symmetric decision tree model is constructed. The burning stage of the fire area is determined in the split results of each part. The loss function is clarified and the decision tree structure is optimized to obtain an ideal model. This model is used to analyze soil moisture data, chlorophyll content data and vegetation coverage content within the fire area, and then the growth status of vegetation is judged. Combined with the coupling results of temperature and vegetation growth, the burning stage of the coalfield fire area is clarified.

[0123] In some embodiments, for the scope of the surface fire area, based on S9, the training set and the test set are randomly divided according to the general 7:3 ratio, and a model is constructed to determine the soil moisture data, chlorophyll content data and vegetation coverage, so as to clarify the vegetation growth status, and the CatBoost gradient boosting tree algorithm is used to mine the subtle differences of similar crops, and match them with the low-temperature oxidation stage (latent period), self-heating stage (self-heating period), spontaneous combustion stage (combustion period), and burnout stage (extinguishing period). In machine learning, a symmetrical decision tree is constructed after steps such as feature traversal and data set division to obtain different predicted values ​​of the split results in each part, clarify the loss function, and repeat to obtain the ideal decision tree structure. The average value of the inversion result is taken as the judgment of the combustion stage of the coalfield fire area.

[0124] S10, for the UAV flight route in S2, the Lingxiu Mini2 atmospheric monitoring system was carried out for a go-around. Since the burning particles and gases in the fire area have horizontal and vertical gradient movement, gas and particle sampling and detection were carried out at different vertical heights (50, 100, 300, 500, 800, 1000) at the same location and different horizontal distances (50, 100, 300, 500, 800, 1000) at the same height.

[0125] In some embodiments, a DJI drone M30T is selected, equipped with the Lingxiu Mini2 atmospheric detection system, to obtain accurate particulate matter (PM2.5 and PM10) and gas spatial distribution information in real time. The gas types are common indicator gases of coal spontaneous combustion, CO, CO2, SO2, O3, NO2, VOCs, and CxHy. The atmospheric monitoring system is used to detect PM2.5 and PM10 and CO, CO2, SO2, O3, NO2, VOCs, and CxHy at 50m, 100m, 300m, 500m, 800m, and 1000m in the vertical and horizontal directions for data collection.

[0126] S11, for the characteristic data obtained by the gas detector, first pre-processing is performed and then normalization is performed, and then interpolation processing is performed to obtain the corresponding grid data to obtain a two-dimensional image of the gas and particle concentration distribution within the combustion area.

[0127] In some embodiments, the preprocessing performed on the particle and gas signature datasets obtained by the gas detector includes radiometric calibration, orthorectification, and stitching.

[0128] S12, overlay and analyze the grid data at different heights in the monitoring area. The fire area range, combustion stage and gas particle data form a three-dimensional distribution map of gas and particles at different combustion stages.

[0129] S13, verify the generated results through field geological survey results.

[0130] In some embodiments, in step S7, a mathematical model is established based on the XGboost method using vegetation index (renormalized difference vegetation index RDVI, red edge amplitude, and red edge area), and its objective function is:

[0131]

[0132] Among them, i is the sample index number, N is the total amount of data imported into the jth tree, K represents all decision trees established, the first term is the loss function, which represents the error between the true value and the measured value, and the second term is the regularization term, Ω(f j) is the regularization function. There are 80 sample data, which are divided into training set and test set in a ratio of 7:3. The chlorophyll content is inverted and the inversion model results are obtained. The red edge amplitude is expressed as the maximum value of the first-order derivative of the spectrum in the red light range; the red edge area is expressed as the sum of the first-order derivatives of the spectrum in the red light range.

[0133] In some embodiments, in step S13, normalization processing is performed based on the acquired landmark data, and a deviation normalization method is used to perform a linear transformation of the original data. The conversion calculation method is as follows:

[0134]

[0135] In the formula, x* represents the normalized value of the concentration sample data; x represents the concentration sample value; max and min represent the maximum and minimum values ​​of the concentration sample data, respectively.

[0136] In some embodiments, chlorophyll is measured in the fire area and the surrounding main vegetation by a chlorophyll meter to ensure that the measurement samples involve vegetation affected by the coal fire and vegetation not affected by the fire area. Each measurement position is selected 10-15 times, and the average value is taken as the SPAD value of the leaf. Since the SPAD value (S) is unitless, in order to convert it into a practical chlorophyll content (in μg / cm 2 The known correlation function between SPAD value and chlorophyll content was used.

[0137]

[0138] Where M is the molar mass of chlorophyll, which is 907 g / mol.

