A method, device and medium for simulating forest fire spread based on combustion probability

Abstract

The embodiment of the present invention discloses a method, device and medium for simulating the spread of forest fires based on burning probability. The method includes: obtaining the initial position of the forest fire; traversing each burning cell in each time step of the simulation, and performing the following operations: S1, determining the neighboring cells adjacent to the current burning cell, and drawing the wind speed vector with the current burning cell as the starting point, and taking the cells through which the wind speed vector passes as the supplementary neighboring cells; S2, for each unburned neighboring cell in all neighboring cells, sequentially performing the following operations: when the vegetation coverage rate of the unburned current neighboring cell is greater than or equal to the burning threshold, matching the slope threshold corresponding to the current neighboring cell according to the wind speed vector; when the slope of the current neighboring cell is greater than or equal to the slope threshold, generating a random number in the interval [0,1]; if the random number is less than or equal to the vegetation coverage rate, igniting the current neighboring cell.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of forest fire spread simulation, and in particular to a forest fire spread simulation method, device and medium based on combustion probability. Background Art

[0002] A forest fire is a type of fire that spreads and expands freely in the forest without human control, causing serious harm to the forest, ecological environment and public safety.

[0003] The cellular automaton model can simulate the spread of fire under different terrains, vegetation types and meteorological conditions. However, existing simulation methods generally have problems such as over-emphasis on forest fire prediction and forecasting, over-simplification of the simulation platform, and insufficient simulation effect.

[0004] How to combine real natural conditions and vegetation data, based on the characteristics of forest fire burning, to achieve simulation of forest fire spread is an urgent problem to be solved. Summary of the invention

[0005] The embodiments of the present invention provide a method, device and medium for simulating forest fire spread based on combustion probability to solve the above problems.

[0006] In a first aspect, an embodiment of the present invention provides a forest fire spread simulation method based on combustion probability, comprising:

[0007] Obtaining the initial position of the forest fire, and initializing the combustion cells in the automatic cellular machine model according to the initial position;

[0008] In each time step of the simulation, each combustion cell is traversed and the following operations are performed on each combustion cell in turn:

[0009] S1. Determine the neighboring cells adjacent to the current burning cell, and draw a wind speed vector with the current burning cell as the starting point, and use the cells passed by the wind speed vector as supplementary neighboring cells, wherein the direction of the wind speed vector is consistent with the wind direction, and the length is proportional to the wind speed;

[0010] S2. For all the unburned neighboring cells in all the neighboring cells, perform the following operations in sequence:

[0011] S2-1, when the vegetation coverage rate of the unburned current neighborhood cell is greater than or equal to the burning threshold, matching the slope threshold corresponding to the current neighborhood cell according to the wind speed vector, wherein the slope threshold is less than 0, and the higher the wind speed, and / or the smaller the angle between the position vector from the current burning cell to the current neighborhood cell and the wind speed vector, the lower the slope threshold;

[0012] S2-2. When the slope of the current neighborhood cell is greater than or equal to the slope threshold, generate a random number in the interval [0,1]; if the random number is less than or equal to the vegetation coverage rate, ignite the current neighborhood cell.

[0013] In a second aspect, an embodiment of the present invention provides an electronic device, the electronic device comprising:

[0014] one or more processors;

[0015] a memory for storing one or more programs,

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the forest fire spread simulation method based on combustion probability described in any embodiment.

[0017] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the forest fire spread simulation method based on combustion probability as described in any embodiment.

[0018] In summary, this embodiment provides a method for simulating the spread of forest fires based on burning probability, which simulates the spread trend of forest fires based on the cellular automaton model, and reflects the effect of wind on the spread of forest fires by supplementing neighborhood grids along the wind speed direction (the direction of the wind); reflects the comprehensive influence of vegetation, terrain and wind on the spread of forest fires by setting vegetation coverage rate thresholds and slope thresholds; and simulates the randomness of ignition of cells by random numbers in the interval [0,1]. In multiple ways, the accuracy of forest fire simulation is improved together.

[0019] In particular, when supplementing the neighborhood, this embodiment does not expand all neighborhood cells outward along the wind speed vector. This is because the simulation itself has certain errors. Whether the original Moore neighborhood cell is ignited is a probabilistic event, while the current burning cell has been burned, which is a given event. Only the extension of the current burning cell along the wind speed direction is retained as a supplement to the neighborhood, which can effectively avoid error diffusion.

