Mountain fire diffusion prediction method, system, equipment, medium and product
By combining convolutional neural networks and cellular automata, the error problem of existing models when considering terrain slope and environmental factors is solved, achieving more accurate wildfire spread prediction and drawing detailed spread time distribution maps.
Patent Information
- Application Number
- CN202511067072.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-31
- Publication Date
- 2025-11-07
AI Technical Summary
Existing wildfire spread models have errors when considering terrain slope and environmental factors, and the influence range of single neural network models is expanded, resulting in insufficient prediction accuracy.
By combining convolutional neural networks and cellular automata, the convolutional neural network is initialized using a training dataset to generate functional relationships, cellular automata are created to calculate the wildfire spread time, and a spread time distribution map is plotted.
It improves the accuracy of wildfire spread prediction, fully considers the complex coupling effect of geographical and environmental parameters, reduces simulation errors, and achieves more accurate simulation of wildfire combustion status.
Smart Images
Figure CN120911287A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of mountain fire spread, more particularly, it relates to a mountain fire spread prediction method, system, device, medium and product. BACKGROUND
[0002] Mountain fire spread simulation is a common means of mountain fire disaster risk assessment, and the simulation results provide scheme planning decision support for delaying and stopping mountain fire spread.
[0003] Common mountain fire spread models are generally divided into experimental-based mathematical models, and the experimental conditions of the mathematical models have weak coverage, and the consideration of terrain slope is insufficient, and the simulation error is large under certain slope conditions, and the model needs to be modified to adapt to the influence of slope factors; or the empirical model based on statistics, the data has strong regional dependence, not suitable for global promotion, the parameters need to be adjusted to adapt to the specific situation.
[0004] Cellular automata is a kind of spatiotemporal discrete local dynamics model, which is a typical method for complex system research, especially suitable for spatiotemporal dynamic simulation research of spatial complex system, and can integrate various complex environmental variables such as terrain, atmosphere, vegetation, human activities and other environmental variables in the complex system evolution simulation process. However, the current cellular automata method more looks at each environmental variable as an independent variable, and simplifies the coupling and dynamic action process of the environmental variables; at the same time, a single neural network can use time series-based observation values to better fit the comprehensive influence function of each factor on mountain fire spread, but its influence range is easy to expand to the global, which violates the correlation principle, that is, the closer the geographical things are, the more similar they are, and the farther the geographical things are, the lower the similarity is, so the environmental factors of two points far away from each other have different influences on mountain fire. Therefore, the separate use of the two methods will lead to insufficient accuracy and precision of the forest fire spread model. SUMMARY
[0005] The purpose of the present application is to provide a mountain fire spread prediction method, system, device, medium and product, which solves the problem of insufficient mountain fire spread prediction precision caused by insufficient consideration of geographical elements by the existing cellular automata.
[0006] The above technical purpose of the present application is realized by the following technical scheme: In a first aspect of the present application, a mountain fire spread prediction method is provided, the method comprising: obtaining the occurrence time of historical mountain fire, the grid area where the current mountain fire may spread, and the geographical parameters and environmental parameters of the grid area; The occurrence time, geographical parameters and environmental parameters of historical forest fires are gridded according to the grid size of the grid area to obtain a training data set containing the occurrence time, geographical parameters, environmental parameters and grid coordinates; The initialized convolutional neural network is trained based on the training data set, and when the training is completed, a function relationship set is output ; A cellular automaton is created with the function relationship and the center coordinates of the cellular grid to calculate the grid burning state; The cellular automaton is used to calculate the forest fire spreading time according to the meteorological parameters and terrain data of the grid area; According to the coordinates of the cellular grid and the forest fire spreading time, a forest fire spreading time distribution map is drawn.
[0007] In an implementation scheme, the grid area where the forest fire can spread is obtained by: dividing the area where the forest fire can spread into multiple grid units of equal length, and taking the multiple grid units as the grid area where the forest fire can spread.
