A method, device, terminal equipment, and storage medium for wildfire alarm on power transmission lines based on dual-polarization radar.

Abstract

This invention discloses a method, device, terminal equipment, and storage medium for wildfire alarm of transmission lines based on dual-polarization radar, belonging to the field of wildfire alarm technology. The method includes: acquiring dual-polarization radar volume scan data and real-time meteorological data of the transmission line buffer area; generating a dynamic threshold for correlation coefficient and a dynamic threshold for differential reflectivity based on the real-time meteorological data; acquiring the real-time differential reflectivity and real-time correlation coefficient of each grid point based on the dual-polarization radar volume scan data; selecting grid points with a real-time differential reflectivity not less than the dynamic threshold for differential reflectivity and a real-time correlation coefficient not greater than the dynamic threshold for correlation coefficient as suspected fire points; and outputting a wildfire alarm for the transmission line based on the dual-polarization radar volume scan data of the suspected fire points. Therefore, by implementing this invention, the problem of insufficient accuracy in fire point identification in existing technologies can be solved.

Description

Technical Field

[0001] This invention relates to the field of wildfire alarm technology, and in particular to a method, device, terminal equipment and storage medium for wildfire alarm of power transmission lines based on dual-polarization radar. Background Technology

[0002] In modern power systems, transmission lines often traverse mountainous areas with dense forests, making them vulnerable to wildfires. The high temperatures and particulate matter from wildfires significantly reduce the insulation strength of the air gaps in transmission lines, leading to air discharge or surface discharge, ultimately causing power outages. Therefore, power maintenance personnel need to respond promptly to wildfire alarms.

[0003] Existing technology relies on dual-polarization Doppler weather radar to detect echo parameters, sets fixed parameter thresholds to determine suspected fire points, and then outputs wildfire warnings. However, because only fixed thresholds are used in determining suspected fire points, grid points in the echo data with differential reflectivity not less than a preset differential reflectivity threshold and correlation coefficient not greater than a correlation coefficient threshold are considered suspected fire points. But the correlation coefficient in the echo data of actual fire points detected by the radar can increase due to high humidity and high wind speed, while the differential reflectivity can decrease due to low temperature and high wind speed. Therefore, it is easy to miss fire points, resulting in insufficient accuracy in fire point identification. Summary of the Invention

[0004] This invention provides a method, device, terminal equipment, and storage medium for wildfire alarm on power transmission lines based on dual-polarization radar, which can solve the problem of insufficient accuracy in fire point identification in the prior art.

[0005] The present invention provides a method for early warning of wildfires on power transmission lines based on dual-polarization radar, comprising: acquiring dual-polarization radar volume scan data and real-time meteorological data of the buffer area of ​​the power transmission line; wherein, the real-time meteorological data includes: real-time humidity, real-time wind speed and real-time temperature; Humidity correction amount is generated based on the deviation between real-time humidity and preset high humidity threshold and preset humidity correction coefficient; first wind speed correction amount is generated based on the deviation between real-time wind speed and preset high wind speed threshold and preset first wind speed correction coefficient; preset correlation coefficient benchmark threshold is dynamically adjusted based on humidity correction amount and first wind speed correction amount to generate correlation coefficient dynamic threshold. A temperature correction amount is generated based on the deviation between the real-time temperature and the preset low temperature threshold and the preset temperature correction coefficient; a second wind speed correction amount is generated based on the deviation between the real-time wind speed and the preset high wind speed threshold and the preset second wind speed correction coefficient; and a preset differential reflectivity benchmark threshold is dynamically adjusted based on the temperature correction amount and the second wind speed correction amount to generate a differential reflectivity dynamic threshold. Real-time differential reflectivity and real-time correlation coefficient of each grid point are obtained based on dual-polarization radar volume scan data; Grid points with real-time differential reflectance not less than the differential reflectance dynamic threshold and real-time correlation coefficient not greater than the correlation coefficient dynamic threshold are selected as suspected fire points. Wildfire warnings for power transmission lines are output based on dual-polarization radar volume scan data of suspected fire points.

[0006] Furthermore, dual-polarization radar volume scan data of the transmission line buffer zone are acquired, including: Read the power grid geographic information system ledger data and the original dual-polarization radar volume scan data; Extract spatial vector information and voltage levels of transmission lines from the power grid geographic information system ledger data; Establish a transmission line buffer zone based on the spatial vector information and voltage level of the transmission line; Based on the filtering of the original dual-polarization radar volume scan data of the transmission line buffer area, dual-polarization radar volume scan data of the transmission line buffer area is obtained.

[0007] Furthermore, the transmission line wildfire alarm output based on dual-polarization radar volume scan data of suspected fire points includes: Historical radar data of suspected fire points were extracted from dual-polarization radar volume scan data of suspected fire points. For each suspected fire point, the mean of differential reflectance, the standard deviation of differential reflectance, the mean of correlation coefficient, and the standard deviation of correlation coefficient are calculated based on real-time meteorological data and historical radar data. The differential reflectance deviation is calculated by standardizing the mean of differential reflectance and the standard deviation of differential reflectance using standard scores. The correlation coefficient deviation is calculated by standardizing the mean of correlation coefficient and the standard deviation of correlation coefficient using standard scores. Suspected fire points with differential reflectance deviations greater than the first preset standard deviation height and correlation coefficient deviations greater than the second preset standard deviation height are selected as significant fire points. Several plume objects are generated from dual-polarization radar volume scan data with significant fire points using a spatial clustering algorithm. Select the confirmed fire plume from a number of smoke plume objects; Extract the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center of the confirmed fire plume; Smoke concentration is calculated based on the reflectivity factor, correlation coefficient, and differential reflectivity at the fire center. Based on smoke concentration, a wildfire warning for the transmission line is output by calling the transmission line tripping probability model.

[0008] Furthermore, the dual-polarization radar volume scan data based on significant fire points generates several plume objects using a spatial clustering algorithm, including: Based on dual-polarization radar volume scan data of significant fire points, significant fire points at the same latitude and longitude but different radar elevation layers are classified into a single plume to be determined. For each plume to be determined, the vertical cumulative debris index is calculated by weighted vertical integration of correlation coefficients based on the dual-polarization radar volume scan data of all significant fire points in the plume to be determined. Filter out plumes with a vertical cumulative debris index greater than a preset index threshold as real plumes; Several plume objects are generated based on dual-polarization radar volume scan data of significant fire points within each real plume using a spatial clustering algorithm.

[0009] Furthermore, the process of filtering out confirmed fire plumes from a plurality of smoke plume objects includes: Based on the dual-polarization radar volume scan data of each grid point within the plume object, calculate the average Doppler spectral width of all grid points within each plume object; Select smoke plumes with an average Doppler width not less than a preset Doppler width threshold as confirmed plumes.

