Wildfire spreading prediction method, device and equipment and storage medium

Abstract

A forest fire spreading prediction method, device, equipment and storage medium are disclosed, and belong to the technical field of forest fire monitoring. The method comprises the following steps: acquiring an environment feature set of a first region based on a remote sensing image of the first region, wherein the environment feature set comprises elevation data, slope, slope direction, vegetation type, combustible water content, wind speed, wind direction angle, temperature and relative humidity; acquiring a fire source feature vector of a forest fire in the first region based on the environment feature set, wherein the fire source feature vector comprises a first fire intensity at a fire starting time; constructing a forest fire spreading state vector based on the environment feature vector and the fire source feature vector, wherein the forest fire spreading state vector comprises a fire line coordinate set, a fire line spreading speed and a fire line spreading direction; and updating the fire line coordinate set in the forest fire spreading state vector based on the fire line spreading speed and the fire line spreading direction, so as to predict the forest fire in the first region. The method can accurately predict the trend of forest fire spreading.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of forest fire monitoring, and in particular to a forest fire spreading prediction method, device, equipment and storage medium. BACKGROUND

[0002] Forest fire is one of the natural disasters that seriously threatens the forest ecosystem and the safety of human life and property. Accurate prediction of forest fire spreading trend is of great significance for fire prevention and control and emergency rescue.

[0003] Traditional forest fire spreading prediction methods mainly rely on a single data source and a simple model, and it is difficult to accurately reflect the interaction relationship between forest fire and environment. SUMMARY

[0004] The present disclosure provides a forest fire spreading prediction method, device, equipment and storage medium, which can accurately predict the trend of forest fire spreading. The technical solution at least includes the following schemes:

[0005] In a first aspect, a forest fire spreading prediction method is provided, comprising: based on a remote sensing image of a first region, obtaining an environment feature set of the first region, the first region being a region where forest fire exists, and the environment feature set including elevation data, slope, slope direction, vegetation type, combustible water content, wind speed, wind direction angle, temperature, and relative humidity; based on the environment feature set, obtaining a fire source feature vector of the forest fire in the first region, the fire source feature vector including a first fire intensity at a fire starting time; based on the environment feature vector and the fire source feature vector, constructing a forest fire spreading state vector, the forest fire spreading state vector including a fire line coordinate set, a fire line spreading speed, and a fire line spreading direction, the fire line spreading speed being determined based on the first fire intensity; and based on the fire line spreading speed and the fire line spreading direction, updating the fire line coordinate set in the forest fire spreading state vector to predict the forest fire in the first region.

[0006] Optionally, the forest fire spreading state vector is represented by the following formula:

[0007]

[0008] wherein, is the forest fire spreading state vector at the time t, is the fire line coordinate set at the time t, is the fire line coordinate set at the time t, is the fire line coordinate set at the time t, is the fire line spreading speed of each fire line coordinate in the fire line coordinate set at the time t, is the fire line coordinate set at the time t, is the fire line spreading speed of each fire line coordinate in the fire line coordinate set at the time t, is the fire line coordinate set at the time t, is the fire line coordinate set at the time t, a fireline spread direction at each fireline coordinate.

[0009] Optionally, represents a set of fireline coordinates at a time instant any one of the fireline coordinates ,

[0010] a fireline spread direction at a fireline coordinate a fireline spread direction at a fireline coordinate is calculated using the following equation:

[0011]

[0012] a fireline spread direction at a fireline coordinate a fireline spread direction at a fireline coordinate is calculated using the following equation:

[0013]

[0014] wherein, is a fireline spread speed at a fireline coordinate at a time instant, is the slope at a fireline coordinate at a time instant, is the wind speed at a fireline coordinate at a time instant, is the wind direction at a fireline coordinate at a time instant, is the aspect at a fireline coordinate at a time instant, is a normalized moisture index at a fireline coordinate at a time instant, is the relative humidity at a fireline coordinate at a time instant, is the fuel moisture content at a fireline coordinate at a time instant, , , , , , , is a weight.

[0015] Optionally, the updating the set of fire line coordinates in the wildfire spread state vector comprises:

[0016] The updating the set of fire line coordinates in the wildfire spread state vector is implemented by using the following formula:

[0017] wherein, is the set of fire line coordinates in the updated wildfire spread state vector, represents the set of fire line coordinates at the time t, is the length of a time step, represents the time difference between the time t and the time t-1.

[0018] Optionally, the method further comprises: generating a wildfire report based on the updated set of fire line coordinates, the wildfire report further comprising a real-time fire intensity of the wildfire, the real-time fire intensity being calculated based on the set of environmental parameters.

[0019] Optionally, the method further comprises: correcting the updated set of fire line coordinates and the real-time fire intensity in the wildfire report.

[0020] In a second aspect, a wildfire spread prediction device is also provided, comprising: a first acquisition module configured to acquire a set of environmental features of a first region based on a remote sensing image of the first region, the first region being a region where a wildfire exists, the set of environmental features comprising elevation data, slope, aspect, vegetation type, combustible water content, wind speed, wind direction angle, temperature, and relative humidity; a second acquisition module configured to acquire a fire source feature vector of the wildfire in the first region based on the set of environmental features, the fire source feature vector comprising a first fire intensity at a fire starting time; a state vector construction module configured to construct a wildfire spread state vector based on the set of environmental features and the fire source feature vector, the wildfire spread state vector comprising a set of fire line coordinates, a fire line spread speed, and a fire line spread direction, the fire line spread speed being determined based on the first fire intensity; and an updating module configured to update the set of fire line coordinates in the wildfire spread state vector based on the fire line spread speed and the fire line spread direction, so as to predict the wildfire in the first region.

