Forest fire spreading trend prediction method, storage medium and equipment

Abstract

The embodiment of the invention discloses a forest fire spreading trend prediction method, a storage medium and equipment, and the method comprises the steps: obtaining a plurality of pieces of historical forest fire data in a preset region range and historical forest fire spreading trend conditions corresponding to the plurality of pieces of historical forest fire data, and then employing a preset CFD model, according to each piece of historical forest fire data and each piece of historical forest fire spreading trend condition, historical forest fire spreading trend simulation data corresponding to each piece of historical forest fire data is generated, and then an initial machine learning model is constructed; and finally, the multiple pieces of historical forest fire data and the multiple pieces of historical forest fire spreading trend simulation data are sequentially input into the initial machine learning model for training, a target machine learning model is obtained, and the target machine learning model is used for predicting forest fire spreading trend detailed conditions in a preset area range. The combined prediction of the forest fire spreading trend is carried out by adopting the CFD model and the machine learning model, so that the technical problems existing in the prior art can be effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, storage medium and device for predicting the spread trend of wildfires. Background Art

[0002] In recent years, with the in-depth development of power resources, more and more transmission lines are crossing complex mountainous areas. The unique topography and climatic conditions in these areas are very likely to cause wildfires, which can cause transmission lines to trip or even burn out towers, causing long-term, irreversible major power accidents. Therefore, wildfires have become one of the main factors affecting the safe and stable operation of transmission lines. With the continuous expansion of the scale of the power grid and the frequent occurrence of extreme drought weather in recent years, the threat of wildfires to the safety of the power grid has become increasingly severe.

[0003] In order to effectively respond to this challenge, accurate analysis of wildfire spread trends to predict wildfire-prone areas and key regions is of great significance for timely and accurate detection of wildfires and their rapid extinguishing. Summary of the invention

[0004] Based on this, it is necessary to propose a wildfire spread trend prediction method, storage medium and equipment to address the above problems. By adopting CFD model and machine learning model to perform combined prediction of wildfire spread trend, it is possible to accurately analyze the wildfire spread trend, predict wildfire high-incidence areas and key areas, and then accurately detect wildfires and quickly extinguish them, thereby effectively solving the technical problems existing in the existing technology.

[0005] To achieve the above object, the present invention provides a method for predicting wildfire spread trend in a first aspect, the method comprising:

[0006] Obtaining multiple historical wildfire data within a preset area and historical wildfire spreading trends corresponding to the multiple historical wildfire data;

[0007] Using a preset CFD model, based on each historical wildfire data and each historical wildfire spread trend, generate historical wildfire spread trend simulation data corresponding to each historical wildfire data;

[0008] Build an initial machine learning model;

[0009] Multiple historical wildfire data and multiple historical wildfire spread trend simulation data are sequentially input into the initial machine learning model for training to obtain a target machine learning model, which is used to predict the details of the wildfire spread trend in the preset area.

[0010] Optionally, the use of a preset CFD model to generate historical wildfire spread trend simulation data corresponding to each historical wildfire data according to each historical wildfire data and each historical wildfire spread trend situation includes:

[0011] According to the nth historical wildfire data, the model parameters of the preset CFD model are adjusted to obtain the nth target CFD model;

[0012] Simulate the nth target CFD model to generate nth initial wildfire spread simulation data;

[0013] Determine whether the nth target CFD model needs to be optimized based on the nth initial wildfire spread simulation data and the nth historical wildfire spread trend;

[0014] When it is determined that there is no need to optimize the nth target CFD model, the nth initial wildfire spread simulation data is used as the historical wildfire spread trend simulation data corresponding to the nth historical wildfire data;

[0015] Among them, n takes integers greater than 0 in sequence until n is equal to the total number of historical wildfire data.

