A system and method for delineating concealed fire zones using multi-source information

Abstract

This invention relates to the field of coal mine fire detection and prevention technology, and provides a multi-source information-based system and method for delineating concealed fire zones. The delineation method includes: scanning the target area to delineate primary suspected fire zones; conducting low-altitude detection on the primary suspected fire zones to acquire high-resolution thermal infrared and multispectral data, and comprehensively evaluating and delineating secondary suspected fire zones; conducting geophysical exploration and gas sampling within the secondary suspected fire zones to verify the fire zone's location, depth, and combustion state; integrating the above multi-source data to construct a fuzzy comprehensive evaluation model of fire zone hazard and calculate the comprehensive membership degree; classifying fire zone hazard levels and generating a fire zone hazard level planning map; formulating prevention and control decisions and conducting dynamic monitoring and updates. This solution achieves comprehensive fire zone identification from macro to micro and from the surface to underground, and combined with the intelligent analysis of a geographic information system, significantly improves the accuracy, timeliness, and systematic nature of concealed fire zone delineation.

Description

Technical Field

[0001] This invention relates to coal mine fire detection and prevention technology, specifically to a multi-source information-based system and method for delineating concealed fire zones. Background Technology

[0002] Hidden fire zones in coal mines refer to areas that are continuously burning below the surface or inside the coal seam and are difficult to detect through direct observation. The existence of these fire zones seriously threatens safe production in mines, not only continuously consuming coal resources and damaging the coal seam structure, but also serving as a potential source of danger for major accidents such as gas and coal dust explosions.

[0003] Currently, the detection of concealed fire zones mainly relies on single technical means, such as surface thermometry, gas detection, or geophysical methods. These traditional methods have significant limitations in practical applications: First, their detection range and accuracy are insufficient. Surface thermometry is greatly affected by vegetation and weather, gas detection is easily diluted by wind, and geophysical methods suffer from multiple interpretations, making it difficult to accurately locate the spatial position of the fire source and the combustion boundary. Second, information coordination is poor; various technologies form data silos, lacking the ability to systematically integrate large-scale initial screening, high-precision identification, and underground verification. This leads to delayed assessments and ambiguous delineation of the concealed combustion state in deep fire zones and fissure-developed areas, resulting in less targeted subsequent firefighting operations and lower prevention and control efficiency.

[0004] Therefore, how to construct an efficient, accurate, and systematic method for delineating hidden fire zones using multi-source information has become a key technical problem that urgently needs to be solved in coal mine fire prevention and control. Summary of the Invention

[0005] To address the problems existing in the background technology, this invention proposes a multi-source information concealed fire zone delineation system and method, which realizes all-round fire zone identification from macro to micro and from the surface to the underground. Combined with the intelligent analysis of geographic information system, it significantly improves the accuracy, timeliness and systematicness of concealed fire zone delineation.

[0006] To achieve the above objectives, the present invention adopts the following solution:

[0007] A method for delineating concealed fire zones using multi-source information includes the following steps:

[0008] Step 1: Scan the target area and, through surface temperature inversion and statistical threshold division, designate areas with temperatures higher than the preset threshold as Level 1 suspected fire zones.

[0009] Step 2: Conduct low-altitude detection on the first-level suspected fire zone to obtain high-resolution thermal infrared and multispectral data. Through comprehensive analysis, accurately delineate the high-temperature core and the area of ​​influence on the ground as the second-level suspected fire zone.

[0010] Step 3: Conduct geophysical exploration and gas sampling within the suspected secondary fire zone to verify the location, depth, and combustion status of the fire zone;

[0011] Step 4: Integrate the multi-source data from Steps 1-3 into the GIS geographic information system platform, construct a fuzzy comprehensive evaluation model for fire zone hazard, and calculate the comprehensive membership degree of each unit;

[0012] Step 5: Based on the comprehensive membership degree, classify the fire zone hazard level and generate a fire zone hazard level planning map;

[0013] Step 6: Formulate prevention and control decisions based on the aforementioned hierarchical planning map, and conduct dynamic monitoring and updates.

[0014] Optionally, in step 2, the comprehensive assessment process includes the following steps:

[0015] Step 2.1: Identify obvious high-temperature points with temperatures greater than 120°C and preliminarily mark them as potential ignition sources or heat flow outlets;

[0016] Step 2.2: For each obviously high temperature point, analyze the distribution characteristics of its surrounding temperature field, and combine it with the background temperature value to delineate the thermal anomaly boundary where the temperature is significantly higher than the background, as the thermal influence range based on physical evidence.

[0017] Step 2.3: Calculate the vegetation index based on multispectral data, and extract areas with a significant decrease in vegetation index as vegetation stress areas; analyze the spatial overlap between the vegetation stress areas and the thermal anomaly boundary, as well as the distribution pattern of vegetation stress, to obtain ecological thermal effect evidence for verifying the authenticity of the thermal anomaly and assisting in defining the scope of influence.

[0018] Step 2.4: The area where the thermal anomaly boundary coincides with the vegetation stress area is taken as the core evidence area for the existence of the fire area and forcibly included in the secondary suspected fire area. For areas with only thermal anomalies and no vegetation stress, they are selectively included based on their spatial correlation with the high temperature core point. For areas with only vegetation stress and no thermal anomalies, their inclusion is determined based on their distribution pattern and their positional relationship with the high temperature core point. Based on the above analysis results, a closed polygonal area is generated as the secondary suspected fire area.