[0021] Optionally, the updating module is further configured to implement the updating of the set of fire line coordinates in the wildfire spread state vector by using the following formula:

[0022]

[0023] wherein, is the set of fire line coordinates in the updated wildfire spread state vector, represents ​The set of fireline coordinates at time [time]. The length of the time step represents Time and The time difference between moments.

[0024] Optionally, the device further includes a generation module for generating a wildfire report based on the updated set of fireline coordinates, the wildfire report also including the real-time fire intensity of the wildfire, the real-time fire intensity being calculated based on the set of environmental parameters.

[0025] Optionally, the device further includes a correction module for correcting the updated set of fireline coordinates and the real-time fire intensity in the wildfire report.

[0026] Thirdly, a computer device is also provided, comprising: a memory and a processor, wherein the memory stores at least one computer program, the at least one computer program being loaded and executed by the processor to perform the wildfire spread prediction method described in the above embodiments.

[0027] Fourthly, a computer-readable storage medium is also provided, wherein at least one computer program is stored in the computer-readable storage medium, the at least one computer program being loaded and executed by a processor to perform the wildfire spread prediction method described in the above embodiments.

[0028] Fifthly, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the method described in the first aspect.

[0029] The beneficial effects of the technical solutions provided in this disclosure include at least the following:

[0030] In this embodiment, an environmental feature set of the first region is obtained from a remote sensing image of the first region, and then a wildfire spread state vector is constructed based on the environmental feature set. This wildfire spread state vector includes a fireline coordinate set, a fireline spread velocity, and a fireline spread direction. By updating the fireline coordinate set using the fireline spread velocity and fireline spread direction, accurate prediction of the fireline coordinate set can be achieved. Furthermore, since the environmental feature set includes elevation data, slope, aspect, vegetation type, combustible material moisture content, wind speed, wind direction angle, temperature, and relative humidity, it is equivalent to using multiple parameters to update the fireline coordinate set. Under the premise of the combined effect of multiple parameters, the updated fireline coordinate set can have a high accuracy even in complex environments. Attached Figure Description

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings needed to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 A flow chart of a mountain fire spreading prediction method provided by an example embodiment of the present disclosure is shown;

[0033] Figure 2 A flow chart of a mountain fire spreading prediction method provided by another example embodiment of the present disclosure is shown;

[0034] Figure 3 A structural schematic diagram of a mountain fire spreading prediction device provided by an example embodiment of the present disclosure is shown;

[0035] Figure 4 A structural schematic diagram of a computer device provided by an example embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0036] Unless otherwise defined, technical terms or scientific terms used herein should be understood as having the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure belongs. The terms “first”, “second”, “third” and similar terms used in the specification and claims of the present patent application do not indicate any order, number or importance, but are only used to distinguish different components. Similarly, the terms “one” or “a” and similar terms do not indicate a quantity limitation, but indicate the existence of at least one. The terms “include” or “contain” and similar terms mean that the elements or objects appearing before “include” or “contain” cover the elements or objects listed after “include” or “contain” and their equivalents, and do not exclude other elements or objects.

[0037] In order to make the purposes, technical solutions and advantages of the present disclosure clearer, the embodiments of the present disclosure will be further described in detail below with reference to the drawings.

[0038] Figure 1 A flow chart of a mountain fire spreading prediction method provided by an example embodiment of the present disclosure is shown, which can be executed by a computer device. Referring to Figure 1 , the method comprises:

[0039] In step 101, based on a remote sensing image of a first region, a set of environmental features of the first region is obtained.

[0040] The first region is a region where a mountain fire exists, and the set of environmental features includes elevation data, slope, slope direction, vegetation type, combustible water content, wind speed, wind direction angle, temperature, and relative humidity.

[0041] Here, the application scenario of the embodiments of the present disclosure is that the remote sensing satellite will continuously monitor the global region, so in the case of a mountain fire in any region, the remote sensing satellite can obtain the remote sensing image of the region before the mountain fire and during the burning process of the mountain fire, and the remote sensing satellite will also send the obtained remote sensing image of the region to the computer device in real time.

[0042] The staff can determine the region where the mountain fire currently exists by monitoring the remote sensing image in the computer device, and then set a certain region as the first region. At this time, the computer device includes multiple remote sensing images of the first region.

[0043] Based on the multiple remote sensing images of the first region, a DEM (Digital Elevation Model) of the first region can be established, and when dealing with the deduction problem of a small range of spreading mountain fires, the earth surface of the studied region (the first region) can be regarded as a plane. In this case, the DEM data of the first region can be represented by a projected coordinate system (PCS).

[0044] The first region in the DEM can be divided into a plurality of discrete regular grids, a proper grid size is defined, a grid index system is established, and each grid point is a set of coordinate data. That is, the coordinates of the first region are discretized.

[0045] Any set of coordinate data in the DEM includes: X coordinate (Easting): eastward distance; Y coordinate (Northing): northward distance; Z coordinate (Elevation): elevation value. In this DEM data, the Z coordinate is a binary function of the X coordinate and the Y coordinate. That is, any set of DEM data can be represented by the coordinate on the X coordinate axis and the coordinate on the Y coordinate axis .

[0046] In this case, the environmental feature set at the coordinate of the first region can be represented as formula (1) as follows.

[0047] (1)

[0048] In formula (1), is the elevation data at , that is, the elevation value, is the slope at , is the aspect at , is the vegetation type at the location. the wind speed at the location the wind direction at the location and the wind direction angle the set of the temperature at the location, the relative humidity at the location; the moisture content of the combustible at the location, the normalized moisture index at the location and the relative humidity the moisture content of the combustible at the location, the normalized moisture index at the location and the relative humidity are calculated.