[0020] At the same time, when setting the slope threshold, this embodiment matches a differentiated slope threshold for each cell according to the wind speed vector. The higher the wind speed, the lower the slope threshold; the smaller the angle between the position vector from the current burning cell to the current neighboring cell and the wind speed vector, the lower the slope threshold; thereby more accurately reflecting the impact of wind force and slope on different cells during the spread of forest fires, and improving the accuracy of forest fire simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 is a flow chart of a forest fire spread simulation method based on combustion probability provided by an embodiment of the present invention;

[0023] Figure 2 is a partial data source of a fire simulation area provided by an embodiment of the present invention, wherein: Figure 2 (a) is Landsat8 image data, Figure 2 (b) is NDVI data, Figure 2 (c) is DEM data;

[0024] Figure 3 is a schematic diagram of a Moore neighborhood provided by an embodiment of the present invention;

[0025] Figure 4 is a flow chart of another forest fire spread simulation method based on combustion probability provided by an embodiment of the present invention;

[0026] Figure 5 is a binary image of a fire spread simulation result provided by an embodiment of the present invention;

[0027] Figure 6 This is a schematic diagram of a forest fire spread simulation result after superimposing Landsat8 images provided by an embodiment of the present invention;

[0028] Figure 7 is a schematic diagram of a change trend of forest fire spread area under the influence of different parameters provided by an embodiment of the present invention, wherein: Figure 7 (a) Figure 7 (b) Figure 7 (c) Figure 7 (d) The directions of the stroke were south, east, north, and west;

[0029] Figure 8 It is a schematic diagram of forest fire simulation results provided by an embodiment of the present invention when the wind direction is north and the wind speed is level 2;

[0030] Fig. 9 It is a schematic diagram of forest fire simulation results provided by an embodiment of the present invention when the wind direction is south and the wind speed is level 2;

[0031] Fig.10 It is a schematic diagram of forest fire simulation results provided by an embodiment of the present invention with the wind direction being east and the wind speed being level 2;

[0032] Fig.11 This is a schematic diagram of forest fire simulation results provided by an embodiment of the present invention when the wind direction is west and the wind speed is level 2;

[0033] Fig.12 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0035] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0036] In the description of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0037] The embodiment of the present invention provides a method for simulating the spread of forest fires based on combustion probability. In order to illustrate the method, the principle of cellular automaton simulation of fire is first introduced. The principle of cellular automaton simulation of the spread of forest fires is based on a discretized space and time model to describe and predict the evolution of the states and interactions of each cell in a complex system. In the cellular automaton method, the fire area is divided into many small units, each of which is called a "cell", and each cell has its own state and attributes; these cells interact according to the law of fire diffusion and the state of adjacent cells to simulate the process of fire spread.

[0038] Specifically, the cellular automaton model consists of four parts: cells, states, neighborhoods, and state transfer rules. The cellular automaton can be regarded as a cell space and a function defined in the space. Its form is defined as:

[0039] CA=(N,S,NC,R)

[0040] Among them, CA represents a cellular automaton system, N represents the grid space composed of cells, S is a finite set used to represent the state of the cell, NC represents the neighborhood, and R represents the transmission law of the cell state and the neighborhood cell state, which determines the state and result of the evolution of the cellular automaton. Therefore, the cellular automaton is a dynamic system defined on a discrete, finite cellular space and evolved in a discrete time dimension according to natural laws. Its principle is to decompose complex natural phenomena or physical systems into several time steps in time, and each time step is decomposed into continuous grids in space. The state of each grid is determined by the state of the adjacent grids around the previous time step and the natural law of state transmission.

[0041] ① State and state transmission law: Each cell has a certain state, which represents the flame state of the area, such as burning, unburned, or completely burned, partially burned, unburned, etc. According to the state of adjacent cells and the law of fire diffusion, the state of each cell will be updated as time steps go by.

[0042] ② Neighborhood relations and interactions: The cellular automaton model updates the state of the current cell based on the state of the neighboring cells and the law of fire diffusion. Specifically, the Moore neighbor relationship can be used to consider the state of the neighboring cells around the current cell.

[0043] ③ Iterative update: The cellular automaton model simulates the spread of fire through iterative updates. In each time step, the state of the current cell and the state of the adjacent cells are calculated to update the state of the current cell. The simulation will continue to iterate until the specified simulation time is reached or the termination condition is met.

[0044] In general, the cellular automaton model can simulate the spread path and changes in fire intensity of fire through different cell states and the diffusion law of fire. This model can better capture the details and local characteristics of the flames and is suitable for fire spread simulation under complex terrain and vegetation conditions.

