A transmission line wildfire risk warning method considering human fire behavior

Abstract

The present application relates to a kind of power transmission line forest fire risk early warning method considering human use fire behavior, comprising the following steps: the early warning area is divided into multiple grids according to preset size;The forest fire weather risk grade, the forest fire combustible risk grade and the ignition point risk grade of each grid are calculated respectively;For any tower in power transmission line, the forest fire weather risk grade, the forest fire combustible risk grade and the ignition point risk grade of all grids in its fixed range are integrated, the forest fire risk grade of the tower is calculated, and the risk size of the tower fixed range occurrence forest fire is evaluated by forest fire risk grade.

Description

TECHNICAL FIELD

[0001] The present application relates to a power transmission line forest fire risk early warning method considering human fire behavior, and belongs to the technical field of forest fire monitoring. BACKGROUND

[0002] With the expansion of the scale of power transmission lines year by year and the increasingly severe situation of power corridor construction land, many high-voltage power transmission line constructions in China choose to cross the mountainous areas with rich vegetation, dense jungle and few people. This area is prone to forest fire disasters, and if the forest fire around the power transmission line spreads to the underlying surface of the power transmission line, it can cause damage to the line equipment and lead to line trip failure. Therefore, if the forest fire can be timely solved and handled when it occurs but has not spread widely, the probability of line failure can be reduced, which is beneficial to the safe and stable operation of the power grid.

[0003] For a long time, the research on forest fires has mainly focused on physical models, and by exploring the mechanism of forest fire formation and analyzing the physical process of forest fire disasters, a number of physical equations are derived to describe the impact of forest fires. However, the physical process of power transmission line forest fire disaster is complex, and the mechanism and parameters of many processes still have a lot of knowledge gaps, and the physical model has problems such as high complexity, many assumptions, and inability to predict future forest fires. In recent years, dynamic forest fire risk early warning and evaluation methods have been proposed, which consider various forest fire influencing factors and conduct early warning work on the forest fire risk of the study area based on the mechanism of forest fire formation and development. Compared with the physical model, it is more reasonable and efficient.

[0004] Patent No. "CN104268655A" discloses a power transmission line forest fire early warning method, which calculates the current period power transmission line forest fire early warning level according to the target area historical satellite monitoring fire point data, precipitation data, and industrial and agricultural fire customs, and makes a warning table according to the warning level. The line operation and maintenance personnel can reasonably arrange the anti-forest fire resources according to the early warning level provided by the present application method, effectively improve the efficiency and resource utilization rate of forest fire prevention, reduce the waste of manpower and material resources, and reduce the power transmission line forest fire trip rate, improve the safety and stability performance of the power system.

[0005] However, the above-mentioned scheme does not consider the data dimension comprehensively when performing forest fire early warning, which may lead to a prediction result that does not conform to the actual situation, and may cause false positives, resulting in waste of manpower and material resources. SUMMARY

[0006] In order to solve the problems existing in the prior art, the present application provides a power transmission line forest fire risk early warning method considering human fire behavior.

[0007] The technical scheme of the present application is as follows:

[0008] The application provides a power transmission line forest fire risk early warning method considering human fire behavior, comprising the following steps:

[0009] The early warning area is divided into a plurality of grids according to a preset size;

[0010] Meteorological conditions of each grid are collected, and the meteorological risk grade of each grid is determined based on the meteorological conditions of each grid;

[0011] The combustible content in each grid is collected, and the combustible risk grade of each grid is determined based on the combustible content in each grid;

[0012] Historical human fire records in the early warning area are collected, cluster analysis is performed on the historical human fire records, the early warning area is divided into multiple human fire zones according to the cluster results, the average human fire and the human fire out-of-control rate of each human fire zone are calculated, and the fire source risk grade of each grid is determined based on the average human fire and the human fire out-of-control rate of each human fire zone;

[0013] For any tower in the power transmission line, the meteorological risk grade, the combustible risk grade and the fire source risk grade of all grids in the fixed range of the tower are integrated to calculate the forest fire risk grade of the tower, and the risk of forest fire in the fixed range of the tower is evaluated through the forest fire risk grade.