[0019] Optionally, step 3 specifically includes: setting up a measurement network using the high-density resistivity method, with an electrode spacing of 5 meters and a detection depth of 80 meters; identifying resistivity anomaly areas by drawing resistivity profile maps to infer the location, range, and burial depth of the fire zone; and simultaneously using a laser gas analyzer to conduct continuous gas monitoring on surface fissures or borehole openings to verify the underground combustion state.

[0020] Optionally, in step 4, the construction and calculation process of the fuzzy comprehensive evaluation model for fire zone hazard is as follows:

[0021] Step 4.1: Construct an evaluation index system and determine weights using the analytic hierarchy process (AHP): Define the evaluation objectives and select four evaluation factors: surface temperature level X1, underground resistivity anomaly intensity X2, distance from high-temperature core point X3, and surface CO concentration level X4. Use the AHP to determine the weights of each evaluation factor as 0.3, 0.3, 0.2, and 0.2, respectively. Then, perform a consistency check by calculating the weight vector of each evaluation factor to ensure the rationality of the weight settings.

[0022] Step 4.2: Establish membership functions for each evaluation factor: Convert the measured values ​​of each factor into fuzzy degrees of membership in hazard. Specifically, the membership function for surface temperature level X1 is linearly positive; the higher the temperature, the higher the membership. Set the attention threshold Tmin = 50℃ and the significant high temperature threshold Tmax = 120℃. The membership function for underground resistivity anomaly intensity X2 is based on normalized resistivity anomaly values; the greater the resistivity anomaly intensity, or the higher the processed anomaly index, the higher the membership. Set the mild anomaly threshold Rmin = 0.2 and the strong anomaly threshold Rmax = 0.8. The membership function for distance from the high-temperature core point X3 is linearly negative; the closer the distance, the higher the membership. Set the high-risk radius Rmin = 20m and the influence radius Rmax = 200m. The membership function for surface CO concentration level X4 is linearly positive; the higher the nitric oxide concentration, the higher the membership. Set the background threshold Cmin = 20ppm and the active indicator threshold Cmax = 100ppm.

[0023] Step 4.3, Divide the evaluation units and calculate the comprehensive membership degree: In GIS, divide the study area or secondary suspected fire zone into regular grids as evaluation units. Extract the factor values ​​of each unit from each data layer, substitute them into the corresponding membership degree function to calculate the membership degree k1, k2, k3, k4 of each evaluation factor, and use the weighted average fuzzy operator to calculate the comprehensive membership degree B of each unit: B=0.3×k1+0.3×k2+0.2×k3+0.2×k4.

[0024] Optionally, in step 5, the fire zone hazard is classified into four levels:

[0025] The region with a membership degree B value in the interval (0.7, 1) is classified as Level I, which is an extremely high-risk area;

[0026] The region with a membership degree B value in the range of (0.5, 0.7) is classified as Level II, and is designated as a high-risk area;

[0027] The region with a membership degree B value in the range of (0.3, 0.5) is classified as Level III, which is considered a medium-risk area.

[0028] Regions with membership B values ​​in the interval (0, 0.3) are classified as Level IV, serving as low-risk areas.

[0029] A multi-source information concealed fire zone delineation system includes a space remote sensing system, an airborne remote sensing system, a ground detection system, and a geographic information system;

[0030] The aerospace remote sensing system is used to perform large-scale scanning of the target area, identify areas of abnormal surface temperature, and initially delineate primary suspected fire zones. The airborne remote sensing system is used to conduct low-altitude flight detection of the primary suspected fire zones, acquire high-resolution thermal infrared and multispectral data, accurately identify high-temperature cores and thermal anomaly boundaries, and form secondary suspected fire zones. The ground detection system is used to detect electrical, magnetic, or elastic wave anomalies in underground media using geophysical methods, and to collect gases escaping from the surface or boreholes, analyze gas concentrations and ratios, and verify the underground combustion status. The geographic information system is used to integrate and analyze multi-source data acquired by the aerospace remote sensing system, airborne remote sensing system, and ground detection system, and generate a fire zone hazard level planning map through spatial overlay analysis, model calculation, and visualization output.

[0031] Optionally, the aerospace remote sensing system includes a satellite remote sensing platform, a thermal infrared sensor, and a data receiving and processing unit, wherein the spatial resolution of the thermal infrared data is not less than 60 meters, and the abnormal temperature threshold is determined by the mean + N times the standard deviation or by the sliding window statistical method.

[0032] Optionally, the airborne remote sensing system includes an unmanned aerial vehicle or a manned airborne platform, a high-resolution thermal imager, and a multispectral imager, wherein the thermal imager has a temperature resolution of not less than 0.05°C and a spatial resolution of not less than 0.1 meters.

[0033] Optionally, the geophysical methods in the ground detection system include high-density resistivity method and transient electromagnetic method; the collected gases include at least CO, CO2 and CH4.