[0016] Optionally, adjusting the model parameters of the preset CFD model according to the nth historical wildfire data to obtain the nth target CFD model includes:

[0017] Constructing a preset geometric model of the preset area;

[0018] Constructing the preset CFD model of the preset geometric model, and setting initial conditions and boundary conditions of the preset CFD model;

[0019] According to the nth historical wildfire data, the model parameters of the preset geometric model are adjusted to obtain the nth target geometric model;

[0020] According to the nth target geometric model, the model parameters of the preset CFD model are adjusted to obtain the nth target CFD model.

[0021] Optionally, constructing a preset geometric model of the preset area range includes:

[0022] Acquire regional data of the preset regional range;

[0023] The preset geometric model is constructed according to the area data.

[0024] Optionally, the method further comprises:

[0025] When it is determined that the nth target CFD model needs to be optimized, adjusting the model parameters and boundary conditions of the nth target CFD model to obtain an adjusted nth target CFD model;

[0026] Simulating the adjusted nth target CFD model to generate adjusted nth initial wildfire spread simulation data;

[0027] Determine whether the adjusted nth target CFD model needs to be optimized based on the adjusted nth initial wildfire spread simulation data and the nth historical wildfire spread trend;

[0028] When it is determined that there is no need to optimize the adjusted nth target CFD model, the adjusted nth initial wildfire spread simulation data is used as the historical wildfire spread trend simulation data corresponding to the nth historical wildfire data;

[0029] When it is determined that the adjusted nth target CFD model needs to be optimized, the adjustment and simulation are continued until optimization is no longer necessary.

[0030] Optionally, adjusting the model parameters and boundary conditions of the nth target CFD model to obtain the adjusted nth target CFD model includes:

[0031] Using an optimization algorithm, searching for the best combination parameters between each parameter of the model parameters of the nth target CFD model and each condition in the boundary conditions;

[0032] According to the optimal combination parameters, the model parameters and boundary conditions of the nth target CFD model are adjusted to obtain the adjusted nth target CFD model.

[0033] Optionally, determining whether to optimize the nth target CFD model according to the nth initial wildfire spread simulation data and the nth historical wildfire spread trend includes:

[0034] In the case where the nth historical wildfire spread trend situation belongs to the nth initial wildfire spread simulation data, it is determined that there is no need to optimize the nth target CFD model;

[0035] In the case where the nth historical wildfire spread trend situation does not belong to the nth initial wildfire spread simulation data, it is determined that the nth target CFD model needs to be optimized;

[0036] According to the adjusted nth initial wildfire spread simulation data and the nth historical wildfire spread trend, it is determined whether the adjusted nth target CFD model needs to be optimized, including:

[0037] In the case where the nth historical wildfire spread trend situation belongs to the adjusted nth initial wildfire spread simulation data, it is determined that there is no need to optimize the adjusted nth target CFD model;

[0038] When the nth historical wildfire spread trend situation does not belong to the adjusted nth initial wildfire spread simulation data, it is determined that the adjusted nth target CFD model needs to be optimized.

[0039] Optionally, simulating the nth target CFD model to generate nth initial wildfire spread simulation data includes:

[0040] Use different types of combustion models and radiation models to simulate the nth target CFD model and generate wildfire spread simulation data corresponding to different types of combustion models and radiation models;

[0041] The wildfire spread simulation data corresponding to different types of combustion models and radiation models are used as the nth initial wildfire spread simulation data;

[0042] The adjusted nth target CFD model is simulated to generate adjusted nth initial wildfire spread simulation data, including:

[0043] Using different types of combustion models and radiation models, simulating the adjusted nth target CFD model, and generating wildfire spread simulation data corresponding to different types of combustion models and radiation models;

[0044] The wildfire spread simulation data corresponding to different types of combustion models and radiation models are used as the nth initial wildfire spread simulation data.

[0045] To achieve the above object, the present invention provides a wildfire spread trend prediction device in a second aspect, the device comprising:

[0046] An acquisition module, used to acquire a plurality of historical wildfire data within a preset area and the historical wildfire spreading trend corresponding to the plurality of historical wildfire data;

[0047] A generation module, for generating historical wildfire spread trend simulation data corresponding to each historical wildfire data, using a preset CFD model according to each historical wildfire data and each historical wildfire spread trend;

[0048] Building blocks for constructing initial machine learning models;

[0049] The training prediction module is used to input multiple historical wildfire data and multiple historical wildfire spread trend simulation data into the initial machine learning model in sequence for training to obtain a target machine learning model. The target machine learning model is used to predict the details of the wildfire spread trend in a preset area.