[0014] As a preferred embodiment of the application, the meteorological conditions of each grid include:

[0015] The wind speed at 14 o'clock of the next day, the air temperature at 14 o'clock of the next day, the relative humidity at 14 o'clock of the next day, the precipitation of the next day, the snow depth of the next day and the number of consecutive days without precipitation until the next day.

[0016] As a preferred embodiment of the application, the grid forest fire meteorological risk grade determination step is:

[0017] A forest fire risk meteorological index is constructed, as shown in the following formula:

[0018]

[0019] wherein I v represents the wind speed index; I T represents the temperature index; represents the relative humidity index; I M represents the consecutive days without precipitation index; λ1 is the wind speed weight coefficient; λ2 is the temperature weight coefficient; λ3 is the relative humidity weight coefficient; λ4 is the consecutive days without precipitation weight coefficient; Cr represents the precipitation correction coefficient; Cs represents the snow correction coefficient;

[0020] Multiple adjacent wildfire risk meteorological index value intervals are preset in order from small to large, each value interval represents a wildfire risk meteorological level, and the wildfire risk meteorological level of the grid is determined based on the value interval in which the wildfire risk meteorological index of the grid falls.

[0021] As a preferred embodiment of the present invention, the steps for determining the wildfire combustible material risk level of the grid are as follows:

[0022] Analyze and collect the combustible load of grids where fires have occurred in the past and the combustible load of other grids in the warning area;

[0023] Based on the combustible load, all grids with historical fires are sorted from large to small to obtain a combustible load set. The combustible load set is then divided into multiple subsets. The minimum value in each subset is taken out, and multiple intervals are constructed in order of size. Each interval represents a wildfire combustible risk level. The wildfire combustible risk level of the grid is determined by the position of the grid's combustible load in the interval.

[0024] As a preferred embodiment of the present invention, the historical records of man-made fire use are specifically the overall data statistics of fire point investigations of historical wildfire alarms in various sections of the transmission line, including statistics of cases where fire points were generated but no fire occurred and statistics of cases where fire occurred.

[0025] As a preferred embodiment of the present invention, the cluster analysis of historical man-made fire records is specifically based on the historical fire point latitude and longitude data in the historical man-made fire records, and the historical man-made fire records are clustered using the K-means clustering algorithm. The specific steps are:

[0026] Construct the fire point latitude and longitude data sample set X = {X1, X2, X3, ..., X n}, where n represents the total number of grids. The fire point longitude and latitude data sample set is divided into K clusters, and K points C = {C1, C2, C3, ..., C K} as the cluster center of each cluster;

[0027] Calculate the distance between each sample in the fire point latitude and longitude data sample set X and each cluster center, and assign the sample to the cluster with the closest cluster center. The distance calculation formula between the sample and the cluster center is as follows:

[0028]

[0029] Where: dis(X i , C j ) represents the sample X i To cluster center C j Example; Xit represents the tth attribute of the ith sample; C jt represents the tth attribute of the jth cluster center; e represents the total number of attributes;

[0030] After the classification of the samples is completed, the new cluster classes are denoted as Q = {Q1, Q2, Q3,..., Q K} and the average value of the samples in each cluster is calculated as a new cluster center, which is specifically shown in the following formula:

[0031]

[0032] wherein C l represents the cluster center of the lth new cluster class; |Q l | represents the total number of samples in the lth new cluster class; Y i represents the ith sample in the lth new cluster class;

[0033] The error sum of squares of all samples is calculated, which is specifically shown in the following formula:

[0034]

[0035] wherein E l represents the error sum of squares of all samples; u l represents the average value of all samples in the cluster class Q l .

[0036] The distance of each sample to each cluster center is recalculated, the sample is assigned to the cluster in which the cluster center with the closest distance is located, and then the error of the error sum of squares of all samples in the current iteration and the error sum of squares of all samples in the last iteration is calculated. If the error is less than a preset threshold, the clustering is completed to obtain a clustering result. If the error is greater than the preset threshold, the above steps are repeated for iteration until the error of the error sum of squares is less than the threshold.

[0037] As a preferred embodiment of the present application, the specific steps of dividing the pre-warning area into multiple fire use areas according to the clustering result are as follows:

[0038] For each cluster class, the minimum area region composed of the grids corresponding to the longitude and latitude positions in the pre-warning area is divided into a fire use area according to the longitude and latitude data of the fire points of all samples in the cluster class.