[0034] Optionally, the geographic information system uses a fuzzy comprehensive evaluation model to assess the fire risk of the fire zone. The evaluation factors of the fuzzy comprehensive evaluation model include at least the surface temperature anomaly intensity, the underground resistivity anomaly gradient, the distance from the high-temperature core point, and the surface carbon monoxide concentration.

[0035] The beneficial effects of this invention are as follows: First, this scheme constructs a progressive detection system consisting of aerospace / airborne remote sensing (macroscopic temperature anomalies) → UAV low-altitude detection → geophysical preliminary gas detection verification, significantly improving the detection accuracy and reliability of underground hidden fire zones. Firstly, a large-scale temperature anomaly screening is conducted using aerospace / airborne remote sensing to delineate primary suspected fire zones, effectively narrowing the detection range. Then, high-precision UAV detection is used within the primary suspected fire zones to generate surface temperature distribution maps, accurately identifying high-temperature core points and delineating secondary suspected fire zones. Finally, ground verification is conducted within the secondary suspected fire zones using resistivity methods and CO concentration monitoring, obtaining evidence of the fire zone's existence through underground structural damage and combustion gas release. This achieves comprehensive fire zone identification from macroscopic to microscopic levels, and from the surface to underground levels. Combined with intelligent analysis using a geographic information system, this significantly improves the accuracy, timeliness, and systematic nature of hidden fire zone delineation.

[0036] Furthermore, this scheme proposes a fuzzy comprehensive evaluation model for fire zone hazard based on the analytic hierarchy process (AHP) and fuzzy comprehensive evaluation, achieving effective fusion of multi-source data and scientific classification of hazard levels. For the acquired detection data such as surface temperature, underground resistivity, distance to high-temperature points, and surface CO concentration, membership degrees are constructed to uniformly convert them into fuzzy degrees of hazard membership, solving the problem of difficult multi-source data fusion. Simultaneously, the AHP is used to determine the weights of each evaluation factor, and the rationality of the weights is verified. Finally, a weighted average fuzzy operator is used to calculate the comprehensive membership degree of each unit, thus scientifically quantifying evidence from different dimensions such as remote sensing, geophysics, and gas chemistry into a unified, spatially visualized hazard index.

[0037] In addition, the generated hidden fire zone hazard level planning map clearly divides the area into four levels: extremely high hazard zone, high hazard zone, medium hazard zone and low hazard zone, and overlays key elements such as high temperature core point, vegetation stress zone, and resistivity abnormal zone, thereby providing a quantitative decision-making basis for the division of priority areas for fire fighting projects, the layout of monitoring points and risk management. Attached Figure Description

[0038] Figure 1 This is a flowchart of the method of the present invention;

[0039] Figure 2 This is a schematic diagram of the system of the present invention;

[0040] Figure 3 This is a schematic diagram of fire zone hazard level planning in an embodiment of the present invention.

[0041] The following are labeled in the diagram: 1. Remote sensing platform; 2. Satellite receiving station; 3. Unmanned aerial vehicle (UAV); 4. Geophysical exploration device; 5. Gas detection device; 6. GIS geographic information system platform. Detailed Implementation

[0042] To make the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the given embodiments are merely one implementation method and do not represent all embodiments.

[0043] Example 1

[0044] Combination Figure 2 This embodiment provides a multi-source information concealed fire zone delineation system, including aerospace remote sensing system, airborne remote sensing system, ground detection system and geographic information system.

[0045] Specifically, the aerospace remote sensing system includes a satellite remote sensing platform 1, a thermal infrared sensor, and a data receiving and processing unit, used to perform large-scale scanning of the target area, identify areas with abnormal surface temperatures, and initially delineate areas with abnormal surface temperatures as first-level suspected fire zones. The data processing unit is a satellite receiving station 2.

[0046] The airborne remote sensing system includes a UAV 3 or a manned airborne platform, a high-resolution thermal imager, and a multispectral imager. In this embodiment, a UAV 3 is used, wherein the thermal imager it carries has a temperature resolution of not less than 0.05°C and a spatial resolution of not less than 0.1 meters.

[0047] The ground detection system includes a geophysical detection device 4 and a gas detection device 5. Furthermore, the geophysical detection device 4 uses geophysical methods to detect electrical, magnetic, or elastic wave anomalies in the underground medium to infer the depth of the ignition source, the thickness of the combustion layer, and the development of fissures. The high-density resistivity method device can be a DUK-2B high-density electrical resistivity meter, and the main device for transient electromagnetic detection is a multi-turn small loop device. The gas detection device 5 collects gases escaping from surface fissures or boreholes, analyzes the concentrations and ratios of gases such as CO, CO2, and CH4, and verifies the underground combustion state. The gas detection device can be a laser gas analyzer.