[0045] Based on the above principles, Figure 1 1 is a flow chart of a method for simulating the spread of forest fires based on combustion probability provided by an embodiment of the present invention. The method is applicable to the case of simulating the spread of forest fires and is executed by an electronic device. Figure 1 As shown, the method specifically includes:

[0046] S110, obtaining an initial position of the forest fire, and initializing the combustion cells in the automatic cellular machine model according to the initial position.

[0047] The spread of forest fire is very similar to the principle of cellular automation. The fire simulation area is regarded as a space composed of multiple square grids, and each grid point is a cell. Figure 2 (a) is used to divide the grid, and each grid is a cell. After the division is completed, the cell where the initial position of the forest fire is located is marked as the burning state. Of course, if the initial position of the forest fire is accurate enough, the actual burning area of ​​the cell can also be used to divide the burning grid into complete burning and partial burning. Optionally, the cell state at time step t is defined as:

[0048]

[0049] in, The value range is: if Indicates that the state of cell (i, j) at time step t is unburned; Indicates that the state of cell (i, j) at time step t is partially burned; Indicates that the state of cell (i, j) at time step t is completely burned. Only cells that are completely burned spread to neighboring cells.

[0050] S120, traversing each combustion cell in each time step of the simulation, and updating the fire status of neighboring cells layer by layer outward according to the status of each combustion cell, and repeating this cycle until the simulation period ends.

[0051] This step uses a specific duration as the time step unit of the simulation, and simulates the spread of the fire layer by layer from the burning cell as the center in each time step. Optionally, the time for the burning cell to spread for a cell length in a windless and slopeless state can be used as the length of the time step. Under this step length, only the state of the outer layer of neighboring cells needs to be updated in each simulation time step.

[0052] In a specific implementation, each combustion cell may be traversed in each time step, and the following operations may be performed on each combustion cell:

[0053] S1. Determine the neighboring cells adjacent to the current burning cell, and draw a wind speed vector with the current burning cell as the starting point, and use the cells passed by the wind speed vector as supplementary neighboring cells, wherein the direction of the wind speed vector is consistent with the wind direction, and the length is proportional to the wind speed. The wind direction here is the direction of the wind, that is, the direction in which the wind blows.

[0054] Optional, you can use Figure 3 The Moore neighborhood shown in FIG. 1 shows that there are 8 neighboring cells adjacent to the current burning cell (i, j). Then, according to the cell side length and wind direction, the wind speed is converted into the number of cells passed per unit time step; according to the number of cells, the wind speed vector starting from the current burning cell is drawn. For example, Figure 3 The cell side length is 30 meters. When the wind direction is 45° to the lower right, the wind speed value is multiplied by the simulation time step and then divided by The wind speed can be converted into the number of cells per unit time step, thus drawing the following Figure 3 The wind speed vector shown by the solid arrow, the dotted cell through which the vector passes is the supplemented neighborhood cell, which is added to the neighborhood cell set. Due to the effect of wind force, the probability of the supplemented neighborhood cell being ignited by the central cell within a unit time step is very high, so the cell state in this direction will be updated in each time step.

[0055] It is worth mentioning that in this embodiment, all neighborhood cells are not extended outward along the wind speed vector, that is, the cells in the lower right corner of all neighborhood cells are not used as supplementary neighborhood cells. This is because the simulation itself has certain errors. Whether the original Moore neighborhood cell is ignited is a probabilistic event, while the current burning cell has been burned, which is a given event. Therefore, only the extension of the current burning cell along the wind speed direction is retained as a supplement to the neighborhood, which can effectively avoid error diffusion.

[0056] S2. Extract the unburned neighboring cells (including the directly adjacent neighboring cells and the supplemented neighboring cells) from the neighboring cell set, and determine whether these cells will be ignited in the current time step, thereby updating the states of these cells as the results of the simulation of the current time step.

[0057] In a specific implementation, each unburned neighboring cell may be traversed, and the following operations may be performed on each unburned neighboring cell in sequence:

[0058] S2-1. When the vegetation coverage rate of the unburned current neighborhood cell is greater than or equal to the burning threshold, the slope threshold corresponding to the current neighborhood cell is matched according to the wind speed vector, wherein the slope threshold is less than 0, and a threshold value less than 0 corresponds to a situation where the current neighborhood cell is lower than the current burning cell.

[0059] S2-2. When the slope of the current neighborhood cell is greater than or equal to the slope threshold, generate a random number in the interval [0,1]; if the random number is less than or equal to the vegetation coverage rate, ignite the current neighborhood cell.

[0060] Among them, the vegetation coverage rate (NAVI) refers to the ratio of the vegetation area in the cell to the cell area, such as Figure 2(b) is shown. The main burning material of forest fire is local vegetation. If the vegetation coverage of a cell is too small, the cell cannot be ignited or cannot cause the spread of forest fire. Therefore, a cell can only be ignited when its vegetation coverage is greater than a certain threshold. In this embodiment, this threshold is called the combustion threshold.