[0039] After the division of the fire use areas corresponding to all cluster classes, the remaining grids in the pre-warning area are divided into a fire use area, and finally K+1 fire use areas are obtained.

[0040] As a preferred embodiment of the present application, the calculation steps of the uncontrolled fire mean value and the fire uncontrolled rate of each fire use area are as follows:

[0041] Collect the number of fire point monitoring records N in each type of fire area f , Number of fire records N w ;

[0042] The calculation formula for the mean value FA of uncontrolled fire in the fire zone is:

[0043]

[0044] Where: S i is the area of ​​the i-th type fire zone;

[0045] The calculation formula for the fire uncontrollable rate FK in the fire zone is:

[0046] As a preferred embodiment of the present invention, the steps for determining the risk level of the fire source point of the grid are:

[0047] According to the average value of uncontrolled fire use and the rate of uncontrolled fire use, each type of fire area is sorted from large to small, with the average value of uncontrolled fire use as the priority indicator when sorting. If the average value of uncontrolled fire use in two types of fire areas is equal, the uncontrolled fire rates of the two types of fire areas are compared. After the sorting is completed, each type of fire area represents a fire source risk level, and the fire source risk level of all grids within this type of fire area is equal to that of this type of fire area.

[0048] As a preferred embodiment of the present invention, the steps for determining the mountain fire risk level of the tower are:

[0049] The highest wildfire meteorological risk level, the highest wildfire combustible risk level, and the highest fire source risk level in all grids within the fixed range of the tower are extracted, and the minimum level is selected as the wildfire risk level of the tower.

[0050] The present invention has the following beneficial effects:

[0051] 1. The present invention considers the differences in human fire behavior in different regions and makes full use of historical transmission line fire data to effectively predict the possibility of wildfire disasters caused by unplanned fire. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] It should be understood that the step numbers used herein are only for the convenience of description and are not limited to the execution sequence of the steps.

[0055] It should be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and the appended claims of the present application, "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0056] The terms "comprise" and "include" indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0057] The term "and / or" means any combination of one or more of the associated listed items and all possible combinations thereof, and includes these combinations.

[0058] Embodiment:

[0059] Referring to Figure 1 A power transmission line forest fire risk early warning method considering human use of fire behavior, comprising the following steps:

[0060] The early warning area is divided into a plurality of grids according to a size of 2.5km x 2.5km; the grid division is to divide the study area into a plurality of square grids with the same size and a side length of 2.5km, and because the longitudinal and latitudinal span of the study area is small, the graph formed by the longitudinal and latitudinal ranges (x1, x2), (y1, y2) can be approximated as a rectangle, and the rectangle is divided into a plurality of grids from left to right and from top to bottom, and the grids outside the range of the study area are cut off, thereby completing the grid division work;

[0061] Collect the meteorological conditions of each grid, and determine the forest fire meteorological risk level of each grid based on the meteorological conditions of each grid;

[0062] Collect the combustible content in each grid, and determine the forest fire combustible risk level of each grid based on the combustible content in each grid;

[0063] Collect the historical human use of fire records in the early warning area, perform cluster analysis on the historical human use of fire records, divide the early warning area into multiple use of fire areas according to the cluster results, calculate the average uncontrollable use of fire and the use of fire uncontrollable rate of each use of fire area, and determine the fire source risk level of each grid based on the average use of fire and the use of fire uncontrollable rate of each use of fire area;

[0064] For any tower in the transmission line, the fire risk level of the tower is calculated by integrating the fire weather risk level, the fire fuel risk level and the fire source risk level of all the grids within 3km range of the tower, and the risk of fire in the fixed range of the tower is evaluated by the fire risk level.

[0065] As a preferred embodiment of the present embodiment, the meteorological condition of each grid comprises:

[0066] the wind speed V at 14:00 of the next day, the air temperature T at 14:00 of the next day, the relative humidity r at 14:00 of the next day RH , the precipitation R from 0:00 to 24:00 of the next day r , the snow depth H from 0:00 to 24:00 of the next day s and the consecutive non-precipitation days M until the next day, where the non-precipitation day refers to the precipitation R from 0:00 to 24:00 of the next day r less than 1mm, that is, if the next day is forecasted to have rain and the forecasted precipitation R from 0:00 to 24:00 of the next day r is less than 1mm, the consecutive non-precipitation days is recorded as 0.