[0048] The geophysical methods include high-density resistivity and transient electromagnetic methods. This embodiment combines the two to form a progressive detection logic, achieving complementary advantages and multiplied detection efficiency:

[0049] The transient electromagnetic method inputs a strong instantaneous current through a transmitting loop and then abruptly cuts it off. A receiving coil observes the secondary magnetic field signal generated by the attenuation of underground eddy currents. Based on the signal attenuation characteristics, it obtains the electrical information of the underground medium (low resistivity corresponds to water-rich / grouting zones, high resistivity corresponds to burned-out voids). Its core function is to rapidly detect the location of mid-to-deep fire sources, delineate large-scale water-rich zones, and construct a three-dimensional electrical model of the fire zone. It is less affected by the shielding effect of the high-resistivity surface layer and the equipment is lightweight and efficient. The high-density resistivity method's core function is to precisely characterize the lateral boundaries, vertical layering structure, and fracture development of shallow-to-medium-level fire zones. It can generate intuitive two-dimensional resistivity profiles with high spatial resolution, accurately identifying the boundaries between the burning layer and the normal zone, as well as water-filled fractures. It can also evaluate the grouting effect by comparing before-and-after treatment.

[0050] This embodiment follows a progressive process: "Aerospace / Airborne Remote Sensing (macroscopic delineation of temperature anomaly zones) → Transient Electromagnetic Method (rapid scanning of mid-to-deep layers to delineate the three-dimensional outline of the anomaly) → High-Density Resistivity Method (fine characterization of shallow and mid-layers to clarify boundaries and internal structures)," forming a complete detection chain from surface to volume to fine structure, compensating for the shortcomings of single methods in terms of detection scale and accuracy. The two methods are mutually verified through multiple parameters, achieving complementary detection blind spots and verification of multi-phase responses: the high-density resistivity method is sensitive to changes in the electrical properties of solid media and water-filled fissures, while the transient electromagnetic method is sensitive to changes in the electrical properties of water-bearing areas and deep layers. Their combination can comprehensively explain high-resistivity (combustion) and low-resistivity (water-filled) anomalies in fire zones, reducing the ambiguity of detection. Simultaneously, the transient electromagnetic method can compensate for the insufficient detection depth of the high-density resistivity method in high-resistivity surface areas, while the high-density resistivity method can compensate for the signal interference defects of the transient electromagnetic method in areas with strong electromagnetic interference and shallow low-resistivity coverage, jointly achieving comprehensive and accurate detection of the fire zone structure.

[0051] The geographic information system (GIS) is used to integrate and analyze multi-source data acquired by the aerospace remote sensing system, airborne remote sensing system, and ground detection system. Through spatial overlay analysis, model calculation, and visualization output, it generates a fire zone hazard level planning map. The GIS is a GIS platform. Furthermore, the GIS employs a fuzzy comprehensive evaluation model for fire zone hazard assessment. The evaluation factors of the fuzzy comprehensive evaluation model include at least the intensity of surface temperature anomalies, the gradient of underground resistivity anomalies, the distance to the high-temperature core point, and the surface carbon monoxide concentration.

[0052] This embodiment successfully resolves the contradiction between the limited detection range and insufficient accuracy of a single technical means by employing a large-scale initial screening with aerospace remote sensing systems, detailed identification of key areas with airborne remote sensing systems, and precise verification with ground-based detection systems. Specifically, the aerospace remote sensing system can rapidly scan an area of ​​tens of square kilometers around the mining area to identify macroscopic thermal anomalies; the UAV 3, equipped with a thermal imager and multispectral imager, can perform detailed temperature measurements on suspicious areas, accurately delineating the high-temperature core and heat diffusion boundaries on the surface; geophysical exploration and gas detection provide verification of underground vertical structures and direct evidence of combustion status, thus constructing a detection system from "area" to "point" to "volume," achieving comprehensive fire zone identification from macroscopic to microscopic, and from the surface to the underground. Moreover, combined with intelligent analysis from a geographic information system, it significantly improves the accuracy, timeliness, and systematic nature of delineating hidden fire zones.

[0053] Example 2

[0054] Combination Figure 1 This embodiment provides a method for delineating concealed fire zones using multi-source information, including the following steps:

[0055] Step 1, preliminary identification by space remote sensing: The Landsat series satellites are used to scan the mine field and surrounding area. The thermal infrared band data is B10, 10.6-11.2μm, with a spatial resolution of 30 meters.

[0056] Thermal infrared data is used to scan the target area, and surface temperature inversion and statistical thresholding are used to initially identify first-level suspected fire zones. Specifically, surface temperature inversion is performed using the radiative transfer equation method to obtain a surface temperature distribution map. Then, the mean and standard deviation of the background temperature are statistically analyzed, and areas with temperatures higher than the mean plus two times the standard deviation are designated as first-level suspected fire zones. Simultaneously, satellite temperature anomaly vector data corresponding to these areas are extracted.

[0057] Step 2, Fine-grained aerial remote sensing: For the Level 1 suspected fire zone, a UAV flight path at an altitude of 150 meters was planned, equipped with a 640×512 pixel infrared thermal imager with a thermal sensitivity of 0.05℃ and a five-band multispectral camera. After stitching and radiometric calibration of the acquired data, a surface temperature distribution map with a spatial resolution of 0.2 meters was generated. Based on this, obvious high-temperature points with temperatures >120℃ were identified in the area, and the temperature diffusion areas centered on each high-temperature point were delineated. At the same time, combined with multispectral data, the vegetation index around the high-temperature points was analyzed to see if there was a significant decrease and whether there were signs of heat stress, and the vector data of the vegetation stress area was extracted. Finally, based on comprehensive judgment, the surface influence range of the fire zone was accurately delineated as the Level 2 suspected fire zone, and high-temperature core points A, B, and C were marked.