[0008] In an implementation scheme, the function relationship is: where t represents the occurrence time of the historical forest fire, represents the vegetation coverage, represents the grid center elevation value, represents the wind speed, represents the wind direction, represents the temperature, represents the humidity, represents the road coverage, represents the water coverage, x represents the grid horizontal coordinate, and y represents the grid vertical coordinate, represents the set of all grid internal propagation speeds.
[0009] In an implementation scheme, the cellular automaton is used to calculate the spreading time of each cellular grid according to the meteorological parameters and terrain data of the grid area, including: Taking a single cell as the center point, the time when the cellular grid starts to burn is obtained according to the function relationship, and the burning state of the current cellular grid is calculated with the time as the starting time; When the burning state is greater than a preset threshold, the spreading speed of the adjacent cellular grid in eight directions of the current cellular grid is calculated according to the geographical parameters and environmental parameters, and the burning time of each direction adjacent cellular grid is calculated according to the spreading speed; The calculation of the burning time of the adjacent cellular grid is repeated until the state of any cellular grid is changed from the burning state to the stop burning state, and the iteration is stopped to obtain the forest fire spreading time.
[0010] In a second aspect, the present application provides a forest fire spreading prediction system, the system comprising: a data acquisition module configured to acquire occurrence time of historical forest fires, grid regions where the current forest fire is likely to spread, and geographical parameters and environmental parameters of the grid regions; a data set construction module configured to rasterize the occurrence time, geographical parameters and environmental parameters of the historical forest fires according to the grid size of the grid regions to obtain a training data set containing occurrence time, geographical parameters, environmental parameters and grid coordinates; a function determination module configured to train an initialized convolutional neural network based on the training data set, and output a function relationship when the training is completed; a cellular automata component module configured to create a cellular automaton based on the burning start time, the burning end time, the function relationship and the coordinates of the center of the cellular grid; a spreading time calculation module configured to calculate the forest fire spreading time based on the meteorological parameters and terrain data of the grid regions by using the cellular automaton; a forest fire spreading prediction module configured to draw a forest fire spreading time distribution map based on the coordinates of the cellular grid and the forest fire spreading time.
[0011] In an implementation scheme, the region where the forest fire is likely to spread is divided into a plurality of grid units of equal length, and the plurality of grid units are taken as the grid regions where the forest fire is likely to spread.
[0012] In an implementation scheme, the function relationship is: wherein t represents the occurrence time of the historical forest fires, represents the vegetation coverage, represents the grid center elevation value, represents the wind speed, represents the wind direction, represents the temperature, represents the humidity, represents the road coverage, represents the water coverage, x represents the grid horizontal coordinate, and y represents the grid vertical coordinate, represents the set of all grid internal propagation speeds.
[0013] In a third aspect, the present application provides an electronic device comprising a memory and a processor; the memory is configured to store a computer program, the computer program comprising program instructions; the processor is configured to execute the program instructions to enable the electronic device to perform the steps of the forest fire spreading prediction method according to the first aspect of the present application.
[0014] In a fourth aspect, the present application provides a computer program product comprising program instructions which, when executed by an electronic device, cause the electronic device to perform the steps of the method for predicting forest fire spread according to the first aspect of the present application.
[0015] In a fifth aspect, the present application provides a computer-readable storage medium comprising a computer program which, when executed by one or more processors, implements the method for predicting forest fire spread according to the first aspect of the present application.
[0016] Compared with the prior art, the present application has the following beneficial effects: The present application takes into account the influence of geographical parameters and environmental parameters (such as factors such as terrain slope, weather, vegetation, etc.) on the initialization state of the cell, improves the prediction accuracy of forest fire spread, and first uses a convolutional neural network to calculate the influence of each geographical factor on the burning state of the cell when calculating the propagation time, more fully considers the complex coupling between geographical factors, and at the same time considers the bidirectional propagation between cells, which weakens the one-way propagation setting when the conventional convolutional neural network model predicts forest fires. At the same time, the state function of each cell is trained through the convolutional neural network, so that the cell can simulate the real burning situation of the forest fire, that is, the cell has a burning period and a propagation period, which avoids the simulation error caused by the grid discretization simplification of geographical factors. BRIEF DESCRIPTION OF DRAWINGS
[0017] The drawings described herein are used to provide further understanding of the embodiments of the present application, form a part of the present application, and do not constitute a limitation on the embodiments of the present application. In the drawings: Figure 1 A flowchart of a method for predicting forest fire spread according to an embodiment of the present application is provided. Figure 2 An algorithm flowchart of a method for predicting forest fire spread according to an embodiment of the present application is provided. Figure 3 A structural block diagram of a forest fire spread prediction system according to an embodiment of the present application is provided. DETAILED DESCRIPTION
[0018] In order to make the purpose, technical scheme and advantages of the present application clearer and more apparent, the present application will be further described in detail below with reference to the embodiments and drawings, and the illustrative embodiments of the present application and their descriptions are only used to explain the present application and do not constitute a limitation on the present application.