[0061] The slope is the slope from the current burning cell to the unburned current neighboring cell, that is:

[0062]

[0063] Where slop represents the slope, in degrees; h2 represents the height of the current neighborhood cell, h1 represents the height of the current burning cell; s represents the horizontal distance between the current neighborhood cell and the current burning cell, s>0. The data for calculating the slope can come from the elevation data (DEM data) of the fire simulation area, such as Figure 2 (c) shows that forest fire spreads easily along the rising slope direction and has difficulty spreading along the descending slope direction. Therefore, this embodiment uses a slope threshold value less than 0 to characterize the influence of the slope on the spread of forest fire. When the cell slope is greater than or equal to the slope threshold value, the cell may be ignited.

[0064] In particular, this embodiment takes into account the impact of wind on the spread of forest fires, and matches different slope thresholds for each cell according to different wind speeds (i.e., wind speed values) and wind directions. Specifically, the higher the wind speed, the lower the slope threshold of the cell; at the same time, the smaller the angle α between the position vector from the current burning cell to the current neighborhood cell and the wind speed vector, the lower the slope threshold, where the angle has a value interval of [0°, 180°]. The position vector here does not take into account the height data, but is actually a plane position vector. Typically, if the current neighborhood cell is a cell that the wind speed vector passes through, combined with Figure 3 ,Right now Figure 3 The cells passed by the solid line wind speed vector, at this time, the angle between the position vector pointing from the current burning cell (i, j) to the current neighboring cell (i+1, j+1) and the wind speed vector is the smallest (0° angle), and the slope thresholds corresponding to these neighboring cells can be adjusted to be lower than the slope thresholds of other neighboring cells to indicate that the ability of forest fire spread to break through the slope threshold under the action of wind speed is improved. If the current neighboring cell is the cell passed by the reverse direction vector of the wind speed vector, combined with Figure 3 ,Right now Figure 3For the cell through which the dotted vector in the opposite direction to the solid wind speed vector passes, the angle between the position vector from the current burning cell (i, j) to the current neighborhood cell (i-1, j-1) and the wind speed vector is the largest (180° angle). The slope threshold corresponding to the neighborhood cell can be increased to be higher than the slope thresholds of other neighborhood cells (but always less than 0) to indicate that the ability of forest fire spread to break through the slope threshold under the action of wind speed is weakened.

[0065] In an optional implementation, the matching slope threshold may also be determined according to the following formula:

[0066]

[0067] Among them, slope T Represents the slope threshold of the current neighborhood cell (i2, j2), (i2, j2) is the cell coordinate, refer to Figure 3 , for the neighborhood cell (i,j+1), (i2,j2)=(i,j+1), and the rest of the neighborhood cells are similar; represents the wind speed vector, represents the slope vector, that is, the vector pointing from the three-dimensional geographic location (i, j, h1) of the current burning cell to the three-dimensional geographic location (i2, j2, h2) of the current neighboring cell; represents the dot product of two vectors, It is the projection of the wind speed vector along the slope vector; a and b are pre-calibrated values. Preferably, when the wind speed unit is m / s, a = 10° and b = 1.5. This formula is obtained by The magnitude of the wind speed, the angle between the position vector of the current burning cell to the current neighboring cell and the wind speed vector, and the influence of the slope value are also considered, so that the assisting effect of the wind is simulated more comprehensively.

[0068] In another optional implementation, the slope threshold may be determined according to the following formula:

[0069]

[0070] in, The modulus of the wind speed vector, i.e., the wind speed value, is preferably, when the wind speed unit is m / s, a=10°, b=1.5. The formula also takes into account the magnitude of the wind speed, as well as the angle between the position vector of the current combustion cell to the current neighboring cell and the wind speed vector, and can provide good simulation results within a certain accuracy range.

[0071] Through the above-mentioned differentiated slope thresholds, the impact of wind and slope on different cells can be more accurately reflected during the spread of forest fires, thereby improving the accuracy of forest fire simulation. Only when the vegetation coverage rate is greater than and the slope is greater than the corresponding threshold, the current neighborhood cell has the possibility of being ignited. With this possibility, the randomness of ignition is simulated by a random number in the [0,1] interval. If the random number is in the [0,NDVI] interval, a random event occurs and the current neighborhood cell is ignited; if the random number is not in the [0,NDVI] interval, the random event does not occur, and the state of the current neighborhood cell remains unchanged, where NDVI represents the vegetation coverage rate of the current neighborhood cell.