[0139] In some embodiments, the calculation method for obtaining raster data by interpolation processing based on the normalized data in step S13 is:

[0140]

[0141] Where Z(S0) is the interpolation prediction value at S0; i is represented by the position weight of the measurement value at the i-th position; n is represented by the sample value around the measurement point; Z(S i ) is in S i The true value obtained.

[0142] In the embodiments of the present application, the study of particulate matter emissions in coalfield fire areas mainly relies on field surveys and measurements, which is time-consuming and labor-intensive, and it is difficult to obtain instantaneous data on hidden fire sources and large-scale fire areas. It is of great significance to use drone remote sensing technology to monitor these situations. The safety of drones in dangerous environments such as handling dangerous fires and detecting gas emissions cannot be ignored. Its flexible flight capabilities enable it to efficiently cover large areas in a short period of time, and its high temporal and spatial resolution provides detailed images and data. The ability of drones to monitor in real time not only reduces manpower and time costs, but also enables timely acquisition of data during the monitoring process. Drones equipped with a variety of sensors help to fully understand the situation in coalfield fire areas and provide timely and accurate monitoring data.

[0143] The coalfield fire zone gas and particulate matter monitoring method provided in the embodiment of the present application, on the one hand, can determine the combustion state of the coalfield fire zone through the surface temperature and the distribution of surface vegetation generation, significantly improving the accurate judgment of the combustion condition of the fire zone and the more comprehensive understanding of the fire zone condition, no longer limited to temperature changes, compared with traditional field geological surveys, this method saves time and effort, can save a lot of computing resources and human resources; at the same time, it can also obtain the basic conditions of high-temperature fire zones that cannot be reached on the spot, providing more comprehensive data for research; on the other hand, combined with CatBoost, a coalfield fire zone combustion stage determination method based on machine learning is proposed, which significantly improves the accuracy and reliability of the determination results; this method is based on the gas detection system, and more accurately understands the particulate matter and gas emissions of the coalfield fire zone at different combustion stages, which provides a basic basis for the subsequent monitoring and supervision of particulate matter and gas emission characteristics; it not only improves the level of judgment on the combustion condition of the fire zone, but also provides more reliable technical support for the monitoring and management of coalfield fire zones.

[0144] Figure 3 Schematic diagram of the composition structure of the coalfield fire area gas and particulate monitoring device provided in the embodiment of the present application. Figure 3As shown, the coalfield fire area gas and particulate matter monitoring device 300 includes: an acquisition module 301, which is used to acquire thermal infrared remote sensing image data and hyperspectral image data of the target coalfield; the hyperspectral image data at least includes soil moisture data, chlorophyll content data and vegetation coverage data; an inversion module 302, which is used to perform temperature inversion on the thermal infrared remote sensing image data based on a preset temperature inversion algorithm to obtain temperature anomaly areas; the inversion module 302 is also used to perform temperature inversion on the soil moisture data, the chlorophyll content data and the vegetation coverage data based on a pre-built hyperspectral detection model. The data are inverted to obtain the surface vegetation growth conditions of the target coalfield; a determination module 303 is used to determine the fire combustion stage by using a preset determination model based on the temperature anomaly area and the surface vegetation growth conditions; the determination module 303 is also used to determine the two-dimensional distribution of the gas and particulate matter in each fire combustion stage based on the acquired atmospheric monitoring data; the analysis module 304 is used to superimpose and analyze the atmospheric monitoring data at different heights in the temperature anomaly area based on the two-dimensional distribution to obtain the three-dimensional distribution of the gas and particulate matter.

[0145] In some embodiments, the acquisition module 301 is also used to acquire the coalfield geological information of the target coalfield; based on the coalfield geological information, determine the coalfield boundary information and flight route of the target coalfield; based on the coalfield boundary information and the flight route, determine the infrared image map; pre-process the infrared image map to obtain the thermal infrared remote sensing image data.

[0146] In some embodiments, the inversion module 302 is also used to obtain the maximum value, minimum value, surface temperature change value between each flight strip tested synchronously, and thermal infrared image brightness value of the thermal infrared remote sensing image data; the temperature change value is determined by the surface temperature tested in real time; the maximum value, the minimum value, the surface temperature change value and the thermal infrared image brightness value are substituted into the preset temperature inversion algorithm for calculation to obtain the temperature anomaly area.

[0147] In some embodiments, the preset temperature inversion algorithm is expressed as:

[0148]

[0149] Among them, Δt represents the surface temperature change value between each flight strip of the synchronous test; G represents the brightness value of the thermal infrared image; and T represents the inversion value of the ground temperature.