[0010] Furthermore, the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center of the fire plume were extracted and confirmed, including: Based on the dual-polarization radar volume scan data of each grid point within the confirmed plume, the average reflectivity factor of each confirmed plume is calculated. For each confirmed fire plume, extract the maximum reflectance factor from the reflectance factors of each grid point; calculate the maximum deviation of the reflectance factor based on the maximum reflectance factor and the average reflectance factor. If the maximum deviation of the reflectivity factor is greater than the preset maximum deviation threshold, the combustion mode of the confirmed fire plume is determined to be the concentrated combustion mode; otherwise, the combustion mode of the confirmed fire plume is determined to be the diffusion combustion mode. If the combustion mode of the fire plume is determined to be a concentrated combustion mode, the grid point corresponding to the maximum reflectivity factor in the confirmed fire plume is taken as the fire point center, and the reflectivity factor, correlation coefficient and differential reflectivity of the fire point center are extracted based on the dual polarization radar volume scan data at the fire point center. If the combustion mode of the confirmed fire plume is determined to be diffusion combustion mode, the weighted centroid of the confirmed fire plume is calculated based on the reflectivity factor of each grid point and the position of each grid point, and the weighted centroid is taken as the fire point center; the weighted average of the reflectivity factor, the weighted average of the correlation coefficient, and the weighted average of the differential reflectivity of all grid points in the confirmed fire plume are taken as the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center, respectively.

[0011] Furthermore, the calculation of smoke concentration based on the reflectivity factor, correlation coefficient, and differential reflectivity at the fire point center includes: The reflectivity factor weight term is obtained by linearly restoring the reflectivity factor at the center of the fire point. A correlation coefficient decay term is generated based on the deviation between the correlation coefficient at the fire center and the perfect correlation. After linear scaling of the differential reflectivity at the fire point center, the correction reference is adjusted based on the result of the linear scaling to obtain the differential reflectivity correction term. The smoke concentration is obtained by adjusting the smoke concentration calibration coefficient together with the reflectivity factor weight term, the correlation coefficient attenuation term, and the differential reflectivity correction term; wherein, the smoke concentration calibration coefficient is obtained by regression fitting calculation through a linear regression model established by synchronously collecting ground smoke concentration and radar echo parameters at a known wildfire site.

[0012] Another embodiment of the present invention provides a power transmission line wildfire alarm device based on dual polarization radar, including: a data acquisition module, a correlation coefficient dynamic threshold generation module, a differential reflectivity dynamic threshold generation module, a grid data acquisition module, a suspected fire point screening module, and an alarm output module; The data acquisition module is used to acquire dual-polarization radar volume scan data and real-time meteorological data of the transmission line buffer area; wherein, the real-time meteorological data includes: real-time humidity, real-time wind speed and real-time temperature; The correlation coefficient dynamic threshold generation module is used to generate a humidity correction amount based on the deviation between real-time humidity and a preset high humidity threshold and a preset humidity correction coefficient; generate a first wind speed correction amount based on the deviation between real-time wind speed and a preset high wind speed threshold and a preset first wind speed correction coefficient; and dynamically adjust the preset correlation coefficient benchmark threshold based on the humidity correction amount and the first wind speed correction amount to generate a correlation coefficient dynamic threshold. The differential reflectivity dynamic threshold generation module is used to generate a temperature correction amount based on the deviation between the real-time temperature and the preset low temperature threshold and the preset temperature correction coefficient; generate a second wind speed correction amount based on the deviation between the real-time wind speed and the preset high wind speed threshold and the preset second wind speed correction coefficient; and dynamically adjust the preset differential reflectivity benchmark threshold based on the temperature correction amount and the second wind speed correction amount to generate a differential reflectivity dynamic threshold. The grid data acquisition module is used to acquire the real-time differential reflectivity and real-time correlation coefficient of each grid point based on dual-polarization radar volume scan data. The suspected fire point screening module filters out grid points with real-time differential reflectivity not less than the differential reflectivity dynamic threshold and real-time correlation coefficient not greater than the correlation coefficient dynamic threshold as suspected fire points. The alarm output module is used to output a wildfire alarm for power transmission lines based on dual-polarization radar volume scan data of suspected fire points. Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of power transmission line wildfire alarm based on dual polarization radar as provided by the present invention.

[0013] Another embodiment of the present invention also provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform the steps of the power transmission line wildfire alarm based on dual polarization radar provided by the present invention.

[0014] The following benefits can be obtained by implementing the present invention: This invention discloses a method for wildfire alarm on power transmission lines based on dual-polarization radar. The method generates correction values ​​based on real-time humidity, wind speed, and temperature data of the power transmission line buffer zone, and dynamically adjusts the judgment thresholds of correlation coefficient and differential reflectivity in real time based on these correction values ​​to obtain corresponding dynamic thresholds for correlation coefficient and differential reflectivity. Suspected fire points are screened based on these dynamic thresholds, and a wildfire alarm is subsequently output based on these selected suspected fire points. Compared to existing methods that use fixed judgment thresholds in fire point identification, this invention, by calculating humidity correction values ​​and a first wind speed correction value using real-time meteorological data, can appropriately increase the judgment threshold of the correlation coefficient under strong winds and high humidity. Simultaneously, calculating temperature correction values ​​and a second wind speed correction value can appropriately decrease the judgment threshold of differential reflectivity under strong winds and low temperatures. This reduces the possibility of missing real fire points due to weather influences, reduces performance fluctuations under complex weather conditions, and improves the accuracy of suspected fire point identification. The high-precision suspected fire point identification can further improve the output accuracy of wildfire alarms. Attached Figure Description

[0015] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating a method for wildfire warning of power transmission lines based on dual-polarization radar, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a power transmission line wildfire alarm device based on dual-polarization radar provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of flame partitioning provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0019] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0020] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0022] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0023] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0024] See Figure 1 To address the insufficient accuracy of fire point identification in existing technologies, an embodiment of the present invention provides a method for wildfire alarm of transmission lines based on dual-polarization radar, comprising: 101. Acquire dual-polarization radar volume scan data and real-time meteorological data of the transmission line buffer zone; wherein the real-time meteorological data includes: real-time humidity, real-time wind speed and real-time temperature.

[0025] In a preferred embodiment, acquiring dual-polarization radar volume scan data of the transmission line buffer zone includes: Read the power grid geographic information system ledger data and the original dual-polarization radar volume scan data; Extract spatial vector information and voltage levels of transmission lines from the power grid geographic information system ledger data; Establish a transmission line buffer zone based on the spatial vector information and voltage level of the transmission line; Based on the filtering of the original dual-polarization radar volume scan data of the transmission line buffer area, dual-polarization radar volume scan data of the transmission line buffer area is obtained.