[0072] Correspondingly, when the vegetation coverage rate of the unburned current neighborhood cell is less than the burning threshold, or the slope is less than the slope threshold matched by the cell, or the random number is greater than the vegetation coverage rate, the current neighborhood cell is not ignited, and the state of the current neighborhood cell remains unchanged.

[0073] In one embodiment, when the domain cell has multiple layers (e.g. Figure 3 As shown, there are two layers of neighborhood cells, the first layer includes 8 Moore neighborhood cells, and the second layer includes 1 supplementary neighborhood cell). The above-mentioned S2-1 to S2-2 operations can be performed on each unburned neighborhood cell layer by layer from near to far according to the distance between each neighborhood cell and the current burning cell, so as to simulate the spread of forest fires per unit time step layer by layer. Optionally, firstly perform the operations S2-1 and S2-2 on each unburned neighborhood cell in the first layer of neighborhood cells; then, use each ignited neighborhood cell in the first layer of neighborhood cells as the new current burning cell, and perform the operations S2-1 and S2-2 on each unburned neighborhood cell adjacent to each new current cell in the second layer of neighborhood cells; and spread outward layer by layer until all layers of neighborhood cells are executed. Figure 3 For example, first take cell (i, j) as the current burning cell and update the status of the 8 Moore neighborhood cells; when cell (i+1, j+1) is ignited, take cell (i+1, j+1) as the current burning cell and update the status of the supplemented neighborhood cell (i+2, j+2).

[0074] Furthermore, the above embodiment takes the cell states of burning and unburned as an example. In another specific implementation, when considering the actual combustion area of ​​the cell, the combustion cell can be further divided into two states: partial combustion and complete combustion. At this time, when executing S1, it should first be determined whether the current combustion cell state is partial combustion or complete combustion; according to the determination result, the operation of S1 can be completed in the following two cases:

[0075] Case 1: The state of the current combustion cell is complete combustion. At this time, the same operation as the above embodiment is performed, that is, the neighboring cells adjacent to the current combustion cell are determined, and the wind speed vector is drawn with the current combustion cell as the starting point, and the cells passed by the wind speed vector are used as supplementary neighboring cells.

[0076] Case 2: The state of the current burning cell is partially burned, indicating that the cell has not been completely burned and does not have the conditions to spread to the surrounding cells. At this time, the newly added burning area of ​​the current burning cell in the current time step can be calculated based on the forest fire spread speed, where the forest fire spread speed is a pre-known quantity, and its calculation process will be described in subsequent embodiments. After the calculation is completed, it is determined whether the ratio of the accumulated burning area of ​​the current burning cell (the total burning area at each time step) to the cell area has reached the vegetation coverage rate of the current burning cell.

[0077] If the ratio does not reach the vegetation coverage rate of the current burning cell, indicating that the cell still cannot be completely burned in the current time step, the partial combustion state of the current burning cell is maintained unchanged, and the operation on the current burning cell is terminated.

[0078] If the ratio reaches the vegetation coverage rate of the current burning cell, indicating that the vegetation in the cell has been completely ignited, the state of the current burning cell can be marked as completely burned, and the neighboring cells adjacent to the current burning cell can be determined. The wind speed vector is drawn with the current burning cell as the starting point, and the cells passed by the wind speed vector are used as supplementary neighboring cells.

[0079] Accordingly, no matter whether S1 is executed according to situation 1 or situation 2, as long as the current neighborhood cell is finally ignited through S2, the state of the neighborhood cell should first be marked as partially burned in S2-2; then, according to the forest fire spread speed, the newly added burning area of ​​the neighborhood cell in the remaining time of the current time step is calculated. If the ratio of the accumulated burning area of ​​the neighborhood cell to the cell area reaches the vegetation coverage rate of the neighborhood cell, the state of the neighborhood cell is marked as completely burned. If the ratio of the accumulated burning area of ​​the neighborhood cell to the cell area does not reach the vegetation coverage rate of the neighborhood cell, the state of the neighborhood cell is kept as partially burned. Then, the operation on the current neighborhood cell is terminated, and the operation on the next unburned neighborhood cell or the next burning cell is started, and this cycle is repeated until all burning cells are traversed, the simulation of the current time step is terminated, and the next time step is started. The above process can also be combined with Figure 4 The flowchart shown can be understood, and the height threshold in the figure is the slope threshold.