[0019] It should be noted that the terms "comprising" or "may include" used in the various embodiments of this application indicate the presence of the claimed function, operation, or element, and do not limit the addition of one or more functions, operations, or elements. Furthermore, as used in the various embodiments of this application, the terms "comprising," "having," and their cognates are intended only to indicate a specific feature, number, step, operation, element, component, or combination of the foregoing, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing, or adding one or more combinations of the foregoing.
[0020] It should be understood that terms such as "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.
[0021] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a wildfire spread prediction method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes: S101 obtains the occurrence time of historical wildfires, the grid area where the current wildfire may spread, and the geographical and environmental parameters of the grid area.
[0022] In this embodiment, based on the terrain data of the potential wildfire spread area, the grid area where the wildfire may spread is divided into multiple grid cells of equal length, and these multiple grid cells are used as the grid area where the wildfire may spread. Specifically: set up Given the length and width of the potential diffusion area, it is divided into sections with equal length and width. The set of grids, denoted as : (1), where , These represent the number of grid divisions in the diffusion region, respectively. This represents the cell with row and column index (n, m).
[0023] Geographical parameters include road networks, rivers, lakes, terrain grids, vegetation cover, etc. within the grid area, while environmental parameters include wind speed, wind direction, temperature, humidity, etc. within the grid area. These are all conventional parameters in this technical field, so this embodiment will not describe them in detail.
[0024] S102, the occurrence time, geographical parameters and environmental parameters of historical forest fires are rasterized according to the grid size of the grid area to obtain a training data set containing occurrence time, geographical parameters, environmental parameters and grid coordinates.
[0025] In the embodiment, DEM elevation data, vegetation cover data, wind speed, wind direction, temperature, humidity, road network and river and lake data corresponding to the period are selected according to historical forest fires, and are rasterized according to a uniform cell size to obtain a training data set. According to the proportion of 8:2, the training data set is divided into a training set and a test set .
[0026] S103, training the initialized convolutional neural network based on the training data set, and outputting a functional relationship when the training is completed.
[0027] In the embodiment, the initialized convolutional neural network provided is defined based on the UNet framework. Since the framework is a conventional technical means at present, the initialized convolutional neural network is trained by using the training set and the test set divided by the training data set, and the existing neural network training method is completed. In the case, a common Dice Loss + ClassWeight loss function combination and Radam + Lookahead acceleration convergence are used, and the embodiment does not make a detailed description.
[0028] The functional relationship is: , wherein t represents the occurrence time of the historical forest fire, represents the vegetation coverage, represents the grid center elevation value, represents the wind speed, represents the wind direction, represents the temperature, represents the humidity, represents the road coverage, represents the water body coverage, x represents the grid horizontal coordinate, and y represents the grid vertical coordinate, represents the set of all grid internal propagation speeds.
[0029] S104, creating a cellular automaton with the burning start time, the burning end time, the functional relationship and the cell grid center coordinates.
[0030] In the embodiment, each element in the initialized cell is defined by a tuple to represent the state, which is represented as: (2), wherein represents the burning start time; represents the burning end time; represents a cell state function; represents a cell grid center coordinate.
[0031] S105, using a cellular automaton, calculates the mountain fire spreading time according to the meteorological parameters and terrain data of the grid area.