[0067] As a preferred embodiment of the present embodiment, the fire weather risk level judgment step of the grid comprises:

[0068] A fire risk weather index is constructed, as shown in the following formula:

[0069]

[0070] wherein I V represents the wind speed index; I T represents the temperature index; represents the relative humidity index; I M represents the consecutive non-precipitation days index; λ1 is the wind speed weight coefficient; λ2 is the temperature weight coefficient; λ3 is the relative humidity weight coefficient; λ4 is the consecutive non-precipitation days weight coefficient; Cr represents the precipitation correction coefficient; Cs represents the snow correction coefficient;

[0071] The numerical classification of each meteorological factor and the corresponding index value are shown in the following table:

[0072]

[0073] The value rules of the precipitation correction coefficient Cr and the snow correction coefficient Cs are as follows:

[0074] when the 24h precipitation R r is greater than or equal to 1mm, Cr=0; when the 24h precipitation R r is less than 1mm, Cr=1;

[0075] when the 24h snow depth H sCs = 0 when H < 0 cm; Cs = 1 when H ≥ 0 cm s Cs = 0 when H < 0 cm; Cs = 1 when H ≥ 0 cm

[0076] The multiple adjacent fire risk meteorological index value intervals are preset in ascending order, each value interval representing a fire risk meteorological grade, and the fire risk meteorological grade of the grid is determined according to the value interval in which the fire risk meteorological index of the grid falls;

[0077] Specifically, in this embodiment, the fire risk meteorological grade RM of the grid is divided into five grades of 0, I, II, III, and IV, and is updated once a day, as shown in the following table:

[0078] I FRM ]] RM Risk rating definition [0，40) 0 Low fire risk [40，50) I Lower fire risk [50，60) II Higher fire risk, increased prevention required [60，75) III High fire risk, increased fire source management required in forest areas [75，100) IV Very high fire risk, all use of fire in forest strictly prohibited

[0079] As a preferred embodiment of this embodiment, the fire risk fuel grade determination step of the grid is:

[0080] The fuel load of the grid in which a fire has occurred in the past in the warning area and the fuel load of other grids are analyzed and collected;

[0081] The fuel load of all grids in which a fire has occurred in the past is sorted in descending order based on the fuel load to obtain a fuel load set, the fuel load set is then divided into multiple subsets, the minimum value in each subset is taken out, and multiple intervals are constructed in size order, each interval representing a fire risk fuel grade, and the fire risk fuel grade of the grid is determined by the position of the fuel load of the grid in the interval.

[0082] In this embodiment, the fuel load data since 2020 in the warning area and the fuel load records of all fire occurrence cases in recent years are obtained, and the fuel risk grade RC of the grid is divided into five grades of 0, I, II, III, and IV according to the following threshold standards, and is updated once a month:

[0083] The fuel load of all grids in which a fire has occurred in the fire occurrence case is sorted in descending order, the minimum load value of the fuel load ranked in the top 20% of all records is recorded as C4, and when the fuel load C of a grid satisfies C ≥ C4, the fire risk fuel grade RC of the grid is rated as IV.

[0084] The minimum load value of the fuel load ranked in the top 40% of all records is recorded as C3, and when the fuel load C of a grid satisfies C3 ≤ C < C4, the fire risk fuel grade RC of the grid is rated as III.

[0085] The minimum combustible material load value in the top 70% of all records is recorded as C2. When the combustible material load C of a grid satisfies C2≤C<C3, the wildfire combustible material risk level RC of the grid is assessed as Level II.

[0086] The minimum combustible material load value that ranks in the top 100% of all records is recorded as C1. When the combustible material load C of a grid satisfies C1≤C<C2, the wildfire combustible material risk level RC of the grid is assessed as level I.

[0087] When the combustible material load C of a grid satisfies C<C1, the wildfire combustible material risk level RC of the grid is assessed as level 0;

[0088] The overall standards are shown in the following table:

[0089]

[0090] As a preferred implementation of this embodiment, the historical records of man-made fire use are specifically the overall data statistics of fire point investigations of historical wildfire alarms in various sections of the transmission line, including statistics of cases where fire points were generated but no fire occurred and statistics of cases where fire occurred.