[0080] It is worth mentioning that this embodiment corrects the combustion state of the cell according to the characteristics of forest fire. When , cell (i,j) is completely burned, Cell (i, j) is partially burned; this state judgment basis can be used when the burning cell is initialized. Compared with the judgment basis of complete combustion, this embodiment is more in line with the spreading law of forest fires with vegetation as the main burning material, and can improve the accuracy of forest fire simulation. When used as a basis for judging complete combustion, the method of this embodiment can also be used to complete fire simulation, but there is a difference in simulation accuracy.

[0081] In another specific embodiment, the state of the cell (i, j) at time t+1 is determined by the combustion state at time t. The burning state of the neighborhood cells at time t Current wind speed wind direction And the NDVI value of the cell location i,j and DEM value (height value)DEM i,j By joint decision, the relationship can be expressed as:

[0082]

[0083] when or NDVI k,l When , the burning state of the neighboring cell (k,l) will spread to the cell (i,j). Then the cell (i,j) is partially burned, if This means that the cell (i, j) is completely burned and begins to spread to the surrounding 8 cells. Before calculating the function, it is necessary to determine the NDVI i,j and DEM i,j , set the flammable threshold α and the differential slope threshold β, if NDVI i,j >α and DEM i+m,j+n -DEM i,j >β, where m and n are the direction vectors of the current wind speed and wind direction. The value of the NDVI vegetation index is set as the flammable probability. The cell has a certain probability of being ignited. If it is ignited, the current wind speed is considered. and wind direction Determine the direction and wind direction of cells (i,j) and cells (k,l) If it is in the downwind direction, the burning area growth rate v of the cell (i, j) is the original growth rate v0 multiplied by the projection of the wind speed in the spreading direction. times, that is If NDVI i,j <α or DEM i,j<β, then no function operation is required and the cell is non-flammable. Always 0.

[0084] The following is a specific application example to show the forest fire spread simulation results of the method of this embodiment. In this example, a mountain forest fire emergency response plan is established based on high-resolution remote sensing and three-dimensional geographic information technology, and the Juyongguan scenic area and the surrounding area are selected as the fire simulation area. The overall terrain of this area is high, with an altitude of 600 to 1000 meters, and the terrain is undulating; the landform is mainly hilly and mountainous, with magnificent peaks and canyons; the vegetation is mainly composed of forests and grasslands, and the mountainous areas are covered with dense coniferous and broad-leaved forests, such as pine trees, cypress trees, and locust trees; the region belongs to the warm temperate semi-humid continental climate, with distinct four seasons and an average annual temperature of about 10°C; the average annual precipitation is about 40mm to 600mm, mainly concentrated in summer, and winter is generally a dry season. There are many mountain springs and streams in the region, and water resources are abundant.

[0085] The simulation range is roughly between 116.01 and 116.142 east longitude and 40.24 and 40.343 north latitude, with an area of ​​about 130.64 square kilometers. The forest fire spread simulation data uses the cloudless Landsat 8 image of June 9, 2021 as the base map, and the simulated burning parameters use NDVI normalized vegetation index and DEM digital elevation model data, with a resolution of 30m. The vegetation distribution, meteorological data, terrain and other data are input into the model to define the spread speed of each cell. The relationship between the spread speed and the terrain slope and wind speed in the simulation process of this model is shown in Table 1. In the calculation, a spread speed is first determined according to the slope. On the basis of this spread speed, the spread speed corresponding to the wind speed is added to obtain the final spread speed. The actual situation will be affected by many factors. The specific value of the spread speed may vary depending on the region, model and fire characteristics. It is necessary to obtain a more accurate numerical relationship through specific field investigation and monitoring.

[0086] Table 1 Example of the relationship between forest fire spread speed, slope and wind speed

[0087]

[0088] In this example, multiple fire simulation analyses were conducted based on the characteristics of meteorological, topographical, and vegetation data, combined with the control variable method. The initial coordinates of the forest fire were set (116.079, 40.316). When the wind direction was south, the wind speed was 1 unit (1 unit wind speed was defined as level 2 wind). Each fire point spread one grid distance around as one time unit. According to the state transfer function of the cell, 12 sets of forest fire spread simulation result binary graphs were generated in 0-120 time units (unit: minutes), as shown in the figure below: Figure 5To better distinguish the forest fire spread area, the forest fire spread simulation results are superimposed with the Landsat8 image of the study area, and the forest fire spread area is set to red, as shown in Figure 2. Figure 6 The fire simulation results show that the heat and smoke released by the fire source spread along the south wind direction, causing the fire line to expand toward the south. Fire spread map ( Figure 5 ) can clearly show the location of the fire boundary and fire line, showing the burning status from the fire source to the fire boundary. The influence of the southerly wind caused the fire to spread rapidly to the south, which may pose a threat to residents and ecosystems in the southern region.