[0058] The specific process of the comprehensive analysis is as follows:

[0059] Step 2.1: Identify obvious high-temperature points (temperature > 120℃) as indicators that underground combustion may have affected the surface or that there are strong heat conduction channels, and preliminarily mark them as potential fire sources or heat flow outlets.

[0060] Step 2.2 involves analyzing the temperature diffusion region centered on each high-temperature point and observing the distribution characteristics of the temperature field. Thermal diffusion caused by concealed fire zones typically manifests in two anomalous states: first, heterogeneous thermal diffusion, where temperature isopleths extend in a striped or beaded pattern, suggesting the orientation of underground fissures or burning coal seams; second, gradient anomalous thermal diffusion, where the temperature gradient around the high-temperature point changes abruptly, indicating a change in the heat transfer medium. Based on the above analysis, and combined with the background temperature value, a boundary for thermal anomalies with temperatures significantly higher than the background is delineated as the thermal influence range based on physical evidence.

[0061] Step 2.3 involves obtaining evidence of the ecological and indirect thermal effects of fire zone activity through multispectral vegetation analysis to verify the authenticity of the thermal anomaly and help define the scope of impact. Specific analyses include: calculating vegetation indices based on multispectral data; analyzing the spatial overlap between areas of significant vegetation index decline and temperature diffusion areas; higher overlap indicates that surface high temperatures have caused sustained damage to the ecosystem, confirming the existence of a long-term underground heat source; and analyzing the distribution patterns of vegetation stress. Isometric stress represents a large-scale baking zone, while linear stress corresponds to heat escape channels on underground fissure zones, providing a basis for inferring the morphology of underground fire zones.

[0062] Step 2.4: Conduct comprehensive analysis of the results: Areas in the Level 1 suspected fire zone with temperatures significantly higher than the threshold (>120℃) are identified as high-temperature core points and sorted and labeled with letters based on factors such as temperature value and distribution area; When accurately delineating the boundaries of Level 2 suspected fire zones, the overlapping area of ​​thermal anomalies and vegetation stress is the core evidence area and must be included in the Level 2 suspected fire zone. Areas with only thermal anomalies and no vegetation stress are designated as new or weak heat source areas, and areas with only vegetation stress and no obvious thermal anomalies are designated as thermal impact edge lag effect areas. For vegetation degradation caused by other reasons, its spatial relationship with the high-temperature core point is considered to determine whether it should be included. Finally, a closed polygonal area is formed as the Level 2 suspected fire zone, which is smaller in scope, has stronger evidence, and has a clearer target than the Level 1 suspected fire zone.

[0063] Step 3, Precise Verification via Ground Detection: Ground-based detection systems are used to conduct geophysical surveys and gas sampling within the suspected secondary fire zone to verify the fire's location, depth, and combustion status. Specifically, within the area delineated by airborne remote sensing, multiple parallel survey lines are laid out using the high-density resistivity method, forming a survey network with an electrode spacing of approximately 5 meters and a detection depth of approximately 80 meters. Resistivity profiles are drawn based on changes in underground resistivity to identify areas of resistivity anomalies, thereby inferring the fire's location, extent, and depth. Simultaneously, a laser gas analyzer is used for continuous detection and monitoring at visible surface fissures and newly constructed verification boreholes. If the CO concentration at the fissure downwind of the high-temperature core point remains consistently between 80-120 ppm, and the CO / CO2 ratio is greater than 0.05, significantly higher than the background value, this indicates the potential presence of active smoldering.

[0064] Step 4, Multi-source information fusion analysis and fuzzy comprehensive evaluation of fire zone hazard: Establish a project database in the GIS geographic information system platform 6, and import the data such as the satellite temperature anomaly vector from Step 1, the high-precision surface temperature distribution map, high temperature points, and vegetation stress zone vector from UAV 3 from Step 2, as well as the resistivity anomaly vector and surface CO concentration from Step 3, and carry out integrated data analysis.

[0065] Furthermore, a fuzzy comprehensive evaluation model for fire zone hazard was constructed to address the fuzziness and uncertainty in fire zone hazard assessment, combining qualitative and quantitative indicators through mathematical methods. Four evaluation factors were selected: surface temperature level (X1), underground resistivity anomaly intensity (X2), distance from the high-temperature core (X3), and surface CO concentration level (X4). The weights of each factor were determined using the analytic hierarchy process (AHP), with values ​​of 0.3, 0.3, 0.2, and 0.2, respectively. Membership functions were established for each evaluation factor, and the study area was divided into regular grid evaluation units. The comprehensive membership degree of the high-hazard level for each unit was calculated.

[0066] The construction and calculation process of the fuzzy comprehensive evaluation model for fire zone hazard is as follows:

[0067] Step 4.1: The analytic hierarchy process (AHP) was used to construct the evaluation index system and determine the weights. Based on the classification of fire zone hazard levels, four primary indicators were determined: surface temperature level, underground resistivity anomaly intensity, distance from the high-temperature core point, and surface CO concentration level. Secondary indicators were determined under the four indicator dimensions. Table 1 shows the hierarchical structure of the index system.