[0091] As a preferred implementation of this embodiment, the cluster analysis of historical man-made fire records is specifically performed based on the historical fire point latitude and longitude data in the historical man-made fire records, using the K-means clustering algorithm to cluster the historical man-made fire records. The specific steps are:

[0092] (1) Construct a sample set of fire point latitude and longitude data X = {X1, X2, X3, ..., X n}, where n represents the total number of grids. The fire point longitude and latitude data sample set is divided into K clusters, and K points C = {C1, C2, C3, ..., C K} as the cluster center of each cluster;

[0093] (2) Calculate the distance between each sample in the fire point latitude and longitude data sample set X and each cluster center, and assign the sample to the cluster with the closest cluster center. The distance calculation formula between the sample and the cluster center is as follows:

[0094]

[0095] Where: dis(X i , C j ) represents the sample X i To cluster center C j Example; X it represents the tth attribute of the i-th sample; C jtrepresents the jth attribute of the tth cluster center; e represents the total number of attributes;

[0096] (3) The new cluster class after the sample classification is denoted as Q = {Q1, Q2, Q3,..., QK}, and the average value of the samples in each cluster is calculated as the new cluster center, which is specifically shown in the following formula: K}, and the average value of the samples in each cluster is calculated as the new cluster center, which is specifically shown in the following formula:

[0097]

[0098] wherein: C l represents the cluster center of the lth new cluster class; |Q l | represents the total number of samples in the lth new cluster class; Y i represents the ith sample in the lth new cluster class;

[0099] (4) The error sum of squares of all samples is calculated, which is specifically shown in the following formula:

[0100]

[0101] wherein: E l represents the error sum of squares of all samples; u l represents the average value of all samples in the cluster class Q l

[0102] The distance of each sample to each cluster center is recalculated, and the sample is assigned to the cluster in which the nearest cluster center is located. After the cluster center of each new cluster class is calculated, the error E l -E l+1 of the error sum of squares of all samples in the current iteration and the error sum of squares of all samples in the last iteration is calculated. If the error is less than a preset threshold δ, the clustering is completed to obtain the clustering result. If the error is greater than the preset threshold δ, the steps (2)-(4) are repeated for iteration until the error of the error sum of squares is less than the threshold.

[0103] As a preferred embodiment of the present embodiment, the specific steps of dividing the pre-warning area into multiple fire use areas according to the clustering result are as follows:

[0104] For each cluster class, the minimum area region composed of the grid corresponding to the latitude and longitude position in the pre-warning area is divided into a fire use area according to the latitude and longitude data of the fire points of all samples in the cluster class, and the fire use area is extended outward by 2 km;

[0105] After the fire use areas corresponding to all cluster classes are divided, the remaining grid in the pre-warning area is divided into a fire use area, and finally K+1 fire use areas are obtained.

[0106] As a preferred embodiment of the present embodiment, the calculation steps of the uncontrolled fire mean value and the fire use uncontrolled rate of each fire use area are as follows: ​

[0107] Collecting the number of fire point monitoring records N in each type of fire area f , the number of fire records N w ;

[0108] The calculation formula of the average uncontrolled fire FA of the fire area is:

[0109]

[0110] Wherein: S i is the area of the i-th type of fire area;

[0111] The calculation formula of the fire control rate FK of the fire area is:

[0112] As a preferred embodiment of the present embodiment, the fire source point risk level determination step of the grid is:

[0113] According to the average uncontrolled fire and the fire control rate of each type of fire area, the average uncontrolled fire is sorted in descending order, and the fire control rate is sorted in descending order. When sorting, the average uncontrolled fire is used as the priority index, and if the average uncontrolled fire of the two types of fire area is equal, the fire control rate of the two types of fire area is compared, and after sorting, each type of fire area represents a fire source point risk level, and the fire source point risk level of all grids in the type of fire area is equal to the type of fire area. The ignition source risk level RI of the grid is evaluated according to the following rules, and it is updated once a year:

[0114] (1) If K = 4, the ignition source risk level RI of each type of grid is divided into five levels of IV, III, II, I and 0 according to the sorting results of the average uncontrolled fire and the fire control rate;

[0115] (2) If K < 4, the ignition source risk level RI of each type of grid is divided into four levels of IV, III, II, I or three levels of IV, III, II or two levels of IV, III according to the sorting results of the average uncontrolled fire and the fire control rate;

[0116] (3) If K > 4, the 5th to Kth fire area is classified as one fire area, and the ignition source risk level RI of each type of grid is divided into five levels of IV, III, II, I and 0 according to the sorting results of the average uncontrolled fire and the fire control rate.