[0089] Furthermore, the spread of forest fires is affected by many factors and there are obvious differences in different time periods. By controlling variables, the spread of fires under different wind speeds, wind directions, terrains, and vegetation conditions is simulated. The results are shown in Table 2.

[0090] Table 2 Changes in forest fire spread area under the influence of different simulated parameters

[0091]

[0092] Note: The wind speeds "1, 2, 3" in the model represent "level 2 wind, level 4 wind, level 6 wind" respectively; "east, south, west, north" represent the four directions of wind respectively.

[0093] By comparing the simulation results under different scenarios or parameter settings (Table 2, Figure 7 ), we can further understand the factors affecting the spread of forest fires and their role in the development of fires. It can be seen that compared with the case of lower wind speed, the fire spread speed under higher wind speed conditions is significantly increased, resulting in rapid spread of fire. In addition, under steep slope terrain, the speed of fire spread on the line is significantly greater than that on gentle slope terrain, which shows that the slope has a significant impact on the spread of fire. At the same time, the spatial area spread simulation results also show that under the influence of different wind directions, the direction of fire spread is opposite to the direction of wind ( Figure 8-Figure 11 ), which is the same direction as the wind. This comparative analysis helps to identify the main influencing factors, provides important information for fire management and emergency response, and provides a scientific basis for formulating effective fire prevention strategies and resource allocation.

[0094] In summary, this embodiment provides a method for simulating the spread of forest fires based on burning probability, which simulates the spread trend of forest fires based on the cellular automaton model, and reflects the effect of wind on the spread of forest fires by supplementing neighborhood grids along the wind speed direction (the direction of the wind); reflects the comprehensive influence of vegetation, terrain and wind on the spread of forest fires by setting vegetation coverage rate thresholds and slope thresholds; and simulates the randomness of ignition of cells by random numbers in the interval [0,1]. In multiple ways, the accuracy of forest fire simulation is improved together.

[0095] In particular, when supplementing the neighborhood, this embodiment does not expand all neighborhood cells outward along the wind speed vector. This is because the simulation itself has certain errors. Whether the original Moore neighborhood cell is ignited is a probabilistic event, while the current burning cell has been burned, which is a given event. Only the extension of the current burning cell along the wind speed direction is retained as a supplement to the neighborhood, which can effectively avoid error diffusion.

[0096] At the same time, when setting the slope threshold, this embodiment matches a differentiated slope threshold for each cell according to the wind speed vector. The higher the wind speed, the lower the slope threshold; the smaller the angle between the position vector from the current burning cell to the current neighborhood cell and the wind speed vector, the lower the slope threshold; the greater the projection of the wind speed along the slope direction, the lower the slope threshold; thereby more accurately reflecting the impact of wind force and slope on different cells in the spread of forest fires, and improving the accuracy of forest fire simulation.

[0097] In addition, when judging complete combustion, this embodiment corrects the combustion state of the cell in combination with the vegetation coverage rate of the cell. When the Compared with the judgment basis of complete combustion, this embodiment is more in line with the fire spread law in the forest fire scene with vegetation as the main combustible material, and can improve the accuracy of forest fire simulation.

[0098] Fig.12 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention is shown in FIG. Fig.12 As shown, the device includes a processor 60, a memory 61, an input device 62 and an output device 63; the number of processors 60 in the device can be one or more. Fig.12 A processor 60 is taken as an example; the processor 60, the memory 61, the input device 62 and the output device 63 in the device can be connected by a bus or other means. Fig.12 The example of connecting through bus is taken in the following.

[0099] The memory 61 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the forest fire spread simulation method based on burning probability in the embodiment of the present invention. The processor 60 executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory 61, that is, realizing the above-mentioned forest fire spread simulation method based on burning probability.

[0100] The memory 61 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 61 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 61 may further include a memory remotely arranged relative to the processor 60, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0101] The input device 62 may be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device 63 may include a display device such as a display screen.

[0102] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the forest fire spread simulation method based on combustion probability of any embodiment.

[0103] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, a device or a device or used in combination with it.