[0068] Table 1 Hierarchical Structure

[0069]

[0070] The evaluation objective is first clearly defined as the classification of fire zone hazard levels, and four primary evaluation factors are selected:

[0071] X1: Surface temperature level, weight 0.3, reflects the intensity of direct thermal anomalies;

[0072] X2: Intensity of underground resistivity anomaly, weighted at 0.3, indicating underground structural damage and combustion voids;

[0073] X3: Distance from the high-temperature core point, weight 0.2, characterizing the spatial decay effect;

[0074] X4: Surface CO concentration level, weighted 0.2, provides evidence of combustion activity.

[0075] Secondly, the weights were determined using the analytic hierarchy process (AHP), and the opinions of experts with extensive experience in the field were consulted. The importance of the factors was quantified using the 1-9 scale method (as shown in Table 2), and a judgment matrix was constructed.

[0076] Table 2 Proportional Scale Table

[0077]

[0078] For each primary evaluation factor in the criterion layer, a pairwise comparison of its importance relative to the target layer is performed to construct a criterion layer judgment matrix, assigning scores to its relative importance. For each primary evaluation factor in the criterion layer, a pairwise comparison of the importance of each secondary evaluation factor in its corresponding indicator layer is performed using the 1-9 scale method, constructing the corresponding indicator layer judgment matrix. Table 3 shows the criterion layer judgment matrix A, using the primary evaluation factors in the criterion layer as an example.

[0079] Table 3 Judgment Matrix A

[0080]

[0081] Then, the weight vector of each evaluation factor is calculated and a consistency check is performed to ensure the rationality of the weight settings. Specifically,

[0082] Calculate the geometric mean Mi of each row of the judgment matrix:

[0083] Where n=4, a ij These are matrix elements.

[0084] ,

[0085] ,

[0086] ,

[0087] ,

[0088] in, To determine the geometric mean of the first row of matrix A; To determine the geometric mean of the second row of matrix A; To determine the geometric mean of the third row of matrix A; To determine the geometric mean of the fourth row of matrix A.

[0089] Normalize Mi to obtain the weights Wi:

[0090] ,

[0091] ,

[0092] but, ,

[0093] ,

[0094] ,

[0095] ,

[0096] Among them, W1-W4 are the weight coefficients calculated from the four primary evaluation factors, and their values ​​reflect the relative importance of each factor in the "fire zone hazard" evaluation. The weighting for calculating the surface temperature level (X1); The weight for calculating the intensity of the underground resistivity anomaly (X2); The weight for calculating the distance (X3) from the high-temperature core point; The weighting for the surface CO concentration level (X4) is used in the calculation.

[0097] Therefore, the above results (0.2999, 0.2999, 0.2000, 0.2000) are consistent with the given weights (0.3, 0.3, 0.2, 0.2) in both concept and approximation, verifying the rationality of the weights. In practical applications, these weights can be used after passing a consistency test (calculating the consistency ratio CR < 0.1).

[0098] Step 4.2: Establish the membership function for each evaluation factor. Convert the measured values ​​of each factor into the degree of fuzziness of membership to the risk level (between 0 and 1, where 0 indicates no membership at all and 1 indicates full membership).

[0099] Among them, the membership function of the surface temperature level (X1) is linearly positively correlated; the higher the temperature, the higher the membership. The attention threshold Tmin = 50℃ and the significant high temperature threshold Tmax = 120℃ are set. For example, for a pixel with a temperature of 85℃, its membership k1 = (85-50) / (120-50) = 0.5.

[0100] The membership function of the underground resistivity anomaly intensity (X2) is based on the normalized resistivity anomaly value. The higher the resistivity anomaly intensity (positive value, indicating a high-resistivity combustion void), or the higher the processed anomaly index, the higher the membership degree. A light anomaly threshold Rmin = 0.2 and a strong anomaly threshold Rmax = 0.8 are set. For example, if the anomaly threshold R = 0.5, its membership degree k2 = (0.5 - 0.2) / (0.8 - 0.2) = 0.5.

[0101] The membership function of distance (X3) from the high-temperature core point is linearly negatively correlated; the closer the distance, the higher the membership. Setting the high-risk radius Rmin = 20m and the influence radius Rmax = 200m, if a unit is 100m from the nearest core point, then K3 = (200-100) / (200-20) ≈ 0.56;

[0102] The membership function of surface CO concentration level (X4) is linearly positively correlated; the higher the nitric oxide concentration, the higher the membership. With a background threshold Cmin = 20 ppm and an active index threshold Cmax = 100 ppm, if the interpolated concentration at a certain location is 60 ppm, then k4 = (60-20) / (100-20) = 0.5.

[0103] Step 4.3: Divide the evaluation units and calculate the comprehensive membership degree. In the GIS geographic information system platform 6, divide the study area (or the secondary suspected fire zone) into regular grids (e.g., 2m×2m) as evaluation units. Extract the factor values ​​of each unit from each data layer, substitute them into the corresponding membership function to calculate the membership degrees k1, k2, k3, and k4 of each factor, and use a weighted average fuzzy operator to calculate the comprehensive membership degree B of each unit: B = 0.3×k1 + 0.3×k2 + 0.2×k3 + 0.2×k4. The value of B is between [0, 1]. The closer the value is to 1, the higher the degree of danger of the unit.