[0026] The power grid geographic information system (GIS) ledger data includes: spatial vector information of transmission lines and voltage levels of transmission lines; the original dual-polarization radar volume scan data includes: reflectivity factor ( Differential reflectivity ), correlation coefficient ( ) and Doppler spectral width ( The specific physical meanings and units are shown in Table 1: Table 1 Specifically, the spatial vector information of transmission lines is read from the power grid geographic information system ledger data to obtain the geometric shape of the transmission lines, which is usually a multi-segment polyline. Based on the geometry of the transmission lines, a buffer sub-region with a preset distance is generated for each segment of the multi-segment polyline. The preset distance (D) is determined according to the voltage level of the transmission line. The relationship between the preset distance (D) and the voltage level of the transmission line is as follows: D ; Boolean summation is performed on all buffer sub-regions to obtain the transmission line buffer area; the transmission line buffer area is converted into a grid mask with the same resolution as the radar grid, generating a grid mask matrix; the original dual-polarization radar volume scan data is filtered using the grid mask matrix, and the original dual-polarization radar volume scan data is multiplied by the elements of the grid mask matrix. The expression of the grid mask matrix is ​​as follows: ; For a certain grid point of the original dual-polarization radar volume scan data If the latitude and longitude coordinates of the grid points If the data belongs to the transmission line buffer zone, the grid points are retained; otherwise, the grid points are discarded. The retained grid point data is used as the dual-polarization radar volume scan data of the transmission line buffer zone.

[0027] In the above steps for acquiring dual-polarization radar volume scan data, based on the premise that transmission lines only account for 5%-10% of the radar coverage area, optimization is performed for the power grid application scenario. The original dual-polarization radar volume scan data is initially screened using the power grid geographic information system ledger data, which reduces the number of grid points that need to be processed and improves the speed of subsequent calculations. The existing technology is not optimized for the power grid application scenario and requires a 360° all-round scan of the entire radar coverage area, which results in excessively long data processing time and makes it difficult to meet the power grid's 6-minute real-time alarm requirement.

[0028] Specifically, for each grid point in the transmission line buffer zone, the position information of the grid point is recorded using radial distance, azimuth angle, and elevation angle, and used as the subscript of the feature vector. Simultaneously, the reflectivity factor, differential reflectivity, correlation coefficient, and Doppler spectral width are used as elements of the feature vector; for a radial distance of... azimuth angle is Angle of elevation is The lattice points, their eigenvectors As shown below: ; In the formula, The reflectivity factor is the grid point, and the differential reflectivity is... The reflectivity factor of the grid points, The correlation coefficient of the grid points. The Doppler spectral width at the lattice point; Based on real-time meteorological data updated by the local meteorological system in the power transmission line buffer zone, real-time humidity, real-time wind speed, and real-time temperature are obtained.

[0029] In the above data acquisition steps, the corresponding data of each grid point in the dual-polarization radar volume scan data is integrated into a feature vector to facilitate subsequent calling and calculation; at the same time, real-time meteorological data is acquired as the basis for calculating the dynamic threshold.

[0030] 102. Generate a humidity correction amount based on the deviation between the real-time humidity and the preset high humidity threshold and the preset humidity correction coefficient; generate a first wind speed correction amount based on the deviation between the real-time wind speed and the preset high wind speed threshold and the preset first wind speed correction coefficient; dynamically adjust the preset correlation coefficient benchmark threshold based on the humidity correction amount and the first wind speed correction amount to generate a dynamic correlation coefficient threshold.

[0031] Specifically, the calculation method for the dynamic threshold of the correlation coefficient is as follows: ; In the formula, The dynamic threshold for the correlation coefficient; The correlation coefficient benchmark threshold is set to 0.7 in this embodiment; Real-time humidity; The preset humidity correction factor is set to 0.002 in this embodiment; Real-time wind speed; The preset first wind speed correction coefficient is set to 0.01 in this embodiment; Under high humidity conditions, such as when the real-time humidity is greater than 70%, the adhesion of water vapor causes the overall correlation coefficient to increase, and the dynamic threshold needs to be relaxed. Under windy conditions, such as when the real-time wind speed is greater than 10 meters per second, the plume is rapidly diluted, the particle dispersion increases, the correlation coefficient increases, and the dynamic threshold also needs to be relaxed.

[0032] In the above-mentioned dynamic threshold calculation steps for correlation coefficient, the correction amount is calculated based on the real-time humidity and real-time wind speed of real-time meteorological data, and then the baseline threshold for correlation coefficient is dynamically adjusted based on the correction amount, appropriately increasing the judgment threshold for correlation coefficient under strong winds and high humidity.

[0033] 103. Generate a temperature correction amount based on the deviation between the real-time temperature and the preset low temperature threshold and the preset temperature correction coefficient; generate a second wind speed correction amount based on the deviation between the real-time wind speed and the preset high wind speed threshold and the preset second wind speed correction coefficient; dynamically adjust the preset differential reflectivity benchmark threshold based on the temperature correction amount and the second wind speed correction amount to generate a differential reflectivity dynamic threshold.

[0034] Specifically, the calculation method for the dynamic threshold of differential reflectivity is as follows: ; In the formula, The dynamic threshold for differential reflectivity; The differential reflectance reference threshold is set to 3.0 dB in this embodiment; Real-time temperature; The preset temperature correction factor is set to 0.03 in this embodiment; Real-time wind speed; The preset first wind speed correction coefficient is set to 0.05 in this embodiment; Under low-temperature weather conditions, such as when the real-time temperature is less than 15 degrees Celsius, the flame combustion is incomplete, the ash production is low, the differential reflectivity decreases, and the dynamic threshold needs to be lowered. Under windy weather conditions, such as when the real-time wind speed is greater than 10 meters per second, the plume is rapidly diluted, the concentration of flat particles decreases, the differential reflectivity is low, and the dynamic threshold also needs to be lowered.

[0035] In an alternative embodiment, the dynamic threshold can be automatically learned and formulated using a machine learning model such as a random forest.

[0036] In the above-mentioned differential reflectance dynamic threshold calculation steps, the correction amount is calculated based on the real-time temperature and real-time wind speed of the real-time meteorological data, and then the differential reflectance benchmark threshold is dynamically adjusted according to the correction amount, appropriately reducing the judgment threshold of differential reflectance during strong winds and low temperatures; since the correlation coefficient and differential reflectance have different dimensions and value ranges, a first wind speed correction coefficient and a second wind speed correction coefficient are set so that the corresponding wind speed correction amount can be scaled to the range of the corresponding benchmark threshold.