[0032] In this embodiment, using a cellular automaton, the spreading time of each cell grid is calculated according to the meteorological parameters and terrain data of the grid area, including: S1051, taking a single cell fire as the center point, the time when the cell grid starts to burn is obtained according to the functional relationship, and the burning state of the current cell grid is calculated with this time as the starting time.
[0033] S1052, when the burning state is greater than a preset threshold, the spreading speed of the adjacent cell grid in eight directions of the current cell grid is calculated according to the geographical parameters and environmental parameters, and the burning time of each direction adjacent cell grid is calculated according to the spreading speed.
[0034] S1053, repeat the calculation of the burning time of the adjacent cell grid until the state of any cell grid is changed from the burning state to the stop burning, stop iteration, and get the mountain fire spreading time.
[0035] Specifically, according to the aforementioned functional relationship the starting burning time of the cell grid is obtained , and the burning state of is calculated with this time as the starting time, when , that is, the overfire area ratio of exceeds 0.75 of the total area of the cell, the cell starts to have the ability to spread; wherein, represents the burning area inside the grid, according to the isotropy of the cell grid, the calculation method is: , that is, the spreading speed learned by the UNet neural network, is the center point coordinate of the grid, represents the area of the grid, and its value is , is the side length of the grid.
[0036] According to the terrain parameters such as slope, aspect, wind speed, wind direction, vegetation coverage, humidity, temperature, etc. and the spreading speed of its eight adjacent in eight directions is calculated, and the burning time of its eight adjacent in eight directions is calculated respectively. The propagation speed between grids is calculated by the following empirical formula: ; wherein: is the propagation speed inside the grid, , is the current grid center point coordinate; is the combustible indicator coefficient normalized representation, which can be obtained by looking up a table; is the slope type indicator, 1 for uphill and -1 for downhill along the propagation direction; is the slope, ranging from 0° to 90°; is the wind speed, unit m / s; is the angle between the wind direction and the propagation direction, ranging from 0° to 360°.
[0037] S106, according to the coordinates of the cellular grid and the mountain fire spreading time, a mountain fire spreading time distribution map is drawn.
[0038] With the coordinate point longitude and latitude as the coordinate parameters and the mountain fire spreading time as the attribute value, a discrete point vector data is generated, then an ordinary Kriging interpolation algorithm is used to generate an interpolation map, finally the interpolation map result is exported as a vector data, and the discretized spreading time result is converted into a continuous value.
[0039] Please refer to Figure 2 , Figure 2 is a structure block diagram of a mountain fire spreading prediction system provided by the embodiment of the present application, as shown in Figure 2 , the system comprises: a data acquisition module 210, configured to acquire the occurrence time of historical mountain fires, the grid area where the current mountain fire is likely to spread, and the geographical parameters and environmental parameters of the grid area; a data set construction module 220, configured to rasterize the occurrence time of historical mountain fires, the geographical parameters and the environmental parameters according to the grid size of the grid area, to obtain a training data set containing the occurrence time, the geographical parameters, the environmental parameters and the grid coordinates; a function determination module 230, configured to train the initialized convolutional neural network based on the training data set, and output a function relationship when the training is completed; a cellular automaton component module 240, configured to create a cellular automaton with the burning start time, the burning end time, the function relationship and the cellular grid center coordinates; a spreading time calculation module 250, configured to calculate the mountain fire spreading time according to the meteorological parameters and the terrain data of the grid area by using the cellular automaton; a mountain fire spreading prediction module 260, configured to draw a mountain fire spreading time distribution map according to the coordinates of the cellular grid and the mountain fire spreading time.
[0040] Correspondingly, the mountain fire spread prediction system provided by the embodiment considers the influence of geographical parameters and environmental parameters (such as factors such as terrain slope, meteorology, vegetation, etc.) on the initialization state of the cell, improves the prediction accuracy of the mountain fire spread, and first uses a convolutional neural network to calculate the influence of each geographical factor on the burning state of the cell when calculating the propagation time, more fully considers the complex coupling between geographical factors, and simultaneously considers the bidirectional propagation between cells, thereby weakening the one-way propagation setting when the conventional convolutional neural network model predicts the mountain fire. At the same time, the state function of each cell is trained through the convolutional neural network, so that the cell can simulate the real burning situation of the mountain fire, that is, the cell has a burning period and a propagation period, thereby avoiding the simulation error caused by the grid discretization simplification of the geographical factors.