[0049] The way of obtaining each parameter in formula (1) is described below.

[0050] Optionally, the slope at the location is calculated by formula (2).

[0051] (2)

[0052] In formula (2), denotes the partial derivative of with respect to , and denotes the partial derivative of with respect to . The two partial derivatives are calculated by the central difference method, for example, can be calculated by formula (3) to formula (4). The meanings of other parameters in formula (2) are the same as those in formula (1), and are omitted here.

[0053] (3)

[0054] (4)

[0055] In formula (3) and formula (4), denotes the step length used to calculate the partial derivative. The meanings of other parameters in formula (3) and formula (4) are the same as those in formula (1) and formula (2), and are omitted here.

[0056] Optionally, the aspect at the location is calculated by formula (5).

[0057] (5)

[0058] In formula (5), ​This is a two-parameter arctangent function. The meanings of the other parameters in formula (5) are the same as those in formulas (1) to (4), and will not be detailed here. There are many implementation methods for the two-parameter arctangent function in related technologies, and will not be detailed here.

[0059] Optionally, Vegetation type According to Normalized Difference Vegetation Index To determine this. Among them... The normalized vegetation index at the location is calculated using formula (6).

[0060] (6)

[0061] In formula (6), for Normalized Difference Vegetation Index (NDVI) at the location for Near-infrared reflectance at that location for Reflectivity in the red light band at that location. and All of these can be obtained from remote sensing images.

[0062] vegetation type It includes four types: bare land or sparse vegetation, shrubland or sparse woodland, medium-density woodland, and high-density woodland. In this case, explain The area is bare ground or sparsely vegetated; In this case, explain The area is shrubland or sparse woodland; In this case, explain The area is classified as medium-density forest; In this case, explain The area is a high-density forest.

[0063] Optionally, Moisture content of combustibles at the location It can be adopted Normalized Moisture Index at the location and relative humidity at the location The calculation yielded, where The normalized water content index at the location is calculated using formula (7). It is calculated using formula (8).

[0064] (7)

[0065] In formula (7), for the normalized water index at the location, the short infrared band reflectivity at the location, can be obtained from the remote sensing image. The meanings of other parameters in formula (7) are the same as those in formula (6), and details are omitted here.

[0066] (8)

[0067] In formula (8), the relative humidity at the location is the weight. The meanings of other parameters in formula (8) are the same as those in formula (1) and formula (7), and details are omitted here.

[0068] In some cases, for example there are rivers, lakes and other water bodies at the location, the normalized water index should be set to the maximum value, or it is directly considered that these locations are not possible to be on fire because the fire line will not spread to rivers, lakes and other water bodies.

[0069] The wind speed and the wind direction angle at the location are obtained from the collection of the near-surface observation data of the first area. The wind speed at the location and the wind direction angle at the location

[0070] can be obtained from the near-surface observation data of the first area. The temperature at the location is calculated from the thermal infrared remote sensing data at the location (for example, calculated by using the following formula (10)).

[0071] The relative humidity at the location can be obtained from the meteorological data of the first area.

[0072] In step 102, based on the collection of environmental characteristics, the fire source feature vector of the wildfire in the first area is obtained.

[0073] The fire source feature vector includes the first fire intensity at the ignition time.

[0074] Remote sensing satellites can acquire remote sensing images of the first area before and during the fire's combustion. These images typically include thermal infrared remote sensing data, which reflects temperature information. Because of the significant temperature difference before and after the fire, if remote sensing images of the first area before and during the fire's combustion are known, the ignition point (i.e., the fire source) and its related parameters can be located relatively accurately, thus determining the initial fire intensity at the time of ignition.

[0075] Optionally, at the location of the fire source... In the case of fire, the moment of ignition First fire intensity It is calculated using formula (9).

[0076] (9)

[0077] In formula (9), The moment of ignition The intensity of the first fire Indicates the time of fire ignition Location of the fire source The temperature at that location Indicates the time of fire ignition The temperature of the background location in the first region (the average value can be taken). The background location here is the location in the first region other than the ignition point. , For weights.

[0078] in, and All can be detected by the time of ignition. The temperature is calculated from the thermal infrared remote sensing data of the first region at that time. For example, the temperature can be calculated from the thermal infrared remote sensing data using formula (10).

[0079] (10)

[0080] In formula (10), The absolute temperature is calculated from thermal infrared remote sensing data, and the unit is K (Kelvin). , These are two constants in Planck's radiation law. , ;in It is Planck's constant. It is the speed of light in a vacuum. It is the Boltzmann constant. It refers to the observation wavelength, and the unit of observation wavelength is m (meter). Blackbody at wavelength radiation intensity at the location of the fire source.

[0081] In this case, the fire source feature vector can be expressed as .

[0082] In step 103, based on the environment feature vector and the fire source feature vector, a wildfire spread state vector is constructed.

[0083] The wildfire spread state vector includes a fire line coordinate set, a fire line spread speed, and a fire line spread direction, and the fire line spread speed is determined based on the first fire intensity.

[0084] In step 104, based on the fire line spread speed and the fire line spread direction, the fire line coordinate set in the wildfire spread state vector is updated to predict the wildfire in the first region.