[0104] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0105] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0106] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for simulating forest fire spread based on combustion probability, characterized in that: include: Obtaining the initial position of the forest fire, and initializing the combustion cells in the automatic cellular machine model according to the initial position; In each time step of the simulation, each combustion cell is traversed and the following operations are performed on each combustion cell in turn: S1. Determine the neighboring cells adjacent to the current burning cell, and draw a wind speed vector with the current burning cell as the starting point, and use the cells passed by the wind speed vector as supplementary neighboring cells, wherein the direction of the wind speed vector is consistent with the wind direction, and the length is proportional to the wind speed; S2. For all the unburned neighboring cells in all the neighboring cells, perform the following operations in sequence: S2-1, when the vegetation coverage rate of the unburned current neighborhood cell is greater than or equal to the burning threshold, matching the slope threshold corresponding to the current neighborhood cell according to the wind speed vector, wherein the slope threshold is less than 0, and the higher the wind speed, and / or the smaller the angle between the position vector from the current burning cell to the current neighborhood cell and the wind speed vector, the lower the slope threshold; S2-2. When the slope of the current neighborhood cell is greater than or equal to the slope threshold, generate a random number in the interval [0,1]; if the random number is less than or equal to the vegetation coverage rate, ignite the current neighborhood cell.

2. The method according to claim 1, characterized in that The traversing each combustion cell in each time step of the simulation includes: The time it takes for the burning cell to diffuse in all directions with a cell side length in a windless and slopeless state is taken as the time step of the simulation.

3. The method according to claim 1, characterized in that The drawing of the wind speed vector with the current combustion cell as the starting point includes: According to the cell side length and wind direction, the wind speed is converted into the number of cells passed per unit time step; According to the number of cells, a wind speed vector starting from the current burning cell is drawn.

4. The method according to claim 1, characterized in that: The following operations are performed in sequence on each unburned neighboring cell in all neighboring cells, including: When there are multiple layers of neighborhood cells, the operations S2-1 and S2-2 are performed on each unburned neighborhood cell in the first layer of neighborhood cells respectively; Each ignited neighboring cell in the first layer of neighboring cells is used as a new current burning cell, and the operations S2-1 and S2-2 are performed on each unburned neighboring cell adjacent to each new current cell in the second layer of neighboring cells; This process spreads outward layer by layer until all neighboring cells in all layers are executed.

5. The method according to claim 1, characterized in that The matching the slope threshold corresponding to the current neighborhood cell according to the wind speed vector includes: Match the slope threshold corresponding to the current neighborhood cell according to the following formula: Among them, slope T Represents the slope threshold of the current neighborhood cell (i2, j2), where (i2, j2) is the cell coordinate; represents the wind speed vector, represents the slope vector pointing from the three-dimensional geographic location (i, j, h1) of the current burning cell to the three-dimensional geographic location (i2, j2, h2) of the current neighboring cell; represents the dot product of two vectors, is the projection of the wind speed vector along the slope vector; a and b are pre-calibrated values.

6. The method according to claim 1, characterized in that The matching the slope threshold corresponding to the current neighborhood cell according to the wind speed vector includes: Match the slope threshold corresponding to the current neighborhood cell according to the following formula: Among them, slope T Represents the slope threshold of the current neighborhood cell (i2, j2), where (i2, j2) is the cell coordinate; represents the modulus of the wind speed vector, i.e., the wind speed; α represents the angle between the plane position vector from the current combustion cell (i, j) to the current neighborhood cell (i2, j2) and the wind speed vector; a and b are pre-calibrated values.

7. The method according to claim 1, characterized in that S2 also includes: When the vegetation coverage of the unburned current neighborhood cell is less than the burning threshold, or the slope is less than the matching slope threshold, or the random number is greater than the vegetation coverage, the state of the current neighborhood cell is kept unchanged.

8. The method according to claim 1, characterized in that S1 includes: When the state of the current burning cell is partially burned, the newly added burning area of ​​the current burning cell in the current time step is calculated according to the forest fire spreading speed; when the ratio of the accumulated burning area of ​​the current burning cell to the cell area reaches the vegetation coverage rate of the current burning cell, the state of the current burning cell is marked as completely burned; when the state of the current burning cell is completely burned, the neighboring cells adjacent to the current burning cell are determined, and the wind speed vector is drawn with the current burning cell as the starting point, and the cells passed by the wind speed vector are used as supplementary neighboring cells; The igniting of the current neighborhood cell in S2-2 includes: igniting the current neighborhood cell and marking the state of the current neighborhood cell as partially burned; The newly added burned area of ​​the current neighborhood cell in the remaining time of the current time step is calculated according to the forest fire spread speed. When the ratio of the accumulated burned area of ​​the current neighborhood cell to the cell area reaches the vegetation coverage rate of the current neighborhood cell, the state of the current neighborhood cell is marked as completely burned.

9. An electronic device, characterized in that: include: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the forest fire spread simulation method based on combustion probability as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the program is executed by a processor, the method for simulating forest fire spread based on combustion probability as described in any one of claims 1 to 8 is implemented.

Images