[0037] 104. Based on dual-polarization radar volume scan data, obtain the real-time differential reflectivity and real-time correlation coefficient of each grid point.

[0038] Specifically, the feature vector generated in step 101 is directly extracted, and the real-time differential reflectance and real-time correlation coefficient in the feature vector are further extracted for subsequent processes.

[0039] 105. Grid points with real-time differential reflectance not less than the differential reflectance dynamic threshold and real-time correlation coefficient not greater than the correlation coefficient dynamic threshold are selected as suspected fire points.

[0040] Specifically, the dynamic threshold test is as follows: ; In the formula, The dynamic threshold for the correlation coefficient; The dynamic threshold for differential reflectivity; Grid points that pass the dynamic threshold test are considered as suspected fire points.

[0041] In the above suspected fire point screening steps, the dynamic thresholds for correlation coefficient and differential reflectance calculated in steps 102 and 103 are used as direct criteria for determining suspected fire points. This is in contrast to traditional methods that directly use fixed thresholds, such as screening grid points with differential reflectance greater than 0.3 dB and correlation coefficient greater than 0.7 as suspected fire points (i.e.: , However, the correlation coefficient increases due to high humidity and high wind speed, while the differential reflectance decreases due to low temperature and high wind speed, which can easily lead to missed fire points. The method introduces a dynamic threshold in the screening step of suspected fire points, which can reduce the situation of missed real fire points due to weather, reduce the fluctuation of recognition performance under complex weather conditions, and improve the recognition accuracy of suspected fire points.

[0042] 106, 106, Based on dual-polarization radar volume scan data of suspected fire points, output power line wildfire alarm.

[0043] In a preferred embodiment, the transmission line wildfire alarm output based on dual-polarization radar volume scan data of suspected fire points includes: Historical radar data of suspected fire points were extracted from dual-polarization radar volume scan data of suspected fire points. For each suspected fire point, the mean of differential reflectance, the standard deviation of differential reflectance, the mean of correlation coefficient, and the standard deviation of correlation coefficient are calculated based on real-time meteorological data and historical radar data. The differential reflectance deviation is calculated by standardizing the mean of differential reflectance and the standard deviation of differential reflectance using standard scores. The correlation coefficient deviation is calculated by standardizing the mean of correlation coefficient and the standard deviation of correlation coefficient using standard scores. Suspected fire points with differential reflectance deviations greater than the first preset standard deviation height and correlation coefficient deviations greater than the second preset standard deviation height are selected as significant fire points. Several plume objects are generated from dual-polarization radar volume scan data with significant fire points using a spatial clustering algorithm. Select the confirmed fire plume from a number of smoke plume objects; Extract the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center of the confirmed fire plume; Smoke concentration is calculated based on the reflectivity factor, correlation coefficient, and differential reflectivity at the fire center. Based on smoke concentration, a wildfire warning for the transmission line is output by calling the transmission line tripping probability model.

[0044] The historical radar data is pre-collected from the baseline database and includes dual-polarization radar volume scan data for each cell within the transmission line buffer zone. Based on combinations of latitude and longitude coordinates, weather type, and seasonal classification, corresponding combinations of historical dual-polarization radar volume scan data are extracted. The differential reflectance mean, differential reflectance standard deviation, correlation coefficient mean, correlation coefficient standard deviation, reflectance factor mean, and reflectance factor standard deviation are calculated based on the corresponding combinations of historical dual-polarization radar volume scan data. The baseline data is stored as follows: In the formula, This is a set of baseline data corresponding to a combination of latitude and longitude coordinates, weather type, and season. The mean of the reflectivity factor, The standard deviation of the reflectivity factor. The average of differential reflectance. The standard deviation of differential reflectance. The average of the correlation coefficients. The standard deviation of the correlation coefficient; radial distance is azimuth angle is Angle of elevation is The method for calculating the latitude and longitude coordinates of the grid points is as follows: (1) Calculate the projected distance and vertical height of the target point on the ground: S ; ; In the formula, S is the projected distance of the grid point on the horizontal plane, in kilometers, used for subsequent geographic coordinate calculation; h is the vertical height difference of the grid point relative to the radar station, in meters. (2) Calculate the geographical offset of the target point relative to the radar station: Spherical geometric calculations based on small-angle approximation: ; ; In the formula, The average radius of the Earth is taken as 6371 km in this embodiment; Latitude of the radar station; This represents the latitudinal offset of the grid points relative to the radar station. This represents the longitude offset of the grid points relative to the radar station. (3) Overlaying the geographical coordinates of the radar station: l ; l ; In the formula, Latitude of the radar station; Longitude of the radar station; l For grid point latitude; l Longitude of the grid points; In illustrative terms, the above formula is a simplified version with a small angle approximation, applicable to scenarios with radial distances less than 200 kilometers. Under the typical radar detection range of this embodiment, the conversion error is approximately 50 to 100 meters, meeting the spatial resolution requirement of 0.5 kilometers. For higher accuracy, the Vincenty formula or the Haversine formula can be used. Weather type is determined based on reflectance factor and correlation coefficient. The specific discriminant formula for weather type is shown below: ; Specifically, for each suspected fire point, the mean differential reflectance, standard deviation of differential reflectance, mean correlation coefficient, and standard deviation of correlation coefficient are extracted from the baseline data. The differential reflectance bias is calculated using standard score standardization based on the mean and standard deviation of differential reflectance. The correlation coefficient bias is calculated using standard score standardization based on the mean correlation coefficient and standard deviation of correlation coefficient. The formula for standard score standardization (Z-score standardization) is shown below: ; ; In the formula, This refers to the differential reflectivity deviation. This refers to the correlation coefficient deviation. Suspected fire points with differential reflectance deviations greater than the first preset standard deviation and correlation coefficient deviations greater than the second preset standard deviation are selected as significant fire points: ; In the formula, both the first preset standard deviation height and the second preset standard deviation height are set to 3 standard deviation heights; In the above significance test steps, by introducing baseline data as the basis for significance testing, normal anomalies such as seasonal insects and slight precipitation can be excluded, thereby reducing the false alarm rate.

[0045] In another preferred embodiment, the dual-polarization radar volume scan data based on significant fire points generates several plume objects using a spatial clustering algorithm, including: Based on dual-polarization radar volume scan data of significant fire points, significant fire points at the same latitude and longitude but different radar elevation layers are classified into a single plume to be determined. For each plume to be determined, the vertical cumulative debris index is calculated by weighted vertical integration of correlation coefficients based on the dual-polarization radar volume scan data of all significant fire points in the plume to be determined. Filter out plumes with a vertical cumulative debris index greater than a preset index threshold as real plumes; Several plume objects are generated based on dual-polarization radar volume scan data of significant fire points within each real plume using a spatial clustering algorithm.