[0085] In the embodiments of the present disclosure, by obtaining the environment feature set of the first region from the remote sensing image of the first region, and then constructing the wildfire spread state vector based on the environment feature set, the wildfire spread state vector includes the fire line coordinate set, the fire line spread speed, and the fire line spread direction, and by updating the fire line coordinate set using the fire line spread speed and the fire line spread direction, the fire line coordinate set can be accurately predicted. Moreover, since the environment feature set includes elevation data, slope, aspect, vegetation type, combustible water content, wind speed, wind direction angle, temperature, and relative humidity, it is equivalent to updating the fire line coordinate set using multiple parameters, and under the joint action of multiple parameters, the updated fire line coordinate set can also have high accuracy in complex environments.

[0086] Figure 2 A flowchart of a wildfire spread prediction method provided by another exemplary embodiment of the present disclosure is shown. The method can be executed by a computer device. Referring to Figure 2 , the method includes:

[0087] In step 201, based on a remote sensing image of a first region, an environment feature set of the first region is obtained.

[0088] In step 202, based on the environment feature set, a fire source feature vector of a wildfire in the first region is obtained,

[0089] The related content of steps 201 to 202 is described in the foregoing steps 101 to 102, and is omitted here.

[0090] In step 203, based on the environment feature vector and the fire source feature vector, a wildfire spread state vector is constructed.

[0091] The wildfire spread state vector includes a fire line coordinate set, a fire line spread speed, and a fire line spread direction, and the fire line spread speed is determined based on the first fire intensity.

[0092] Optionally, the mountain fire spreading state vector is expressed by the following formula:

[0093]

[0094] wherein, is the mountain fire spreading state vector at the time t, is the set of fire line coordinates at the time t, is the set of fire line coordinates at the time t, is the fire line spreading speed of each fire line coordinate in the set of fire line coordinates at the time t, is the set of fire line coordinates at the time t, is the fire line spreading direction of each fire line coordinate in the set of fire line coordinates at the time t.

[0095] Herein, the time t can be the current time, or a certain time in the past. can be an array containing all the fire line coordinates of the fire line at the time t, which can be obtained by the remote sensing image of the first area at the time t. Herein, the fire line is not identical to the boundary of the burning area, and the fire line refers to the part of the boundary of the burning area that can continue to burn outward and can affect the unburned area. There can be a part of the boundary of the burning area that cannot continue to burn, for example, when a river, a lake or other water bodies are located at the boundary of the burning area, the fire line does not include the part of the river, the lake or other water bodies. Exemplarily, represents any fire line coordinate in the set of fire line coordinates at the time t,

[0096] i.e. In this case, the fire line spreading speed of the fire line coordinate at the time t in the set of fire line coordinates is calculated by formula (11). the fire line spreading direction of the fire line coordinate at the time t in the set of fire line coordinates is calculated by formula (12).

[0097] (11)

[0098] (12)

[0099] ​​​​​​​​​​​​In formula (11) and formula (12), is the fireline coordinate at time is the fireline spread rate at the fireline coordinate is the fireline spread direction at the fireline coordinate at time is the fireline spread rate at the fireline coordinate at time is wherein is an integer and is greater than 0. is the slope at the fireline coordinate at time is calculated using formula (2). is the wind speed at the fireline coordinate at time is calculated using formula (6). is the wind direction angle at the fireline coordinate at time is calculated using formula (7). is the aspect at the fireline coordinate at time is calculated using formula (5). is the normalized moisture index at the fireline coordinate at time is calculated using formula (7). is the relative humidity at the fireline coordinate at time is calculated using formula (8). is the fuel moisture at the fireline coordinate at time is calculated using formula (8). , , , , , , is the weight.

[0100] wherein, , can be obtained through near-surface meteorological observation station data, or through meteorological radar data inversion, or through scatterometer data satellite remote sensing wind speed products. can also be obtained through meteorological observation data.

[0101] In the above formula (11), the slope affects the mountain fire burning speed, the mountain fire burning speed is different on uphill, downhill and flat land; the wind speed plays an additive role on the mountain fire burning, the slope direction affects the wind direction angle, for example, the wind direction angle is actually equal to the slope direction near the ground; the combustible moisture content plays a role in inhibiting the mountain fire burning (the higher the combustible moisture content, the lower the burning efficiency), and the relative humidity also has a certain inhibitory effect on the mountain fire burning. The fire line spreading direction is related to the slope and the wind direction angle.

[0102] Therefore, based on the environmental characteristics in the formula (11) and the formula (12), the fire line spreading speed and the fire line spreading direction can be calculated.

[0103] Suppose the fire starting moment is , the current moment is , the remote sensing satellite sends the remote sensing image of the first area at each time step, then from moment to the current moment, each time step can be represented in turn as , , … .

[0104] According to the formula (11), the fire line spreading speed at the moment is calculated according to the fire line spreading speed at the moment, the fire line spreading speed at the moment is calculated according to the fire line spreading speed at the moment… the fire line spreading speed at the moment is calculated according to the fire line spreading speed at the moment. Therefore, the fire line spreading speed at the moment is related to the fire line spreading speed at the

[0105] Alternatively, the fire line spreading speed at the moment is calculated by using the formula (13).

[0106] (13)

[0107] In the formula (13), is the fire line spreading speed at the fire source position at the moment, and is the weight. The meanings of other parameters in the formula (13) are the same as those of the parameters in the formula (9), and details are omitted here.

[0108] Through the formula (11) to the formula (13), the fire line coordinate set at the The fireline position of any one of the two firelines calculates the response of the fireline spread speed and the fireline direction, thereby constructing the fire spread vector of the wildfire.

[0109] In step 204, based on the fireline spread speed and the fireline spread direction, the fireline coordinate set in the wildfire spread state vector is updated to predict the wildfire in the first region.

[0110] Optionally, the formula (14) is used to update the fireline coordinate set in the wildfire spread state vector.