[0150] In some embodiments, the chlorophyll content data is determined by: based on the preset physiological characteristic differences of vegetation at different burning stages, a correlation analysis method is used to select characteristic bands of the high light spectrum image data to obtain target characteristic sensitive bands; the target characteristic sensitive bands can best indicate the chlorophyll content data;

[0151] The method for selecting characteristic bands through correlation analysis is as follows:

[0152]

[0153] Among them, r xy is the Pearson correlation coefficient of variables x and y, indicating the degree of linear association between two sets of continuous variables; n represents the number of observations; x i represents the i-th observation value of x; y i is the i-th observation value of y.

[0154] In some embodiments, the determination module 303 is further used to perform preprocessing, normalization processing and interpolation processing on the atmospheric monitoring data in sequence to obtain the two-dimensional distribution of the gas and particulate matter within the combustion area.

[0155] In some embodiments, the device further comprises: a construction module for performing adaptive learning rate and category feature processing using a CatBoost gradient boosting tree algorithm; the adaptive learning rate processing is calculated as follows: Where t is the number of iterations, η t is the learning rate of the tth iteration, α t is the average learning rate of the previous t rounds of iterations; the objective function of the CatBoost gradient boosting tree algorithm is: Among them, y i is the true label of sample i, is the predicted label of the sample, and θ is the model parameter; a symmetric decision tree model is constructed, and the combustion stage of the fire zone is determined in the split results of each part; based on the determined loss function, the decision tree structure is optimized to obtain the target symmetric decision tree model.

[0156] It should be noted that the description of the device in the embodiment of the present application is similar to the description of the above method embodiment, and has similar beneficial effects as the same method embodiment, so it will not be repeated. For technical details not disclosed in the embodiment of the device, please refer to the description of the method embodiment of the present application for understanding.

[0157] It should be noted that in the embodiments of the present application, if the above-mentioned coalfield fire zone gas and particulate matter monitoring method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a terminal to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM, Read Only Memory), a magnetic disk or an optical disk. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.

[0158] Correspondingly, the embodiment of the present application provides a coalfield fire zone gas and particulate matter monitoring device, Figure 4 Schematic diagram of the composition structure of the coalfield fire area gas and particulate matter monitoring device provided in the embodiment of the present application, such as Figure 4 As shown, the coalfield fire area gas and particle monitoring device 400 at least includes: a processor 401 and a computer-readable storage medium 402 configured to store executable instructions, wherein the processor 401 generally controls the overall operation of the coalfield fire area gas and particle monitoring device. The computer-readable storage medium 402 is configured to store instructions and applications executable by the processor 401, and can also cache data to be processed or processed by each module in the processor 401 and the coalfield fire area gas and particle monitoring device 400, which can be implemented through flash memory (FLASH) or random access memory (RAM, Random Access Memory).

[0159] The present application embodiment provides a storage medium storing executable instructions, wherein the executable instructions are stored. When the executable instructions are executed by a processor, the processor will execute the method provided by the present application embodiment, for example, Figure 2 The method shown.

[0160] In some embodiments, the storage medium can be a computer-readable storage medium, such as a ferroelectric random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disk, or a compact disk read-only memory (CD-ROM); it can also be various devices including one or any combination of the above memories.

[0161] In some embodiments, executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0162] As an example, executable instructions may, but need not necessarily, correspond to a file in a file system, may be stored as part of a file storing other programs or data, for example, in one or more scripts in a Hyper Text Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions). As an example, executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed at multiple locations and interconnected by a communication network.

[0163] The above is only an embodiment of the present application and is not intended to limit the protection scope of the present application. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.

[0164] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0165] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of further restrictions, an element defined by the statement "comprises one..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed.

[0166] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for monitoring gas and particulate matter in a coalfield fire area, characterized in that: The method comprises: Acquire thermal infrared remote sensing image data and hyperspectral image data of the target coalfield; the hyperspectral image data at least includes soil moisture, chlorophyll content and vegetation coverage data; Based on a preset temperature inversion algorithm, temperature inversion is performed on the thermal infrared remote sensing image data to obtain temperature anomaly areas; Based on a pre-built hyperspectral detection model, the soil moisture data, the chlorophyll content data and the vegetation coverage data are inverted to obtain the surface vegetation growth conditions of the target coalfield; Based on the abnormal temperature area and the growth of the surface vegetation, a preset judgment model is used to determine the fire combustion stage; Determine the two-dimensional distribution of the gas and particulate matter at each of the fire combustion stages based on the acquired atmospheric monitoring data; Based on the two-dimensional distribution, the atmospheric monitoring data at different heights in the temperature anomaly area are superimposed and analyzed to obtain the three-dimensional distribution of the gas and particulate matter.