[0050] To achieve the above-mentioned object, the present invention provides in a third aspect a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the method as described in any one of the first aspects.

[0051] To achieve the above-mentioned purpose, the present invention provides a computer device in a fourth aspect, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the method as described in any one of the first aspects.

[0052] The embodiment of the present invention has the following beneficial effects: the above method obtains multiple historical wildfire data in a preset area and the historical wildfire spread trends corresponding to the multiple historical wildfire data, and then uses a preset CFD model to generate historical wildfire spread trend simulation data corresponding to each historical wildfire data according to each historical wildfire data and each historical wildfire spread trend, and then constructs an initial machine learning model. Finally, the multiple historical wildfire data and the multiple historical wildfire spread trend simulation data are sequentially input into the initial machine learning model for training to obtain a target machine learning model, which is used to predict the detailed wildfire spread trend in the preset area; that is, by using the CFD model and the machine learning model to perform a combined prediction of the wildfire spread trend, the wildfire spread trend can be accurately analyzed to predict wildfire high-incidence areas and key areas, and then accurately detect wildfires and quickly extinguish them, thereby effectively solving the technical problems existing in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] in:

[0055] Figure 1 A schematic diagram of a method for predicting wildfire spread trends in an embodiment of the present application;

[0056] Figure 2 A schematic diagram of a wildfire spread trend prediction device in an embodiment of the present application;

[0057] Figure 3 1 is a diagram of the internal structure of a computer device in some embodiments. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 are within the scope of protection of the present invention.

[0059] In recent years, with the in-depth development of power resources, more and more transmission lines are crossing complex mountainous areas. The unique topography and climatic conditions in these areas are very likely to cause wildfires, which can cause transmission lines to trip or even burn out towers, causing long-term, irreversible major power accidents. Therefore, wildfires have become one of the main factors affecting the safe and stable operation of transmission lines. With the continuous expansion of the scale of the power grid and the frequent occurrence of extreme drought weather in recent years, the threat of wildfires to the safety of the power grid has become increasingly severe.

[0060] In order to effectively respond to this challenge, accurate analysis of wildfire spread trends to predict wildfire-prone areas and key regions is of great significance for timely and accurate detection of wildfires and their rapid extinguishing.

[0061] Other methods for predicting wildfire spread trends have been studied before this application, but because they rely on a single physical model or machine learning model, the accuracy of the prediction is limited, that is, the effect of accurate prediction cannot be achieved.

[0062] In response to the above problems, the present application proposes a wildfire spread trend prediction method, storage medium and device. By adopting CFD model and machine learning model to perform combined prediction of wildfire spread trend, it is possible to accurately analyze the wildfire spread trend to predict wildfire high-incidence areas and key areas, and then accurately detect wildfires and quickly extinguish them, thereby effectively solving the technical problems existing in the prior art. The specific implementation principle will be described in detail in the following embodiments.

[0063] In a first aspect, the present application provides a method for predicting wildfire spread trends.

[0064] See also Figure 1 , is a schematic diagram of a method for predicting wildfire spread trend in an embodiment of the present application, the method comprising:

[0065] Step 110: Acquire a plurality of historical wildfire data within a preset area and the historical wildfire spreading trends corresponding to the plurality of historical wildfire data.

[0066] The preset area range may be an area range pre-set by an operator, and wildfires may frequently occur in the area range, so it is necessary to predict the spread trend of wildfires.

[0067] It should be noted that historical wildfire data and historical wildfire spread trends are both historical data, and there is a one-to-one correspondence between historical power wildfire data and historical wildfire spread trends.