[0111] (14)

[0112] In formula (14), is the updated fireline coordinate set in the wildfire spread state vector, that is, the predicted fireline coordinate set at time is the fireline coordinate set at time is the length of the time step, that is, the time difference between adjacent two time steps, for example, the time difference between time and time .

[0113] is a two-dimensional unit vector. The speed in the two-dimensional plane can be decomposed into the speed in the direction and the speed in the direction according to the sine value and the cosine value. The speed in the direction can update the coordinate in the direction, and the speed in the direction can update the coordinate in the direction. Based on this principle, the above formula (14) can be expanded into the form of formula (15) and formula (16). For the fireline coordinate set at time any coordinate , the coordinate is updated (predicted) to at time , where

[0114] is represented by formula (15), is represented by formula (16).

[0115] (15)

[0116] (16)

[0117] In formula (15) and formula (16), represents the updated coordinate of the coordinate, that is, in​​​​​​​​ The meanings of other parameters in formulas (15) and (16) are the same as those in formulas (11) and (12), and details are omitted here.

[0118] In the case of meeting the stopping condition, step 204 can be stopped. The stopping condition includes but is not limited to: reaching a preset maximum time step, a fire line spreading speed being less than a preset speed threshold, and a real-time fire intensity of all points on the fire line being less than a preset fire intensity threshold (indicating that the burning is approaching to stop, and the updating can be stopped), and all fire line coordinates reaching a preset geographical boundary.

[0119] In step 205, a forest fire report is generated based on the updated fire line coordinate set.

[0120] The real-time fire intensity of the forest fire in the forest fire report is also included, and the real-time fire intensity is calculated based on the set of environmental parameters.

[0121] The definition of the real-time fire intensity (FI) is: the heat released per unit time and per unit length of the fire line. Alternatively, the real-time fire intensity is calculated by formula (17).

[0122] (17)

[0123] In formula (17), is the real-time fire intensity at the point, is the heat value of the unit mass of combustible at the point (unit: kJ / kg), is the load amount of the combustible per unit area at the point (unit: kg / m 2 ), is the fire line spreading speed at the point, which is calculated by formula (11). The heat value of the unit mass of combustible and the vegetation type

[0124] have a corresponding relationship, which can be pre-stored in the computer device. The real-time is substituted into the corresponding relationship to obtain the real-time at the point.

[0125] Alternatively, the load amount of the combustible per unit area at the point is calculated by formula (18).

[0126] (18)​​​​​

[0127] In formula (18), is the basic combustible load at the point is the weight. The meanings of other parameters in formula (18) are the same as those in formula (7) and formula (17), and details are omitted here.

[0128] wherein, is the basic combustible load at the point is calculated by formula (19).

[0129] (19)

[0130] In formula (19), , is the weight. The meanings of other parameters in formula (19) are the same as those in formula (6) and formula (18), and details are omitted here.

[0131] Through formula (17) to formula (19), the real-time fire intensity of the first area can be obtained.

[0132] The mountain fire report can include an updated fire line coordinate set, a real-time fire intensity distribution of each fire line coordinate, and the like. The updated fire line coordinate set can be further processed to generate a spread path map, and the area affected by the mountain fire can be calculated.

[0133] In step 206, the updated fire line coordinate set and the real-time fire intensity in the mountain fire report are corrected.

[0134] Ideally, the predicted fire line coordinates can be obtained according to formula (14) to formula (16), and the real-time fire intensity can be calculated by formula (17) to formula (19). However, in the actual burning process of the mountain fire, many factors will affect the predicted fire line coordinates and the real-time fire intensity. Therefore, the predicted fire line coordinates and the real-time fire intensity need to be corrected.

[0135] There is a complex interaction between the mountain fire and the environment. The mountain fire is not only affected by environmental factors, but also significantly changes the surrounding environmental conditions, thereby affecting its own spread behavior. Therefore, in the simulation of mountain fire spread, the dynamic two-way coupling relationship needs to be considered, especially in large and long-lasting mountain fire events. That is, the mountain fire will most likely affect the environment.

[0136] ​The impact of forest fire on the environment will cause errors in the calculated parameters in the set of environmental characteristics. These errors can be considered in the following three aspects: errors in wind speed and wind direction angle, errors in temperature, and errors in combustible load. Errors in wind speed and wind direction angle, as well as errors in temperature, will affect the accuracy of the predicted fire line coordinates; errors in combustible load will affect the accuracy of real-time fire intensity.

[0137] The large amount of heat energy released during the burning of forest fires will change the local atmospheric environment, resulting in the phenomenon of "fire weather". The high temperature generated by forest fires will cause the surrounding air to warm up rapidly, forming strong thermal convection, which will change the local wind field structure and form convergent airflow towards the fire source center. This self-generated wind field is often stronger than the original environmental wind, which can further intensify the fire, especially in large forest fires, which can also form "fire whirls". Therefore, forest fire burning will cause errors in wind speed and wind direction angle.

[0138] During the burning of forest fires, a large amount of oxygen is consumed and carbon dioxide, water vapor, and smoke are released, changing the composition of the local atmosphere. In particular, the production of smoke will significantly reduce the intensity of solar radiation, affecting the distribution of surface temperature. At the same time, the smoke produced by the fire will reduce visibility, affecting fire field observation and rescue operations. Therefore, forest fire burning will cause errors in temperature.

[0139] Forest fires will cause a dramatic change in the surface vegetation coverage, reducing the combustible load in the burned area and changing the surface albedo, which will further affect the local radiation balance and energy exchange process. Therefore, forest fire burning will cause errors in combustible load.

[0140] The following formulas (20) to (24) can be used to correct the above three aspects respectively.