[0041] In an embodiment, the possible spread area of the mountain fire is divided into a plurality of grid units of equal length, and the plurality of grid units are used as the grid area where the mountain fire can spread.
[0042] In an embodiment, the function relationship is: wherein t represents the occurrence time of the historical mountain fire, represents the vegetation coverage, represents the grid center elevation value, represents the wind speed, represents the wind direction, represents the temperature, represents the humidity, represents the road coverage, represents the water coverage, x represents the grid horizontal coordinate, and y represents the grid vertical coordinate, represents the set of all internal propagation speeds of the grid.
[0043] The mountain fire spread prediction system in the embodiment of the application and the mountain fire spread prediction method shown in the above Figure 1 The mountain fire spread prediction method is a technical solution based on the same inventive concept, and through the detailed description of the mountain fire spread prediction method provided by the above embodiment, a person skilled in the art can clearly understand the implementation process of each module of the mountain fire spread prediction system in the embodiment, so as to avoid repetition.
[0044] The embodiment of the application further provides an electronic device. The electronic device includes a processor, a memory, a communication interface, and at least one communication bus for connecting the processor, the memory, and the communication interface. The memory includes, but is not limited to, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (PROM), or a portable read-only memory (CD-ROM), and is used for related instructions and data.
[0045] The communication interface is configured to receive and send data. The processor can be one or more CPUs, and in the case of one CPU, the CPU can be a single-core CPU or a multi-core CPU. The processor in the electronic device is configured to read one or more programs stored in the memory and perform the following operations: obtain a grid area where a wildfire can spread, geographical parameters and environmental parameters at the time of occurrence of a historical wildfire; rasterize the occurrence time, geographical parameters and environmental parameters of the historical wildfire according to the grid size of the grid area to obtain a training data set containing the occurrence time, geographical parameters, environmental parameters and grid coordinates; train an initialized convolutional neural network based on the training data set, and output a function relationship when the training is completed; create a cellular automaton based on the start time of combustion, the end time of combustion, the function relationship and the coordinates of the center of the cellular grid; calculate the wildfire spread time according to the meteorological parameters and the terrain data of the grid area by using the cellular automaton; and draw a wildfire spread time distribution map according to the coordinates of the cellular grid and the wildfire spread time.
[0046] It should be noted that the specific implementation of each operation can be described above with respect to the method embodiments Figure 1 It should be noted that the specific implementation of each operation can be described above with respect to the method embodiments
[0047] The embodiment of the present application further provides a computer readable storage medium, which is a memory device in a computer device and is used for storing programs and data. It can be understood that the computer readable storage medium herein can include an internal storage medium in the computer device and can also include an extended storage medium supported by the computer device. The computer readable storage medium provides a storage space, which stores an operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium herein can be a high-speed RAM memory or a non-volatile memory such as at least one disk memory. The one or more instructions stored in the computer readable storage medium can be loaded and executed by the processor to implement the corresponding steps of the forest fire spreading prediction method in the above embodiment. It should be understood by those skilled in the art that the embodiment of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can be in the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. In addition, the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.
[0048] The embodiment of the present application further provides a computer program product containing program instructions. The computer program product can be a software or program product containing program instructions, which can be run on a computing device or stored in any available medium. When the computer program product is run on at least one electronic device, the at least one electronic device is caused to perform a forest fire spreading prediction method.