[0068] In some embodiments, historical wildfire data may include terrain parameters, vegetation parameters, and meteorological parameters, among which terrain parameters include but are not limited to parameters such as slope, slope aspect, soil type, and water body location, vegetation parameters include but are not limited to parameters such as vegetation type, vegetation density, vegetation height, and vegetation moisture content, and meteorological parameters include but are not limited to parameters such as wind speed, wind direction, temperature, and humidity.

[0069] Step 120: Using a preset CFD model, based on each historical wildfire data and each historical wildfire spread trend, generate historical wildfire spread trend simulation data corresponding to each historical wildfire data.

[0070] Among them, CFD stands for Computational Fluid Dynamics, which is a discipline that uses numerical analysis and algorithms to solve problems involving fluid flow.

[0071] In this application, it can be used to simulate the spread trend of wildfires. The preset CFD model can simulate the spread trend of wildfires, the diffusion of smoke, and the impact of the fire on the surrounding environment.

[0072] In some embodiments, a preset CFD model may be used to combine each historical wildfire data and the corresponding wildfire spread trend to generate wildfire spread trend simulation data to reproduce a wildfire spread trend situation that is more detailed than the historical wildfire spread trend situation.

[0073] Furthermore, in some embodiments, the preset CFD model can be based on historical wildfire data to consider multiple factors such as slope, slope direction, vegetation type, vegetation density, wind speed, wind direction, etc. These factors will affect the spread speed and direction of the wildfire. By adjusting these parameters, the preset CFD model can simulate the wildfire spread trend under different conditions, thereby generating wildfire spread trend simulation data corresponding to the historical wildfire data.

[0074] Understandably, the generated wildfire spread trend simulation data is very important for understanding the wildfire spread trend mechanism, predicting future fires, and formulating effective fire fighting strategies.

[0075] Step 130: Build an initial machine learning model.

[0076] The initial machine learning model here refers to a machine learning model that has not yet been trained.

[0077] In some embodiments, the initial machine learning model may be an XGboost machine learning model.

[0078] Step 140: Input multiple historical wildfire data and multiple historical wildfire spread trend simulation data into the initial machine learning model in sequence for training to obtain a target machine learning model, which is used to predict the details of the wildfire spread trend in a preset area.

[0079] Among them, the target machine learning model here refers to a trained machine learning model, which can be directly used to predict and output detailed information about the wildfire spread trend based on the input wildfire data. Its wildfire data is similar to the historical wildfire data in the above step 110, and will not be repeated here.

[0080] It should be noted that, assuming that the detailed wildfire spread trend and the historical wildfire spread trend correspond to the same wildfire data, it can be considered that the detailed wildfire spread trend is a more detailed wildfire spread trend, while the historical wildfire spread trend is a more rough wildfire spread trend, that is, the detailed wildfire spread trend and the historical wildfire spread trend are in an inclusive relationship, that is, the detailed wildfire spread trend includes the historical wildfire spread trend. Therefore, by using the CFD model and the machine learning model to perform a combined prediction of the wildfire spread trend, the wildfire spread trend output by the machine learning model prediction can be made more detailed and clear, so as to effectively improve the accuracy of the prediction, thereby achieving the effect of accurate prediction.

[0081] It should be further explained that after multiple historical wildfire data and multiple historical wildfire spread trend simulation data are successively input into the initial machine learning model for training, after training to a certain extent, a trained target machine learning model can be obtained; among them, each historical wildfire spread trend simulation data can be used as the true value of the initial machine learning model, that is, by comparing the true value with each historical wildfire spread trend simulation data output during the training process, it can be determined whether the initial machine learning model is well trained and has met the expected requirements.

[0082] It should also be noted that after obtaining the target machine learning model, in subsequent applications, you only need to input the wildfire data into the target machine learning model to automatically perform predictions and obtain detailed information on the wildfire spread trend as predicted output.