[0046] Specifically, significant fire points at the same latitude and longitude but different radar elevation layers are grouped into a single plume to be determined; the vertical cumulative debris index is calculated by weighted vertical integration using correlation coefficients based on the dual-polarization radar volume scan data of all significant fire points in the plume to be determined. ; In the formula, The vertical cumulative debris index is M, which is the total number of elevation angle layers, and in this embodiment, it is 9 to 11 layers. For the first The vertical thickness corresponding to the elevation angle of the significant fire point of the layer; for the elevation angle of the k-th layer The vertical height of the significant fire point is: ; The vertical thickness between two adjacent layers is: ; When the elevation angle θ < 10°, due to We can approximate that the r values ​​are the same for each layer, that is... , At this point: ; When the elevation angle is large, such as θ>10°, the ground projection coordinates shift significantly. Therefore, a "vertical column" tolerance matching method is used, defining the vertical tolerance radius. If the projection offset is less than the radius, they are considered to be different height layers within the same vertical column. Filter out plumes with a vertical cumulative debris index greater than a preset index threshold as true plumes: ; In the formula, the preset exponential threshold is set to 15. ; Based on dual-polarization radar volume scan data of significant fire points within each real plume, several plume objects are generated using the spatial clustering (DBSCAN) algorithm: ; In the formula, For adaptive neighborhood radius, the baseline clustering radius , The radial distance of the significant fire point; The output consists of several independent plume objects. Each object contains several significant fire points.

[0047] In an optional embodiment, a hierarchical density clustering algorithm (Hierarchical DBSCAN) is used to automatically determine the number of clusters. This algorithm does not require a preset cluster radius and can adaptively determine the cluster radius.

[0048] In the above steps of generating smoke plume objects, ground clutter or high-altitude birds that only appear in a single layer are first removed by vertical cumulative debris index verification, and the vertical continuity of the smoke plume is confirmed; then, discrete significant fire point pixels are aggregated into independent smoke plume objects by DBSCAN algorithm, and distance adaptive clustering radius is used to avoid over-segmentation at close range and omission of small fire points at long range.

[0049] In yet another preferred embodiment, the step of filtering the confirmed fire plume among a plurality of smoke plume objects includes: Based on the dual-polarization radar volume scan data of each grid point within the plume object, calculate the average Doppler spectral width of all grid points within each plume object; Select smoke plumes with an average Doppler width not less than a preset Doppler width threshold as confirmed plumes.

[0050] Specifically, calculate each plume object The average Doppler spectral width of all significant fire points within the interior: ; In the formula, For plume objects The average Doppler spectral width; For plume objects The number of significant fire points included; Significant fire point The corresponding Doppler spectral width; Filter out plumes with an average Doppler width not less than a preset Doppler width threshold as confirmed plumes: ; The above In this process, based on the fact that wildfire plumes are caused by intense thermal convection and generate violent turbulence, the Doppler spectrum width is usually 5-10 m / s; insect swarms have relatively consistent flight speeds, with a spectrum width of usually 2-3 m / s; and precipitation raindrops have small dispersion in their falling speeds, with a spectrum width of usually 1-2 m / s; it is necessary to distinguish between normal scatterers (such as insect swarms, bird flocks, and precipitation raindrops) and real plumes.

[0051] In other preferred embodiments, the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center of the fire plume are extracted and confirmed, including: Based on the dual-polarization radar volume scan data of each grid point within the confirmed plume, the average reflectivity factor of each confirmed plume is calculated. For each confirmed fire plume, extract the maximum reflectance factor from the reflectance factors of each grid point; calculate the maximum deviation of the reflectance factor based on the maximum reflectance factor and the average reflectance factor. If the maximum deviation of the reflectivity factor is greater than the preset maximum deviation threshold, the combustion mode of the confirmed fire plume is determined to be the concentrated combustion mode; otherwise, the combustion mode of the confirmed fire plume is determined to be the diffusion combustion mode. If the combustion mode of the fire plume is determined to be a concentrated combustion mode, the grid point corresponding to the maximum reflectivity factor in the confirmed fire plume is taken as the fire point center, and the reflectivity factor, correlation coefficient and differential reflectivity of the fire point center are extracted based on the dual polarization radar volume scan data at the fire point center. If the combustion mode of the confirmed fire plume is determined to be diffusion combustion mode, the weighted centroid of the confirmed fire plume is calculated based on the reflectivity factor of each grid point and the position of each grid point, and the weighted centroid is taken as the fire point center; the weighted average of the reflectivity factor, the weighted average of the correlation coefficient, and the weighted average of the differential reflectivity of all grid points in the confirmed fire plume are taken as the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center, respectively.

[0052] Specifically, calculate the average reflectance factor for each confirmed fire plume, and extract the maximum reflectance factor; calculate the maximum deviation of the reflectance factor based on the maximum reflectance factor and the average reflectance factor. ; In the formula, This represents the maximum deviation of the reflectivity factor. This represents the maximum value of the reflectivity factor. This represents the average reflectivity factor. If the maximum deviation of the reflectance factor is greater than the preset maximum deviation threshold ( If there is a clear fire core, the combustion mode of the fire plume is determined to be a concentrated combustion mode; otherwise, it indicates that the smoke plume is evenly diffused, and the combustion mode of the fire plume is determined to be a diffusion combustion mode. If the combustion mode of the fire plume is confirmed to be a concentrated combustion mode, the grid point corresponding to the maximum reflectivity factor within the confirmed fire plume is taken as the fire center. The reflectivity factor, correlation coefficient, and differential reflectivity of the fire center are extracted based on the dual-polarization radar volume scan data at the fire center. The fire center is located by identifying the point of maximum intensity. ; In the formula, The latitude of the fire's center; Longitude of the fire's center; The latitude of the grid point corresponding to the maximum reflectivity factor; The longitude of the grid point corresponding to the maximum reflectivity factor; If the combustion mode of the confirmed fire plume is determined to be diffusion combustion mode, the weighted centroid of the confirmed fire plume is calculated based on the reflectivity factor and position of each grid point, and the weighted centroid is taken as the center of the fire point: ; ; In the formula, l For grid points i latitude; l For grid points i longitude; For grid points i Reflectivity factor; The weighted average of reflectance factors, the weighted average of correlation coefficients, and the weighted average of differential reflectance for all grid points within the fire plume will be used as the reflectance factor, correlation coefficient, and differential reflectance of the fire center, respectively. ; ; ; In the formula, This is the weighted average of the reflectivity factors; This is the weighted average of the correlation coefficients; This is the weighted average of the differential reflectance. For grid points i Reflectivity factor; For grid points i Correlation coefficient; For grid points i Differential reflectivity.