[0117] As a preferred embodiment of the present embodiment, the mountain fire risk level determination step of the tower pole is:

[0118] The highest wildfire weather risk level, the highest wildfire fuel risk level and the highest fire source point risk level in all grids in the extraction tower pole fixed range (specifically, a circle with a radius of 3 km and the tower pole as the center as the research range) are extracted, and the minimum level among them is selected as the wildfire risk level of the tower pole, that is, the wildfire risk level RF of the tower pole is divided into 0, I, II, III, IV five levels, and is updated once a day.

[0119] In the embodiments of the present application, "at least one" refers to one or more, and "multiple" refers to two or more. The "and / or" describes the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Wherein A, B can be singular or plural. The character " / " generally represents that the associated objects before and after are a kind of "or" relationship. "At least one of the following" and the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b and c can represent: a, b, c, a and b, a and c, b and c or a and b and c, wherein a, b, c can be single or multiple.

[0120] Those of ordinary skill in the art can realize that the units and algorithm steps described in the embodiments disclosed herein can be realized by electronic hardware, computer software and a combination of electronic hardware and computer software. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0121] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0122] In several embodiments provided in the present application, any function, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0123] The above description is only some embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for early warning of transmission line wildfire risk taking into account human fire behavior, characterized in that: The following steps are involved: Divide the warning area into multiple grids according to preset sizes; Collect the meteorological conditions of each grid and determine the wildfire meteorological risk level of each grid based on the meteorological conditions of each grid; Collect the combustible content in each grid and determine the wildfire combustible risk level of each grid based on the combustible content in each grid; Collect historical records of man-made fires in the warning area, perform cluster analysis on these records, and divide the warning area into multiple fire zones based on the clustering results. Calculate the mean uncontrolled fire rate and uncontrolled fire rate for each type of fire zone. Determine the fire source risk level for each grid based on the mean fire rate and uncontrolled fire rate for each type of fire zone. For any tower on a transmission line, the wildfire risk level of the tower is calculated by combining the wildfire meteorological risk level, wildfire combustible risk level, and fire source risk level of all grids within its fixed range. The wildfire risk level is then used to assess the risk of wildfire within the fixed range of the tower. The calculation steps for the mean value of uncontrolled fire use and the rate of uncontrolled fire use in each type of fire use area are as follows: Collect the number of fire point monitoring records N in each type of fire area f , Number of fire records N w ; The calculation formula for the mean value FA of uncontrolled fire in the fire zone is: Where: S i is the area of ​​the i-th type fire zone; The calculation formula for the fire uncontrollable rate FK in the fire zone is: The steps for determining the risk level of the fire source point in the grid are as follows: According to the average value of uncontrolled fire use and the rate of uncontrolled fire use of each type of fire area, the fire area is sorted from large to small. The average value of uncontrolled fire use is used as the priority indicator when sorting. If the average value of uncontrolled fire use of two types of fire areas is equal, the uncontrolled fire rates of the two types of fire areas are compared. After the sorting is completed, each type of fire area represents a fire source risk level, and the fire source risk level of all grids within this type of fire area is equal to that of this type of fire area. The steps for determining the mountain fire risk level of the tower are as follows: The highest wildfire meteorological risk level, the highest wildfire combustible risk level, and the highest fire source risk level in all grids within the fixed range of the tower are extracted, and the minimum level is selected as the wildfire risk level of the tower.

2. The method for early warning of transmission line wildfire risk considering human fire behavior according to claim 1, characterized in that: The meteorological conditions of each grid include: Wind speed at 2:00 PM on the next day, temperature at 2:00 PM on the next day, relative humidity at 2:00 PM on the next day, precipitation on the next day, snow depth on the next day, and the number of consecutive days without precipitation up to the next day.