[0083] In an embodiment of the present application, by obtaining multiple historical wildfire data within a preset area and the historical wildfire spread trends corresponding to the multiple historical wildfire data, and then using a preset CFD model, according to each historical wildfire data and each historical wildfire spread trend, historical wildfire spread trend simulation data corresponding to each historical wildfire data is generated, and then an initial machine learning model is constructed, and finally, multiple historical wildfire data and multiple historical wildfire spread trend simulation data are sequentially input into the initial machine learning model for training to obtain a target machine learning model, which is used to predict the detailed wildfire spread trend within a preset area; that is, by using a CFD model and a machine learning model to perform a combined prediction of the wildfire spread trend, the wildfire spread trend can be accurately analyzed to predict wildfire high-incidence areas and key areas, and then accurately detect wildfires and quickly extinguish them, thereby effectively solving the technical problems existing in the prior art.

[0084] In a feasible implementation, step 120 in the above embodiment uses a preset CFD model to generate historical wildfire spread trend simulation data corresponding to each historical wildfire data according to each historical wildfire data and each historical wildfire spread trend, including: adjusting the model parameters of the preset CFD model according to the nth historical wildfire data to obtain the nth target CFD model; simulating the nth target CFD model to generate the nth initial wildfire spread simulation data; determining whether the nth target CFD model needs to be optimized according to the nth initial wildfire spread simulation data and the nth historical wildfire spread trend; when it is determined that the nth target CFD model does not need to be optimized, using the nth initial wildfire spread simulation data as the historical wildfire spread trend simulation data corresponding to the nth historical wildfire data; wherein n successively takes integers greater than 0 until n is equal to the total number of historical wildfire data.

[0085] In an embodiment of the present application, by adjusting the preset CFD model and judging whether it needs to be optimized according to each historical wildfire data and each historical wildfire spread trend, and generating historical wildfire spread trend simulation data corresponding to each historical wildfire data, the accuracy of the simulation data can be improved, the generalization ability of the model can be enhanced, the prediction efficiency can be improved, and a scientific basis can be provided for wildfire prevention and control.

[0086] It is understandable that improving the accuracy of simulation data: by adjusting the parameters of the CFD model for each historical wildfire data and judging whether the model needs to be optimized, it can ensure that the generated wildfire spread trend simulation data more accurately reflects the actual more detailed wildfire spread trend, thereby improving the accuracy of subsequent machine learning model training; enhancing the generalization ability of the model: since each historical wildfire data has been personalized Model adjustment and optimization, the trained machine learning model can better adapt to the wildfire spread trend under different terrain and climatic conditions, enhancing the generalization ability of the model; improving prediction efficiency: by iteratively processing each historical wildfire data and generating wildfire spread trend simulation data, it can provide rich training samples for the machine learning model, thereby speeding up the model training speed and improving prediction efficiency; providing a scientific basis for wildfire prevention and control: the generated wildfire spread trend simulation data and the trained machine learning model can provide a scientific basis for wildfire prevention and control, helping relevant departments to accurately predict wildfire high-incidence areas and key areas, take timely response measures, and reduce the threat of wildfires to power grid security.

[0087] In addition, this method generates more accurate and detailed simulation data on the historical wildfire spread trend by iterating each historical wildfire data, adjusting and judging whether the CFD model needs to be optimized, providing strong support for subsequent machine learning model training and improving the overall prediction effect of wildfire spread trend prediction.

[0088] In a feasible implementation method, the above embodiment adjusts the model parameters of the preset CFD model according to the nth historical wildfire data to obtain the nth target CFD model, including: constructing a preset geometric model of a preset area range; constructing a preset CFD model of the preset geometric model, and setting the initial conditions and boundary conditions of the preset CFD model; adjusting the model parameters of the preset geometric model according to the nth historical wildfire data to obtain the nth target geometric model; adjusting the model parameters of the preset CFD model according to the nth target geometric model to obtain the nth target CFD model.

[0089] It should be noted that the preset geometric model is constructed based on the preset area range, and the preset geometric model includes various characteristic parameters of the preset area range, such as terrain parameters, vegetation parameters, meteorological parameters, building parameters and other characteristic parameters.