[0053] In the above-mentioned steps for extracting the fire point center parameters, the combustion mode of the fire plume is determined and confirmed based on the reflectivity factor, and the fire point center is located based on the combustion mode. Finally, the fire point center parameters are calculated, providing a physical basis for subsequent smoke concentration calculations.

[0054] In a further preferred embodiment, the calculation of smoke concentration based on the reflectivity factor, correlation coefficient, and differential reflectivity at the fire point center includes: The reflectivity factor weight term is obtained by linearly restoring the reflectivity factor at the center of the fire point. A correlation coefficient decay term is generated based on the deviation between the correlation coefficient at the fire center and the perfect correlation. After linear scaling of the differential reflectivity at the fire point center, the correction reference is adjusted based on the result of the linear scaling to obtain the differential reflectivity correction term. The smoke concentration is obtained by adjusting the smoke concentration calibration coefficient together with the reflectivity factor weight term, the correlation coefficient attenuation term, and the differential reflectivity correction term; wherein, the smoke concentration calibration coefficient is obtained by regression fitting calculation through a linear regression model established by synchronously collecting ground smoke concentration and radar echo parameters at a known wildfire site.

[0055] Specifically, the formula for calculating smoke concentration is as follows: C ; ; In the formula, C is the smoke concentration, with the unit being mg / m³; The value of is the smoke concentration calibration coefficient, which ranges from 0.03 to 0.08 in this embodiment. The reflectivity factor at the center of the fire point; The correlation coefficient for the center of the fire point; The differential reflectivity at the center of the fire point; The regression fitting function for a linear regression model established to simultaneously collect ground smoke concentration and radar echo parameters at a known wildfire site.

[0056] In an optional embodiment, the smoke concentration calculation formula may incorporate a vertical extension height correction: ; In the formula, The plume height is in meters and is calculated using a radar vertical profile. In another alternative embodiment, the smoke concentration estimation formula can also employ a neural network model for direct regression.

[0057] In the above-mentioned steps for calculating smoke concentration, the smoke concentration is calculated based on the parameters of the fire center and the smoke concentration calibration coefficient, providing an important basis for subsequent transmission line tripping risk assessment.

[0058] In a further embodiment of the present invention, the step of outputting a wildfire alarm for a transmission line based on smoke concentration by calling a transmission line tripping probability model includes: Using the power grid GIS system, the transmission lines within a 5-kilometer radius of the fire center were searched, and one of the lines was selected. The line parameters are read and input together with the fire center parameters into the transmission line tripping probability model. The specific wildfire alarm output process is as follows: ; In the formula, For power transmission lines Alarm coefficient; For power transmission lines Voltage level; transmission lines altitude above the ground; transmission lines The distance between the phases; For power transmission lines The altitude of the route; From the fire center to the transmission line Shortest distance: ; In the formula, For power transmission lines The point on; The distance is the great circle distance; (1) Fire intensity: ; In the formula, I represents the fire intensity (kW / m); q represents the calorific value per unit mass of combustible material (kJ / kg), and its reference range is shown in Table 2; Wf represents the combustible material load (t / hm2), which is experimentally measured; R represents the flame spread rate (m / min), which is obtained by correcting the initial spread rate for wind speed, slope, and combustible material type. Table 2 (2) Flame height: ; In the formula, The height of the flame. I Fire intensity γ This represents the regional correction factor, which is usually taken as... γ =1; like Figure 3 The diagram shows flame zoning. The flame is vertically divided into three zones: a continuous flame zone, a discontinuous flame zone, and a smoke / dust zone. The lengths of the continuous and discontinuous flame zones are calculated as follows: ; ; In the formula, The length of the continuous flame zone; The length of the discontinuous flame region; (3) Calculation of each correction factor: Altitude correction factor: ,in Altitude; Vegetation density correction factor: , where Wf is the combustible load; Smoke particle correction factor: According to the experimental results, the breakdown voltage gradient in the pure smoke zone is 85% of the air gap. The smoke particle correction coefficient is further adjusted according to the smoke concentration. Vegetation type correction factor: C k The values ​​were obtained by looking up tables, for example: Metasequoia: 1; Yunnan pine: 1.13; Eucalyptus: 0.96; Shrubs: 0.93; Thatch: 0.88; Vegetation moisture content correction factor: Where WD represents the absolute water content of vegetation, obtained by looking up a table based on vegetation type and season; the correspondence between absolute water content of vegetation and vegetation type and season is shown in Table 3: Table 3 Flame temperature correction factor: Where T0 is the highest temperature at the top of the flame on the vegetation, T f The temperature of the flame at the height of the wire is calculated. ; ; in T For ambient temperature, ΔT The temperature rise relative to the ambient temperature. I Fire intensity The height of the flame. The height of the line; (4) Breakdown voltage calculation: According to the flame height With line height The relationship can be calculated in three cases: Case 1: Smoke and Dust Bridging ( The gap between the conductor and the ground is bridged sequentially by the smoke zone, the discontinuous flame zone, and the continuous flame zone. Smoke length: ; Line-to-ground breakdown voltage: ; In the formula, Ug This is the breakdown voltage of the line-to-ground gap. Hs The length of the smoke zone. HX The length of the discontinuous flame region. HF E represents the length of the continuous flame zone. a The standard air gap breakdown voltage gradient is 359.2 kV / m, C a C is the altitude correction factor. t E is the flame temperature correction factor. HX The average breakdown voltage gradient in the discontinuous region of a standard flame is 173.7 kV / m, C p C is the particle correction factor. d E is the vegetation density correction factor. HF For a standard flame continuous zone average breakdown voltage gradient of 60 kV / m, C K C is the vegetation type correction factor. w This is the correction factor for vegetation moisture content; Phase-to-phase breakdown voltage: ; In the formula U ab E is the gap breakdown voltage. a The standard air gap breakdown voltage gradient is 359.2 kV / m. C is the distance between phases. a C is the altitude correction factor. t C is the flame temperature correction factor. p This is the particle correction factor; Case 2: Bridging of flame discontinuity zone ( ); Line-to-ground breakdown voltage: ; Phase-to-phase breakdown voltage: ; Case 3: Bridging in the flame continuity zone ( ); Line-to-ground breakdown voltage: ; In the formula, E HF The average breakdown voltage gradient in the standard flame continuous zone is 60 kV / m; Phase-to-phase breakdown voltage: ; (5) Calculation of tripping probability: Risk of line breakdown : ; Alternating breakdown risk : ; In the formula, For power transmission lines Voltage level (kV); The maximum allowable offset coefficient is 1.15 for 220kV and below, and 1.1 for 330kV and above. Ultimate breakdown risk : ; (6) Calculate the alarm coefficient: Based on the risk of final breakdown From the fire point to the transmission line shortest distance Calculate transmission lines Alarm coefficient : ; (7) Output the final alarm coefficient: Traverse all line segments around the fire point, find the line with the highest alarm coefficient, and use the highest alarm coefficient as the final alarm coefficient: ; In the formula, This is the final alarm coefficient; (8) Hierarchical alarm output: The alarm level and recommended measures are output based on the final alarm coefficient; the relationship between the final alarm coefficient, alarm level, and recommended measures is shown in Table 4. Table 4 In an optional embodiment, visualization and alarm push notifications are also included: a GIS map displays fire point markers, and power transmission lines are displayed using risk level color coding, while vector arrows predicting plume movement based on wind direction are shown, and distances are presented using equidistant circles (1km, 2km, 3km).