[0141] Alternatively, the wind speed and wind direction angle set at can be corrected using formula (20) .

[0142] (20)

[0143] In formula (20), is the corrected ; is the fire-induced wind field coefficient, which can be obtained through wind tunnel experiments, CFD (Computational Fluid Dynamics) simulation, or historical fire data regression analysis, and is usually related to fire intensity and local topography. is the distance from to the fire line, is a characteristic distance parameter, which is a learnable value. is the direction to the fire location The unit vector of the direction from the fire location to the point is calculated using equation (21). The meanings of other parameters in equation (20) are the same as those in equation (1) and equation (17), and are omitted here.

[0144] (21)

[0145] In equation (21), is the distance between the point and the fire location. The meanings of other parameters in equation (21) are the same as those in equation (20), and are omitted here. Optionally, the temperature at the point is corrected using equation (22).

[0146] .

[0147] (22)

[0148] In equation (22), is the corrected , is the temperature at the point, is the temperature decay coefficient, which can be obtained by regression fitting based on measured data (e.g., ground temperature changes in the fire area) or by using an atmospheric transmission model to calculate the obscuring effect of smoke on solar radiation. . is the smoke concentration at the point. Optionally, the smoke concentration at the point

[0149] can be estimated based on remote sensing data, which can be represented by equation (23). (23)

[0150] In equation (23), is the aerosol optical depth data collected by a remote sensing satellite (e.g., MODIS, VIIRS, Sentinel-5P, etc.) at the point,

[0151] is an empirical regression coefficient, the unit is kg / m 3 . The meanings of other parameters in equation (23) are the same as those in equation (22), and are omitted here. Optionally, the combustible load at the point is corrected using equation (24).

[0152] .​​​​​​​

[0153] (24)

[0154] In formula (24), is the corrected , is the combustible consumption coefficient, which can be determined by burning experiments (such as the burning rate of different vegetation types) , and can also be calculated using existing burning models (such as fuel consumption in the Rothermel fire behavior model) . The meanings of other parameters in formula (24) are the same as those in formula (17) and formula (22), and are omitted here.

[0155] When modifying the above three aspects of parameters, the model can be further refined according to actual conditions. For example, in a large-scale long-duration forest fire, the heat generated by the fire may even affect the local weather system, inducing the formation of thunderstorm clouds, and then possibly causing rainfall or triggering new fire points. This “fire-weather” coupling effect can form a self-reinforcing positive feedback loop, making it difficult to control the fire. For such more complex situations, a higher-level atmospheric-fire coupling model needs to be introduced for simulation in specific implementation, considering the impact of fire heat release on the atmospheric boundary layer structure, and possible precipitation feedback.

[0156] Generally, After modifying the parameters at time t, the modified parameters should be substituted into the operation at time t. .

[0157] When the research area is large and the first area cannot be considered as a plane, the latitude and longitude coordinates are needed, and then the coordinate conversion is performed to obtain the projection coordinates , and then the calculation is performed. In implementation, the UTM (Universal Transverse Mercator Grid System) projection can be used to perform coordinate conversion on latitude and longitude coordinates.

[0158] In the case of complex climate and terrain in the first area, the complex climate and terrain will affect the burning of forest fires in the first area, and the influence of complex climate and terrain on forest fires needs to be corrected. The complex climate and terrain include but are not limited to: large diurnal temperature difference, rainfall, canyon, etc.

[0159] Parameters such as temperature and relative humidity are not updated in real time, but rather as weather forecast data is updated, typically on an hourly basis. Therefore, when there are large diurnal temperature variations, the temperature and relative humidity may fluctuate significantly between adjacent hours. This means that the temperature and relative humidity at each time step are unclear, which can affect the predicted fire line coordinate set. Therefore, corrections to the temperature and relative humidity are necessary in such cases.

[0160] Rainfall has a significant inhibitory effect on wildfires. Rainfall can reduce the real-time fire intensity and affect the speed of fire spread, which in turn affects the predicted fire line coordinate set and the real-time fire intensity.

[0161] The spread of wildfires under complex terrain conditions is even more complicated. Taking the spread of wildfires in canyons as an example, due to the canyon effect, the wind speed in canyons is often high, which leads to errors in wind speed and wind direction angle, and thus affects the predicted set of fire line coordinates.

[0162] The following formulas (25) to (28) can be used to correct for complex climate and terrain conditions.

[0163] Optionally, when there is a large diurnal temperature range, a diurnal temperature variation model can be used to determine the temperature. Temperature at any given time. This process can be represented by formula (25).

[0164] (25)

[0165] In formula (25), This is a model of daily temperature variation, representing the... Temperature at any moment; The average temperature of the day, It is the amplitude of temperature throughout the day. It is the time of day when the highest temperature occurs. It is the current moment. Furthermore, in calculating formula (25), 1 day needs to be converted to... and Same units, for example, in and The unit is seconds, so one day needs to be converted to 86,400 seconds before calculation.

[0166] Optionally, when there is a large diurnal temperature range, formula (26) can be used to calculate. Relative humidity at any given time.

[0167] (26)

[0168] In formula (26), express Relative humidity at any given time is a reference humidity of relative humidity, is a temperature-humidity relationship coefficient. The meanings of other parameters in equation (26) are the same as those in equation (25) and are omitted here.

[0169] Optionally, in the case where there is a rainfall area in the first area, equation (27) is used to correct the fire intensity in the rainfall area.

[0170] (27)

[0171] In equation (27), represents the corrected real-time fire intensity in the case where represents the real-time fire intensity in the case where is a rainfall influence coefficient, represents the rainfall intensity at

[0172] Optionally, in the case where there is a canyon landform in the first area, equation (28) is used to correct the wind speed and wind direction angle in the canyon landform.