[0104] This step uses AHP to determine weights, constructs fuzzy membership functions for standardization, and finally performs weighted fusion. It scientifically quantifies evidence from different dimensions such as remote sensing, geophysics, and gas chemistry into a unified, spatially visualized hazard index, providing accurate and quantitative decision-making basis for the division of priority areas for fire fighting projects, the deployment of monitoring points, and risk management.

[0105] Step 5, Hazard Level Classification and Thematic Map Generation: Based on the comprehensive membership degree (B value) of each evaluation unit calculated in Step 4, the fire zone hazard is divided into four levels: Level I is an extremely high-risk area (B > 0.7), Level II is a high-risk area (0.5 < B ≤ 0.7), Level III is a medium-risk area (0.3 < B ≤ 0.5), and Level IV is a low-risk area (B ≤ 0.3). In the GIS platform 6, the B values ​​of all evaluation units are rendered to generate a raster map of the fire zone hazard level distribution. Key elements such as high-temperature core points (A, B, C...), vegetation stress zone boundaries, resistivity anomaly zone outlines, and CO concentration contour lines are overlaid to form the "Hidden Fire Zone Hazard Level Planning Map," as shown below. Figure 3 As shown, the four colors red, yellow, green, and blue correspond to extremely high-risk areas, high-risk areas, medium-risk areas, and low-risk areas, respectively, to visually represent different hazard levels, thereby providing a quantitative basis for decision-making regarding the division of priority areas for firefighting projects, the deployment of monitoring points, and risk management.

[0106] Step 6: Formulate prevention and control decisions based on the aforementioned risk level planning map, and conduct dynamic monitoring and updates. Specifically, according to the planning map, ground drilling and gel injection fire extinguishing projects are prioritized for Level I extremely high-risk areas; for Level II high-risk areas, surface fissure sealing treatment is carried out simultaneously while strengthening monitoring. Three months after the project implementation, the process is repeated for monitoring. The results show that the temperature anomalies in the original Level I and Level II areas have significantly decreased, and the CO concentration has dropped to background levels, confirming the effectiveness of the prevention and control measures and that the fire area has been controlled. For Level III medium-risk areas, routine monitoring and preventive management measures are implemented, including: strengthening regular manual inspections, timely clearing of combustibles in the area, setting up firebreaks, promoting the planting of fire-resistant vegetation, and strengthening the monitoring and prevention of fissure development to prevent the risk of fire spread.

[0107] The specific embodiments of the present invention have been described in detail above with reference to the figures, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and these variations still fall within the protection scope of the present invention.

Claims

1. A method for delineating a multi-element information concealed fire area, characterized in that, Includes the following steps: Step 1: Scan the target area and, through surface temperature inversion and statistical threshold division, designate areas with temperatures higher than the preset threshold as Level 1 suspected fire zones. Step 2: Conduct low-altitude detection on the primary suspected fire zone to acquire high-resolution thermal infrared and multispectral data. Through comprehensive analysis, accurately delineate the high-temperature core and the area of ​​influence on the surface as the secondary suspected fire zone. The comprehensive analysis process includes the following steps: Step 2.1: Identify obvious high-temperature points with temperatures greater than 120°C and preliminarily mark them as potential ignition sources or heat flow outlets; Step 2.2: For each obviously high temperature point, analyze the distribution characteristics of its surrounding temperature field, and combine it with the background temperature value to delineate the thermal anomaly boundary where the temperature is higher than the background, as the thermal influence range based on physical evidence. Step 2.3: Calculate the vegetation index based on multispectral data, and extract areas with a significant decrease in vegetation index as vegetation stress areas; analyze the spatial overlap between the vegetation stress areas and the thermal anomaly boundary, as well as the distribution pattern of vegetation stress, to obtain ecological thermal effect evidence for verifying the authenticity of the thermal anomaly and assisting in defining the scope of influence. Step 2.4: The area where the thermal anomaly boundary coincides with the vegetation stress area is taken as the core evidence area for the existence of the fire area and forcibly included in the secondary suspected fire area. For areas with only thermal anomalies and no vegetation stress, they are selectively included based on their spatial correlation with the high temperature core point. For areas with only vegetation stress and no thermal anomalies, their inclusion is determined based on their distribution pattern and their positional relationship with the high temperature core point. Based on the above analysis results, a closed polygonal area is generated as the secondary suspected fire area. Step 3: Conduct geophysical exploration and gas sampling within the suspected secondary fire zone to verify the location, depth, and combustion status of the fire zone; Step 4: Integrate the multi-source data from steps 1-3 into the GIS geographic information system platform (6), construct a fuzzy comprehensive evaluation model for fire zone hazard, and calculate the comprehensive membership degree of each unit; the construction and calculation process of the fuzzy comprehensive evaluation model for fire zone hazard is as follows: Step 4.1: Construct an evaluation index system and determine weights using the analytic hierarchy process (AHP): Define the evaluation objectives and select four evaluation factors: surface temperature level X1, underground resistivity anomaly intensity X2, distance from high-temperature core point X3, and surface CO concentration level X4. Use the AHP to determine the weights of each evaluation factor as 0.3, 0.3, 0.2, and 0.2, respectively. Then, perform a consistency check by calculating the weight vector of each evaluation factor to ensure the rationality of the weight settings. Step 4.2: Establish membership functions for each evaluation factor: Convert the measured values ​​of each factor into fuzzy degrees of membership in hazard. Specifically, the membership function for surface temperature level X1 is linearly positive; the higher the temperature, the higher the membership. Set the attention threshold Tmin = 50℃ and the significant high-temperature threshold Tmax = 120℃. The membership function for underground resistivity anomaly intensity X2 is based on normalized resistivity anomaly values; the greater the resistivity anomaly intensity or the higher the processed anomaly index, the higher the membership. Set the mild anomaly threshold Rmin = 0.2 and the strong anomaly threshold Rmax = 0.