[0059] In the above alarm output steps, the radar identification results are linked with the power grid GIS system to output line-level risk alarms; a quantitative conversion relationship between radar echo parameters and smoke concentration is established to provide input for the tripping model; quantitative assessment of power grid risks is realized to support accurate decision-making, and maintenance personnel can clearly identify line risks; the tripping probability can be quantitatively assessed to determine whether to reduce load; combined with visual push notifications, the arrival time can be further predicted to rationally allocate fire-fighting resources.

[0060] By implementing the above embodiments, accurate results of suspected fire point identification can be obtained; in addition, based on the accurate results of suspected fire point identification, accurate wildfire alarms for power transmission lines can be output, improving the ability of maintenance personnel to handle wildfires on power transmission lines.

[0061] like Figure 2 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided; An embodiment of the present invention provides a power transmission line wildfire alarm device based on dual polarization radar, including: a data acquisition module, a correlation coefficient dynamic threshold generation module, a differential reflectivity dynamic threshold generation module, a grid data acquisition module, a suspected fire point screening module, and an alarm output module; The data acquisition module is used to acquire dual-polarization radar volume scan data and real-time meteorological data of the transmission line buffer area; wherein, the real-time meteorological data includes: real-time humidity, real-time wind speed and real-time temperature; The correlation coefficient dynamic threshold generation module is used to generate a humidity correction amount based on the deviation between real-time humidity and a preset high humidity threshold and a preset humidity correction coefficient; generate a first wind speed correction amount based on the deviation between real-time wind speed and a preset high wind speed threshold and a preset first wind speed correction coefficient; and dynamically adjust the preset correlation coefficient benchmark threshold based on the humidity correction amount and the first wind speed correction amount to generate a correlation coefficient dynamic threshold. The differential reflectivity dynamic threshold generation module is used to generate a temperature correction amount based on the deviation between the real-time temperature and the preset low temperature threshold and the preset temperature correction coefficient; generate a second wind speed correction amount based on the deviation between the real-time wind speed and the preset high wind speed threshold and the preset second wind speed correction coefficient; and dynamically adjust the preset differential reflectivity benchmark threshold based on the temperature correction amount and the second wind speed correction amount to generate a differential reflectivity dynamic threshold. The grid data acquisition module is used to acquire the real-time differential reflectivity and real-time correlation coefficient of each grid point based on dual-polarization radar volume scan data. The suspected fire point screening module filters out grid points with real-time differential reflectivity not less than the differential reflectivity dynamic threshold and real-time correlation coefficient not greater than the correlation coefficient dynamic threshold as suspected fire points. The alarm output module is used to output a wildfire alarm for power transmission lines based on dual-polarization radar volume scan data of suspected fire points.

[0062] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can realize the transmission line wildfire alarm method based on dual-polarization radar provided by any of the above-described method embodiments of the present invention.

[0063] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0064] Based on the above embodiments of the transmission line wildfire alarm method based on dual-polarization radar, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the transmission line wildfire alarm method based on dual-polarization radar according to any embodiment of the present invention.

[0065] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0066] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0067] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0068] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the power transmission line wildfire alarm method based on dual-polarization radar as described in any of the above-described method embodiments of the present invention.

[0069] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0070] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for early warning of wildfires on power transmission lines based on dual-polarization radar, characterized in that, include: Acquire dual-polarization radar volume scan data and real-time meteorological data of the transmission line buffer zone; wherein, the real-time meteorological data includes: real-time humidity, real-time wind speed and real-time temperature; Humidity correction amount is generated based on the deviation between real-time humidity and preset high humidity threshold and preset humidity correction coefficient; first wind speed correction amount is generated based on the deviation between real-time wind speed and preset high wind speed threshold and preset first wind speed correction coefficient; preset correlation coefficient benchmark threshold is dynamically adjusted based on humidity correction amount and first wind speed correction amount to generate correlation coefficient dynamic threshold. A temperature correction amount is generated based on the deviation between the real-time temperature and the preset low temperature threshold and the preset temperature correction coefficient; a second wind speed correction amount is generated based on the deviation between the real-time wind speed and the preset high wind speed threshold and the preset second wind speed correction coefficient; and a preset differential reflectivity benchmark threshold is dynamically adjusted based on the temperature correction amount and the second wind speed correction amount to generate a differential reflectivity dynamic threshold. Real-time differential reflectivity and real-time correlation coefficient of each grid point are obtained based on dual-polarization radar volume scan data; Grid points with real-time differential reflectance not less than the differential reflectance dynamic threshold and real-time correlation coefficient not greater than the correlation coefficient dynamic threshold are selected as suspected fire points. Wildfire warnings for power transmission lines are output based on dual-polarization radar volume scan data of suspected fire points.

2. The method for wildfire alarm of transmission lines based on dual-polarization radar as described in claim 1, characterized in that, Acquire dual-polarization radar volume scan data of the transmission line buffer zone, including: Read the power grid geographic information system ledger data and the original dual-polarization radar volume scan data; Extract spatial vector information and voltage levels of transmission lines from the power grid geographic information system ledger data; Establish a transmission line buffer zone based on the spatial vector information and voltage level of the transmission line; Based on the filtering of the original dual-polarization radar volume scan data of the transmission line buffer area, dual-polarization radar volume scan data of the transmission line buffer area is obtained.