[0173] (28)

[0174] In equation (28), is the corrected (here, is located in the canyon landform of the first area), is a canyon effect coefficient, is a canyon depth, is a canyon width. The meanings of other parameters in equation (28) are the same as those in equation (1) and are omitted here.

[0175] For other special cases, they can also be corrected according to actual situations.

[0176] In actual situations, humans will also have an impact on the mountain fire, for example, fire fighting actions will suppress the fire intensity. In this case, the corrected real-time fire intensity can be represented by equation (29).

[0177] (29)

[0178] In equation (29), is the corrected (here, is located in the area of the first area affected by the fire fighting action), is a fire extinguishing efficiency coefficient, representing the fire extinguishing efficiency, The greater the value, the higher the fire extinguishing efficiency. represents the fire-fighting force put in, represents the distance to the fire line.

[0179] The weights and coefficients involved in the above formulas (1) to (29) (such as to , , , etc.) can be taken as empirical values or obtained using a deep learning method. If the weights are obtained based on empirical values, the sources of empirical data can include historical fire case data, field test data, expert experience values, etc. Statistical regression analysis, parameter sensitivity analysis, Monte Carlo simulation, etc. can be used to analyze the empirical data to obtain the empirical values.

[0180] If the deep learning method is used to obtain the weights, a model can be established, which includes CNN, LSTM and Transformer. Among them, CNN is used to process spatial features, LSTM is used to process time series features, and Transformer is used to process multi-modal data. The model is trained by inputting multi-spectral remote sensing images, DEM data, meteorological data, historical fire data, etc. to the model. Finally, the trained model is used to obtain the weights.

[0181] In actual application, the empirical values and the deep learning method can be combined for use. For example, the empirical values can be used to quickly establish a model at the beginning, and as data accumulates, the deep learning method can be gradually introduced. The two methods are combined for use and mutual verification, and the parameters are updated and optimized regularly. It is also necessary to establish a parameter verification mechanism, consider the adaptability of different regions, and pay attention to the real-time performance of the model in the application of mountain fire disaster emergency.

[0182] The following is an apparatus embodiment of the present application. For details not described in detail in the apparatus embodiment, reference can be made to the above method embodiments.

[0183] Figure 3 The structure schematic diagram of the mountain fire spread prediction apparatus provided by one example embodiment of the present application is shown. Referring to Figure 3 , the mountain fire spread prediction apparatus 300 includes a first acquisition module 301, a second acquisition module 302, a state vector construction module 303, and an update module 304.

[0184] The first acquisition module 301 is configured to acquire an environmental feature set of a first region based on a remote sensing image of the first region, the first region being a region where there is a mountain fire, and the environmental feature set including elevation data, slope, slope direction, vegetation type, combustible water content, wind speed, wind direction angle, temperature, and relative humidity.

[0185] The second acquisition module 302 is configured to acquire a fire source feature vector of the wildfire in the first region based on the set of environmental features, the fire source feature vector including a first fire intensity at a fire starting moment;

[0186] The state vector construction module 303 is configured to construct a wildfire spreading state vector based on the environmental feature vector and the fire source feature vector, the wildfire spreading state vector including a set of fire line coordinates, a fire line spreading speed, and a fire line spreading direction, the fire line spreading speed being determined based on the first fire intensity;

[0187] The update module 304 is configured to update the set of fire line coordinates in the wildfire spreading state vector based on the fire line spreading speed and the fire line spreading direction, so as to predict the wildfire in the first region.

[0188] Optionally, the update module 304 is further configured to update the set of fire line coordinates in the wildfire spreading state vector by using the following formula:

[0189]

[0190] wherein, is the set of fire line coordinates in the updated wildfire spreading state vector, represents the set of fire line coordinates at the moment t, is the set of fire line coordinates at the moment t, is the length of a time step, represents a time difference between the moment t and the moment t+1.

[0191] Optionally, the apparatus further includes a generation module 305 and a correction module 306.

[0192] The generation module 305 is configured to generate a wildfire report based on the updated set of fire line coordinates, the wildfire report further including a real-time fire intensity of the wildfire, the real-time fire intensity being calculated based on the set of environmental parameters.

[0193] The correction module 306 is configured to correct the updated set of fire line coordinates and the real-time fire intensity in the wildfire report.

[0194] It should be noted that the wildfire spreading prediction apparatus provided in the above embodiments is used to perform wildfire spreading prediction, and the division of the above functional modules is used as an example for illustration, and in actual applications, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the apparatus is divided into different functional modules to complete all or part of the functions described above. In addition, the wildfire spreading prediction apparatus and the wildfire spreading prediction method provided in the above embodiments belong to the same concept, and the specific implementation process is described in the method embodiments, which will not be repeated here.

[0195] ​​The division of the modules in the embodiments of the present disclosure is illustrative, and is merely a logical function division. In actual implementation, another division manner can be used. In addition, each function module in each embodiment of the present disclosure can be integrated in one processor, or can be physically separated, or two or more modules can be integrated in one module. The integrated module can be implemented in the form of hardware or in the form of a software function module.

[0196] When the integrated module is implemented in the form of a software function module and sold or used as an independent product, the integrated module can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present disclosure, essentially or in other words, the part that contributes to the prior art or the whole or part 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 several instructions for causing an end device (which can be a personal computer, a mobile phone, or a communication device, etc.) or a processor to execute all or part of the steps of the method of each embodiment of the present disclosure. The foregoing 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.