[0090] Furthermore, the preset CFD model is constructed based on the preset geometric model, and the model parameters of the preset CFD model also include terrain parameters, vegetation parameters, meteorological parameters, building parameters, and the like.

[0091] In some embodiments, when the nth historical wildfire data includes terrain parameters, vegetation parameters and meteorological parameters, the model parameters of the preset geometric model include terrain parameters, vegetation parameters, meteorological parameters and building parameters, and the model parameters of the preset CFD model include terrain parameters, vegetation parameters, meteorological parameters and building parameters, the terrain parameters, vegetation parameters and meteorological parameters in the model parameters of the preset geometric model can be adjusted according to the terrain parameters, vegetation parameters and meteorological parameters in the nth historical wildfire data to obtain the nth target geometric model, and then the terrain parameters, vegetation parameters and meteorological parameters in the model parameters of the preset CFD model are adjusted according to the nth target geometric model to obtain the nth target CFD model.

[0092] In some embodiments, the initial conditions of the preset CFD model include but are not limited to the starting temperature, velocity field, etc.; the boundary conditions of the preset CFD model include but are not limited to the inlet (such as wind speed, direction, temperature and humidity, etc.), outlet (such as pressure boundary, free outlet condition, heat flux, etc.) and wall (such as different thermal properties and friction conditions of the ground, vegetation, buildings, etc.).

[0093] In an embodiment of the present application, by adjusting the model parameters of the preset CFD model based on the constructed preset geometric model according to the nth historical wildfire data to obtain the nth target CFD model, the accuracy of the model can be improved, the applicability of the model can be enhanced, and the prediction efficiency can be improved.

[0094] It can be understood that improving the accuracy of the model: by adjusting the model parameters of the preset geometric model and CFD model according to the historical wildfire data, it can ensure that the model is closer to the actual and detailed wildfire occurrence and spread trends, which helps to improve the accuracy of the simulation results, thereby generating more reliable historical wildfire spread trend simulation data; enhancing the applicability of the model: since each historical wildfire data has its own unique terrain, vegetation, meteorological and other characteristics, therefore, by personalizing the adjustment of model parameters, it can ensure that the model has high applicability in a wide range of wildfire prevention and control scenarios; improving prediction efficiency: by iteratively processing each historical wildfire data and generating wildfire spread trend simulation data, it can provide rich training samples for the machine learning model. These training samples not only help to speed up the training of the model, but also improve the prediction efficiency of the model, enabling it to respond to wildfire events faster and give accurate prediction results.

[0095] In a feasible implementation, constructing a preset geometric model of a preset area range in the above embodiment includes: acquiring area data of the preset area range; and constructing a preset geometric model according to the area data.

[0096] The regional data are various characteristic parameters corresponding to the preset regional range.

[0097] In some embodiments, regional data may include terrain parameters, vegetation parameters, meteorological parameters and building parameters; wherein terrain parameters include but are not limited to parameters such as slope, slope direction, soil type and water body location, vegetation parameters include but are not limited to parameters such as vegetation type, vegetation density, vegetation height and vegetation moisture content, meteorological parameters include but are not limited to parameters such as wind speed, wind direction, temperature and humidity, and building parameters include but are not limited to parameters such as building materials, building spacing and ventilation conditions.

[0098] In the embodiment of the present application, by acquiring regional data of a preset area range and constructing a preset geometric model based on the regional data, the simulation accuracy and the applicability of the model can be improved.

[0099] It can be understood that improving simulation accuracy: by collecting and analyzing regional data such as terrain parameters, vegetation parameters, meteorological parameters and building parameters in a preset area in detail, a geometric model that is closer to reality can be constructed. This precise geometric model can more accurately reflect the natural and man-made environmental characteristics of the preset area, thereby improving the accuracy of the simulation results; enhancing model applicability: the preset geometric model can be customized or adjusted according to different preset area ranges to adapt to different terrains, vegetation, meteorology and buildings. This flexibility enables the model to be widely used in different types of areas, enhancing the applicability of the model.