8. The membership function for distance from the high-temperature core point X3 is linearly negative; the closer the distance, the higher the membership. Set the high-risk radius Rmin = 20m and the influence radius Rmax = 200m. The membership function for surface CO concentration level X4 is linearly positive; the higher the nitric oxide concentration, the higher the membership. Set the background threshold Cmin = 20ppm and the active indicator threshold Cmax = 100ppm. Step 4.3, Divide the evaluation units and calculate the comprehensive membership degree: In the GIS geographic information system platform (6), the secondary suspected fire area is divided into regular grids as evaluation units. The factor values ​​of each evaluation unit are extracted from each data layer and substituted into the corresponding membership degree function to calculate the membership degree k1, k2, k3, k4 of each evaluation factor. The comprehensive membership degree B of each unit is calculated by using the weighted average fuzzy operator: B=0.3×k1+0.3×k2+0.2×k3+0.2×k4; Step 5: Based on the comprehensive membership degree, classify the fire zone hazard level and generate a fire zone hazard level planning map; Step 6: Formulate prevention and control decisions based on the aforementioned hierarchical planning map, and conduct dynamic monitoring and updates.

2. The method of claim 1, wherein: Step 3 specifically includes: setting up a measurement network using the high-density resistivity method, with an electrode spacing of 5 meters and a detection depth of 80 meters; identifying resistivity anomaly areas by drawing resistivity profile maps to infer the location, range, and burial depth of the fire zone; and simultaneously using a laser gas analyzer to conduct continuous gas monitoring on surface fissures or borehole openings to verify the underground combustion status.

3. The method of claim 1, wherein: In step 5, the fire zone hazard is classified into four levels: The region with a membership degree B value in the interval (0.7, 1) is classified as Level I, which is an extremely high-risk area; The region with a membership degree B value in the range of (0.5, 0.7) is classified as Level II, and is designated as a high-risk area; The region with a membership degree B value in the range of (0.3, 0.5) is classified as Level III, which is considered a medium-risk area. Regions with membership B values ​​in the interval (0, 0.3) are classified as Level IV, serving as low-risk areas.

4. A system for delineating a multi-element information concealed fire area, for implementing the method for delineating as claimed in any one of claims 1 to 3, characterized in that: This includes space-based remote sensing systems, airborne remote sensing systems, ground-based detection systems, and geographic information systems; The aerospace remote sensing system is used to perform large-scale scanning of the target area, identify areas with abnormal surface temperature, and preliminarily delineate first-level suspected fire zones; the aerospace remote sensing system includes a UAV (3) or a manned airborne platform, a high-resolution thermal imager and a multispectral imager, wherein the thermal imager has a temperature resolution of not less than 0.05℃ and a spatial resolution of not less than 0.1 meters; The airborne remote sensing system is used to conduct low-altitude flight detection of the first-level suspected fire zone, acquire high-resolution thermal infrared and multispectral data, and accurately identify the high-temperature core and thermal anomaly boundary to form a second-level suspected fire zone. The ground detection system is used to detect electrical, magnetic or elastic wave anomalies in underground media using geophysical methods, and to collect gases escaping from the surface or boreholes, analyze gas concentrations and ratios to verify underground combustion status. The geographic information system is used to integrate and analyze multi-source data acquired by the aerospace remote sensing system, airborne remote sensing system, and ground detection system. Through spatial overlay analysis, model calculation, and visualization output, it generates a fire zone hazard level planning map. The geographic information system uses a fuzzy comprehensive evaluation model to assess the fire zone hazard. The evaluation factors of the fuzzy comprehensive evaluation model include at least the surface temperature anomaly intensity, underground resistivity anomaly gradient, distance from the high-temperature core point, and surface carbon monoxide concentration.

5. The multi-source information concealed fire zone delineation system according to claim 4, characterized in that: The aerospace remote sensing system includes a star remote sensing platform (1), a thermal infrared sensor and a data receiving and processing unit, wherein the spatial resolution of the thermal infrared data is not less than 60 meters, and the abnormal temperature threshold is determined by the mean + N times the standard deviation or the sliding window statistical method.

6. The multi-source information concealed fire zone delineation system according to claim 4, characterized in that: The geophysical methods in the ground detection system include high-density resistivity method and transient electromagnetic method; the collected gases include at least CO, CO2 and CH4.

Images