[0197] Figure 4 is a structural schematic diagram of a computer device provided by the embodiments of the present disclosure. As shown in Figure 4 the computer device 400 includes a processor 401 and a memory 402.

[0198] Processor 401 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 401 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 401 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 401 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 401 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0199] The memory 402 may include one or more computer-readable storage media, which may be non-transitory. The memory 402 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 402 are used to store at least one instruction, which is executed by the processor 401 to implement the wildfire spread prediction method provided in this disclosure embodiment.

[0200] Those skilled in the art will understand that Figure 4 The structure shown does not constitute a limitation on the computer device 400, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0201] This disclosure also provides a non-transitory computer-readable storage medium, wherein when the instructions in the storage medium are executed by the processor of a computer device, the computer device is able to execute the wildfire spread prediction method provided in this disclosure.

[0202] This disclosure also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the wildfire spread prediction method provided in this disclosure.

[0203] The above merely describes optional embodiments of the present disclosure, and is not intended to limit the present disclosure. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for predicting the spread of wildfires, characterized in that, The method includes: Based on the remote sensing image of the first region, an environmental feature set of the first region is obtained. The first region is an area where wildfires exist. The environmental feature set includes elevation data, slope, aspect, vegetation type, combustible water content, wind speed, wind direction angle, temperature, and relative humidity. Based on the set of environmental features, the fire source feature vector of the wildfire in the first region is obtained, and the fire source feature vector includes the first fire intensity at the time of ignition. Based on the environmental feature vector and the fire source feature vector, a wildfire spread state vector is constructed. The wildfire spread state vector includes a set of fire line coordinates, a fire line spread speed, and a fire line spread direction. The fire line spread speed is determined based on the first fire intensity. Based on the fire spread speed and the fire spread direction, update the fire line coordinate set in the wildfire spread state vector to predict wildfires in the first region; The wildfire spread state vector is represented by the following formula: in, for The wildfire spread state vector at time t, for The set of fireline coordinates at time [time]. for The set of fireline coordinates at the specified time The fire spread speed at each fireline coordinate. for The set of fireline coordinates at the specified time The direction of fire spread at each fireline coordinate; express The set of fireline coordinates at the specified time Any fireline coordinate in , Timing of the fireline coordinates The speed of fire spread at the location The following formula is used to calculate: Timing of the fireline coordinates The direction of the fire line spread The following formula is used to calculate: in, for Always at the front line coordinates The speed at which the fire spreads. for Always at the front line coordinates The slope mentioned therein, for Always at the front line coordinates The wind speed at that location, for Always at the front line coordinates The wind direction angle at that location, for Always at the front line coordinates The slope direction at that location, for Always at the front line coordinates Normalized water content at the location, for Always at the front line coordinates The relative humidity at that location, for Always at the front line coordinates The moisture content of the combustible material at the location, , , , , , , For weights.

2. The method according to claim 1, characterized in that, The updating of the fireline coordinate set in the wildfire spread state vector includes: The fireline coordinate set in the wildfire spread state vector is updated using the following formula: in, The set of fireline coordinates in the updated wildfire spread state vector represents... The set of fireline coordinates at time [time]. The length of the time step represents Time and The time difference between moments.

3. The method according to claim 1, characterized in that, The method further includes: A wildfire report is generated based on the updated set of fireline coordinates. The wildfire report also includes the real-time fire intensity, which is calculated based on the set of environmental parameters.

4. The method according to claim 3, characterized in that, The method further includes: The updated set of fireline coordinates and the real-time fire intensity in the wildfire report have been corrected.

5. A wildfire spread prediction device, characterized in that, The device includes: The first acquisition module is used to acquire an environmental feature set of the first region based on the remote sensing image of the first region, where the first region is a region where wildfires exist. The environmental feature set includes elevation data, slope, aspect, vegetation type, combustible water content, wind speed, wind direction angle, temperature, and relative humidity. The second acquisition module is used to acquire the fire source feature vector of the wildfire in the first region based on the environmental feature set, wherein the fire source feature vector includes the first fire intensity at the time of ignition. The state vector construction module is used to construct a wildfire spread state vector based on the environmental feature vector and the fire source feature vector. The wildfire spread state vector includes a fire line coordinate set, a fire line spread speed, and a fire line spread direction. The fire line spread speed is determined based on the first fire intensity. An update module is used to update the fire line coordinate set in the wildfire spread state vector based on the fire line spread speed and the fire line spread direction, so as to predict wildfires in the first area; In the state vector construction module, the wildfire spread state vector is represented by the following formula: in, for The wildfire spread state vector at time t, for The set of fireline coordinates at time [time]. for The set of fireline coordinates at the specified time The fire spread speed at each fireline coordinate. for The set of fireline coordinates at the specified time The direction of fire spread at each fireline coordinate; express The set of fireline coordinates at the specified time Any fireline coordinate in , Timing of the fireline coordinates The speed of fire spread at the location The following formula is used to calculate: Timing of the fireline coordinates The direction of the fire line spread The following formula is used to calculate: in, for Always at the front line coordinates The speed at which the fire spreads. for Always at the front line coordinates The slope mentioned therein, for Always at the front line coordinates The wind speed at that location, for Always at the front line coordinates The wind direction angle at that location, for Always at the front line coordinates The slope direction at that location, for Always at the front line coordinates Normalized water content at the location, for Always at the front line coordinates The relative humidity at that location, for Always at the front line coordinates The moisture content of the combustible material at the location, , , , , , , For weights.

6. A computer device, characterized in that, The computer device includes a memory and a processor, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to implement the method of any one of claims 1 to 4.

8. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 1 to 4.

Images