2. The method according to claim 1, characterized in that The step of obtaining thermal infrared remote sensing image data of the target coalfield includes: Acquiring coalfield geological information of the target coalfield; Based on the coalfield geological information, determining the coalfield boundary information and flight route of the target coalfield; Determining an infrared image map based on the coalfield boundary information and the flight route; The infrared image is preprocessed to obtain the thermal infrared remote sensing image data.

3. The method according to claim 1, characterized in that The method of performing temperature inversion on the thermal infrared remote sensing image data based on a preset temperature inversion algorithm to obtain an abnormal temperature area includes: Obtaining the maximum value, minimum value, surface temperature change value between each flight band tested synchronously and thermal infrared image brightness value of the thermal infrared remote sensing image data; the temperature change value is determined by the surface temperature tested in real time; The maximum value, the minimum value, the surface temperature change value and the thermal infrared image brightness value are substituted into the preset temperature inversion algorithm for calculation to obtain the temperature anomaly area.

4. The method according to claim 1, characterized in that: The preset temperature inversion algorithm is expressed as: Among them, Δt represents the surface temperature change value between each flight strip of the synchronous test; G represents the brightness value of the thermal infrared image; and T represents the inversion value of the ground temperature.

5. The method according to claim 1, characterized in that The chlorophyll content data is determined by: Based on the differences in physiological characteristics of vegetation at different burning stages, a correlation analysis method is used to select characteristic bands of the high light spectrum image data to obtain target characteristic sensitive bands; the target characteristic sensitive bands can best indicate chlorophyll content data; The method for selecting characteristic bands through correlation analysis is as follows: Among them, r xy is the Pearson correlation coefficient of variables x and y, indicating the degree of linear association between two sets of continuous variables; n represents the number of observations; x i represents the i-th observation value of x; y i is the i-th observation value of y.

6. The method according to claim 1, characterized in that Determining the two-dimensional distribution of the gas and the particulate matter at each fire combustion stage based on the acquired atmospheric monitoring data includes: The atmospheric monitoring data is sequentially preprocessed, normalized and interpolated to obtain the two-dimensional distribution of the gas and particulate matter within the combustion zone.

7. The method according to claim 1, characterized in that The method further comprises: The CatBoost gradient boosting tree algorithm is used to perform adaptive learning rate and category feature processing; the calculation method of the adaptive learning rate processing is as follows: Where t is the number of iterations, η t is the learning rate of the tth iteration, α t is the average learning rate of the first t rounds of iterations; The objective function of the CatBoost gradient boosting tree algorithm is: Among them, y i is the true label of sample i, is the predicted label of the sample, θ is the model parameter; A symmetric decision tree model is constructed, and the burning stage of the fire zone is determined in each partial split result; Based on the determined loss function, the decision tree structure is optimized to obtain the target symmetric decision tree model.

8. A coalfield fire zone gas and particle monitoring device, characterized in that: The device comprises: An acquisition module is used to acquire thermal infrared remote sensing image data and hyperspectral image data of a target coalfield; the hyperspectral image data at least includes soil moisture data, chlorophyll content data and vegetation coverage data; An inversion module is used to perform temperature inversion on the thermal infrared remote sensing image data based on a preset temperature inversion algorithm to obtain an abnormal temperature area; The inversion module is further used to invert the soil moisture data, the chlorophyll content data and the vegetation coverage data based on a pre-built hyperspectral detection model to obtain the surface vegetation growth conditions of the target coalfield; A determination module, used to determine the fire combustion stage by using a preset determination model based on the abnormal temperature area and the growth of the surface vegetation; The determination module is further used to determine the two-dimensional distribution of the gas and particulate matter at each of the fire combustion stages based on the acquired atmospheric monitoring data; The analysis module is used to perform superposition analysis on the atmospheric monitoring data at different heights in the temperature anomaly area based on the two-dimensional distribution to obtain the three-dimensional distribution of the gas and particulate matter.

9. An electronic device, characterized in that: include: A memory for storing executable instructions; The processor is used to implement the coalfield fire area gas and particulate matter monitoring method as described in any one of claims 1 to 7 when executing the executable instructions stored in the memory.

10. A computer-readable storage medium, characterized in that: Executable instructions are stored, which are used to cause the processor to execute the executable instructions to implement the coalfield fire area gas and particulate matter monitoring method described in any one of claims 1 to 7.

Images