[0049] The above specific embodiments further explain the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims
1. A method of predicting the spread of a bushfire, characterised by, The method comprises: acquiring the occurrence time of historical forest fires, the grid area where the current forest fire is likely to spread, and geographical parameters and environmental parameters of the grid area; rasterizing the occurrence time, geographical parameters and environmental parameters of historical forest fires according to the grid size of the grid area to obtain a training data set containing occurrence time, geographical parameters, environmental parameters and grid coordinates; training an initialized convolutional neural network based on the training data set, and outputting a function relationship set when the training is completed; creating a cellular automaton with the burning start time, the burning end time, the function relationship and the center coordinates of the cellular grid; calculating the forest fire spread time according to the meteorological parameters and terrain data of the grid area by using the cellular automaton; drawing a forest fire spread time distribution map according to the coordinates of the cellular grid and the forest fire spread time.
2. A method of predicting the spread of a bushfire according to claim 1, characterised in that, The grid area where the forest fire is likely to spread comprises: dividing the area where the forest fire is likely to spread into a plurality of grid units of equal length, and taking the plurality of grid units as the grid area where the forest fire is likely to spread.
3. The method of claim 1, wherein, The function relationship is: where t represents the occurrence time of historical bushfires, represents the vegetation coverage, represents the grid center elevation value, represents the wind speed, represents the wind direction, represents the temperature, represents the humidity, represents the road coverage, represents the water coverage, x represents the grid horizontal coordinate, y represents the grid vertical coordinate, represents the set of all grid internal propagation speeds.
4. The method of claim 1, wherein, calculating the spread time of each cellular grid according to the meteorological parameters and terrain data of the grid area by using the cellular automaton, comprising: taking a single cellular fire as a center point, obtaining the time when the cellular grid starts to burn according to the function relationship, taking the time as the start time, and calculating the burning state of the current cellular grid; when the burning state is greater than a preset threshold, calculating the spread speed of the adjacent cellular grid in eight directions of the current cellular grid according to the geographical parameters and environmental parameters, and calculating the burning time of each direction adjacent cellular grid according to the spread speed; repeating the calculation of the burning time of the adjacent cellular grid until the state of any cellular grid is changed from the burning state to the stop burning state, stopping iteration, and obtaining the forest fire spread time.
5. A bushfire spread prediction system characterised by, The system comprises: a data acquisition module for acquiring the occurrence time of historical forest fires, the grid area where the current forest fire is likely to spread, and geographical parameters and environmental parameters of the grid area; a data set construction module for rasterizing the occurrence time, geographical parameters and environmental parameters of historical forest fires according to the grid size of the grid area to obtain a training data set containing occurrence time, geographical parameters, environmental parameters and grid coordinates; a function determination module for training an initialized convolutional neural network based on the training data set, and outputting a function relationship when the training is completed; a cellular automaton construction module for creating a cellular automaton with the burning start time, the burning end time, the function relationship and the center coordinates of the cellular grid; a spread time calculation module for calculating the forest fire spread time according to the meteorological parameters and terrain data of the grid area by using the cellular automaton; a forest fire spread prediction module for drawing a forest fire spread time distribution map according to the coordinates of the cellular grid and the forest fire spread time.
6. A bushfire spread prediction system according to claim 5, characterised in that, The grid area where the forest fire is likely to spread comprises: dividing the area where the forest fire is likely to spread into a plurality of grid units of equal length, and taking the plurality of grid units as the grid area where the forest fire is likely to spread.
7. A bushfire spread prediction system according to claim 5, characterised in that, The function relationship is: where t represents the occurrence time of historical bushfires, represents the vegetation coverage, represents the grid center elevation value, represents the wind speed, represents the wind direction, represents the temperature, represents the humidity, represents the road coverage, represents the water body coverage, x represents the grid horizontal coordinate, and y represents the grid vertical coordinate, represents the set of all grid internal propagation speeds.
8. An electronic device, comprising: comprising a memory and a processor; the memory is used to store a computer program, and the computer program comprises program instructions; A processor configured to execute the program instructions to cause the electronic device to perform the steps of the method for predicting the spread of a wildfire according to any one of claims 1 to 4.
9. A computer program product comprising program instructions, characterized in that, The program instructions, when executed by an electronic device, cause the electronic device to perform the steps of the method for predicting the spread of a wildfire according to any one of claims 1 to 4.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium comprises a computer program which, when executed by one or more processors, implements the method for predicting the spread of a wildfire according to any one of claims 1 to 4.