3. The method for wildfire alarm of transmission lines based on dual-polarization radar as described in claim 2, characterized in that, The power line wildfire alarm based on dual-polarization radar volume scan data of suspected fire points includes: Historical radar data of suspected fire points were extracted from dual-polarization radar volume scan data of suspected fire points. For each suspected fire point, the mean of differential reflectance, the standard deviation of differential reflectance, the mean of correlation coefficient, and the standard deviation of correlation coefficient are calculated based on real-time meteorological data and historical radar data. The differential reflectance deviation is calculated by standardizing the mean of differential reflectance and the standard deviation of differential reflectance using standard scores. The correlation coefficient deviation is calculated by standardizing the mean of correlation coefficient and the standard deviation of correlation coefficient using standard scores. Suspected fire points with differential reflectance deviations greater than the first preset standard deviation height and correlation coefficient deviations greater than the second preset standard deviation height are selected as significant fire points. Several plume objects are generated from dual-polarization radar volume scan data with significant fire points using a spatial clustering algorithm. Select the confirmed fire plume from a number of smoke plume objects; Extract the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center of the confirmed fire plume; Smoke concentration is calculated based on the reflectivity factor, correlation coefficient, and differential reflectivity at the fire center. Based on smoke concentration, a wildfire warning for the transmission line is output by calling the transmission line tripping probability model.

4. The method for wildfire alarm of transmission lines based on dual-polarization radar as described in claim 3, characterized in that, The dual-polarization radar volume scan data based on significant fire points is used to generate several plume objects through a spatial clustering algorithm, including: Based on dual-polarization radar volume scan data of significant fire points, significant fire points at the same latitude and longitude but different radar elevation layers are classified into a single plume to be determined. For each plume to be determined, the vertical cumulative debris index is calculated by weighted vertical integration of correlation coefficients based on the dual-polarization radar volume scan data of all significant fire points in the plume to be determined. Filter out plumes with a vertical cumulative debris index greater than a preset index threshold as real plumes; Several plume objects are generated based on dual-polarization radar volume scan data of significant fire points within each real plume using a spatial clustering algorithm.

5. The method for wildfire alarm of transmission lines based on dual-polarization radar as described in claim 4, characterized in that, The process of filtering out confirmed fire plumes from a plurality of smoke plume objects includes: Based on the dual-polarization radar volume scan data of each grid point within the plume object, calculate the average Doppler spectral width of all grid points within each plume object; Select smoke plumes with an average Doppler width not less than a preset Doppler width threshold as confirmed plumes.

6. The method for wildfire alarm of transmission lines based on dual-polarization radar as described in claim 5, characterized in that, Extract the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center of the confirmed fire plume, including: Based on the dual-polarization radar volume scan data of each grid point within the confirmed plume, the average reflectivity factor of each confirmed plume is calculated. For each confirmed fire plume, extract the maximum reflectance factor from the reflectance factors of each grid point; calculate the maximum deviation of the reflectance factor based on the maximum reflectance factor and the average reflectance factor. If the maximum deviation of the reflectivity factor is greater than the preset maximum deviation threshold, the combustion mode of the confirmed fire plume is determined to be the concentrated combustion mode; otherwise, the combustion mode of the confirmed fire plume is determined to be the diffusion combustion mode. If the combustion mode of the fire plume is confirmed to be a concentrated combustion mode, the grid point corresponding to the maximum reflectivity factor in the confirmed fire plume is taken as the fire point center, and the reflectivity factor, correlation coefficient and differential reflectivity of the fire point center are extracted based on the dual polarization radar volume scan data at the fire point center. If the combustion mode of the confirmed fire plume is determined to be diffusion combustion mode, the weighted centroid of the confirmed fire plume is calculated based on the reflectivity factor of each grid point and the position of each grid point, and the weighted centroid is taken as the fire point center; the weighted average of the reflectivity factor, the weighted average of the correlation coefficient, and the weighted average of the differential reflectivity of all grid points in the confirmed fire plume are taken as the reflectivity factor, correlation coefficient, and differential reflectivity of the fire point center, respectively.

7. The method for wildfire alarm of transmission lines based on dual-polarization radar as described in claim 6, characterized in that, The calculation of smoke concentration based on the reflectivity factor, correlation coefficient, and differential reflectivity at the fire point center includes: The reflectivity factor weight term is obtained by linearly restoring the reflectivity factor at the center of the fire point. A correlation coefficient decay term is generated based on the deviation between the correlation coefficient at the fire center and the perfect correlation. After linear scaling of the differential reflectivity at the fire point center, the correction reference is adjusted based on the result of the linear scaling to obtain the differential reflectivity correction term. The smoke concentration is obtained by adjusting the smoke concentration calibration coefficient together with the reflectivity factor weight term, the correlation coefficient attenuation term, and the differential reflectivity correction term; wherein, the smoke concentration calibration coefficient is obtained by regression fitting calculation through a linear regression model established by synchronously collecting ground smoke concentration and radar echo parameters at a known wildfire site.

8. A power transmission line wildfire alarm device based on dual-polarization radar, characterized in that, include: The module includes a data acquisition module, a correlation coefficient dynamic threshold generation module, a differential reflectivity dynamic threshold generation module, a grid data acquisition module, a suspected fire point screening module, and an alarm output module. The data acquisition module is used to acquire dual-polarization radar volume scan data and real-time meteorological data of the transmission line buffer area; wherein, the real-time meteorological data includes: real-time humidity, real-time wind speed and real-time temperature; The correlation coefficient dynamic threshold generation module is used to generate a humidity correction amount based on the deviation between real-time humidity and a preset high humidity threshold and a preset humidity correction coefficient; generate a first wind speed correction amount based on the deviation between real-time wind speed and a preset high wind speed threshold and a preset first wind speed correction coefficient; and dynamically adjust the preset correlation coefficient benchmark threshold based on the humidity correction amount and the first wind speed correction amount to generate a correlation coefficient dynamic threshold. The differential reflectivity dynamic threshold generation module is used to generate a temperature correction amount based on the deviation between the real-time temperature and the preset low temperature threshold and the preset temperature correction coefficient; generate a second wind speed correction amount based on the deviation between the real-time wind speed and the preset high wind speed threshold and the preset second wind speed correction coefficient; and dynamically adjust the preset differential reflectivity benchmark threshold based on the temperature correction amount and the second wind speed correction amount to generate a differential reflectivity dynamic threshold. The grid data acquisition module is used to acquire the real-time differential reflectivity and real-time correlation coefficient of each grid point based on dual-polarization radar volume scan data. The suspected fire point screening module filters out grid points with real-time differential reflectivity not less than the differential reflectivity dynamic threshold and real-time correlation coefficient not greater than the correlation coefficient dynamic threshold as suspected fire points. The alarm output module is used to output a wildfire alarm for power transmission lines based on dual-polarization radar volume scan data of suspected fire points.

9. A terminal device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the power transmission line wildfire alarm method based on dual-polarization radar as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the power transmission line wildfire alarm method based on dual-polarization radar as described in any one of claims 1-7.

Images