3. The method for early warning of power line wildfire risk considering human fire behavior according to claim 2, characterized in that: The steps for determining the wildfire meteorological risk level of the grid are as follows: Construct a wildfire risk meteorological index as shown below: in: Indicates the wildfire risk meteorological index; I V Indicates wind speed index; I T represents the temperature index; Represents relative humidity index; I M represents the index of consecutive days without precipitation; λ1 is the wind speed weight coefficient; λ2 is the temperature weight coefficient; λ3 is the relative humidity weight coefficient; λ4 is the weight coefficient of consecutive days without precipitation; Cr represents the precipitation correction coefficient; Cs represents the snowfall correction coefficient; Multiple adjacent wildfire risk meteorological index value intervals are preset in order from small to large, each value interval represents a wildfire risk meteorological level, and the wildfire risk meteorological level of the grid is determined based on the value interval in which the wildfire risk meteorological index of the grid falls.

4. The method for early warning of power line wildfire risk considering human fire behavior according to claim 1, characterized in that: The steps for determining the wildfire combustible risk level of the grid are as follows: Analyze and collect the combustible load of grids where fires have occurred in the past and the combustible load of other grids in the warning area; Based on the combustible load, all grids with historical fires are sorted from large to small to obtain a combustible load set. The combustible load set is then divided into multiple subsets. The minimum value in each subset is taken out, and multiple intervals are constructed in order of size. Each interval represents a wildfire combustible risk level. The wildfire combustible risk level of the grid is determined by the position of the grid's combustible load in the interval.

5. The method for early warning of power transmission line wildfire risk considering human fire behavior according to claim 1, characterized in that: The historical records of man-made fire use are specifically the overall data statistics of fire point investigations of historical wildfire alarms in various sections of the transmission line, including statistics of cases where fire points were generated but no fire occurred and statistics of cases where fire occurred.

6. The method for early warning of power line wildfire risk considering human fire behavior according to claim 1, characterized in that: The cluster analysis of historical fire records is specifically performed based on the longitude and latitude data of historical fire points in the historical fire records, using the K-means clustering algorithm to cluster the historical fire records. The specific steps are as follows: Construct the fire point latitude and longitude data sample set X={X1,X2,X3,…,X n }, where n represents the total number of grids. The fire point longitude and latitude data sample set is divided into K clusters, and K points C = {C1, C2, C3, ..., C K } as the cluster center of each cluster; Calculate the distance between each sample in the fire point latitude and longitude data sample set X and each cluster center, and assign the sample to the cluster with the closest cluster center. The distance calculation formula between the sample and the cluster center is as follows: Where: dis(X i ,C j ) represents the sample X i To cluster center C j Example; X it represents the tth attribute of the i-th sample; C jt represents the tth attribute of the jth cluster center; e represents the total number of attributes; After the sample classification is completed, the new cluster is recorded as Q = {Q1, Q2, Q3, ..., Q K }, calculate the sample average value in each cluster as the new cluster center, as shown in the following formula: Where: C l represents the cluster center of the lth new cluster; |Q l | represents the total number of samples of the lth new cluster; Y i Represents the i-th sample in the l-th new cluster; Calculate the sum of squared errors of all samples, as shown in the following formula: Where: E l represents the sum of squared errors of all samples; u l Represents cluster Q l The average value of all samples in ; Recalculate the distance between each sample and each cluster center, and assign the sample to the cluster with the nearest cluster center. Then calculate the cluster center of each new cluster and calculate the difference between the sum of square errors of all samples and the sum of square errors of all samples in the previous iteration. If the error is less than the preset threshold, clustering is completed and the clustering result is obtained. If the error is greater than the preset threshold, repeat the above steps and iterate until the sum of square errors is less than the threshold.

7. The method for early warning of power transmission line wildfire risk considering human fire behavior according to claim 6, characterized in that: The specific steps of dividing the warning area into multiple types of fire zones according to the clustering results are: For each cluster, based on the longitude and latitude data of the fire points of all samples in it, the smallest area composed of grids corresponding to the longitude and latitude positions in the warning area is divided into a type of fire zone; After the fire zones corresponding to all clusters are divided, the remaining grids in the warning area are divided into one type of fire zones, and finally K+1 types of fire zones are obtained.

Images