[0100] In a feasible implementation, the method in the above embodiment also includes: when it is determined that the nth target CFD model needs to be optimized, adjusting the model parameters and boundary conditions of the nth target CFD model to obtain an adjusted nth target CFD model; simulating the adjusted nth target CFD model to generate adjusted nth initial wildfire spread simulation data; judging whether the adjusted nth target CFD model needs to be optimized based on the adjusted nth initial wildfire spread simulation data and the nth historical wildfire spread trend; when it is determined that the adjusted nth target CFD model does not need to be optimized, using the adjusted nth initial wildfire spread simulation data as the historical wildfire spread trend simulation data corresponding to the nth historical wildfire data; when it is determined that the adjusted nth target CFD model needs to be optimized, continuing to adjust and simulate until no optimization is required.

[0101] Here, "belongs" refers to an inclusion relationship.

[0102] In the embodiments of the present application, by introducing an optimization process for the CFD model, the accuracy and practicality of the simulation of the wildfire spread trend are significantly improved, and the simulation accuracy can be improved.

[0103] It is understandable that the simulation accuracy can be improved by continuously adjusting the model parameters and boundary conditions of the CFD model, and based on the comparison results between the wildfire spread trend simulation data and the historical wildfire spread trend situation, the real and detailed wildfire spread trend can be gradually approached, thereby improving the simulation accuracy. This iterative optimization process helps to discover and correct errors and deficiencies in the model, making the model more in line with the actual and detailed situation.

[0104] In a feasible implementation, the adjusting of the model parameters and boundary conditions of the nth target CFD model in the above embodiment to obtain the adjusted nth target CFD model includes: using an optimization algorithm to search for the best combination parameters between the various parameters of the model parameters of the nth target CFD model and the various conditions in the boundary conditions; and adjusting the model parameters and boundary conditions of the nth target CFD model according to the best combination parameters to obtain the adjusted nth target CFD model.

[0105] In some embodiments, the optimization algorithm may be a genetic algorithm, particle swarm optimization (PSO), or Bayesian optimization or other optimization algorithms.

[0106] In an embodiment of the present application, by using an optimization algorithm to adjust the model parameters and boundary conditions of the nth target CFD model, an adjusted nth target CFD model is obtained, which can improve model simulation accuracy, increase efficiency, and reduce costs.

[0107] It is understandable that improving model simulation accuracy: the optimization algorithm can perform a global search in a wide parameter space by simulating natural processes or probability-based search strategies, so as to find a solution that is closer to the global optimum, which helps to avoid falling into the local optimum and improve the accuracy and reliability of the model's wildfire spread trend simulation data. The optimization algorithm can automatically adjust the model parameters and boundary conditions, reducing the time and cost of manual trial and error, which is especially important for complex CFD models because manual parameter adjustment can be time-consuming and error-prone. Improve efficiency and reduce costs: the optimization algorithm can find the best parameter combination faster through intelligent search strategies, thereby reducing the number of model adjustments, which helps to improve overall efficiency. Through the optimization algorithm, the required parameter combination can be searched and determined more accurately, thereby avoiding unnecessary calculations and reducing resource consumption.

[0108] In a feasible implementation method, the above-mentioned embodiment determines whether the nth target CFD model needs to be optimized based on the nth initial wildfire spread simulation data and the nth historical wildfire spread trend, including: when the nth historical wildfire spread trend belongs to the nth initial wildfire spread simulation data, determining that the nth target CFD model does not need to be optimized; when the nth historical wildfire spread trend does not belong to the nth initial wildfire spread simulation data, determining that the nth target CFD model needs to be optimized.

[0109] In the above embodiment, whether the adjusted nth target CFD model needs to be optimized is determined based on the adjusted nth initial wildfire spread simulation data and the nth historical wildfire spread trend, including: when the nth historical wildfire spread trend belongs to the adjusted nth initial wildfire spread simulation data, determining that the adjusted nth target CFD model does not need to be optimized; when the nth historical wildfire spread trend does not belong to the adjusted nth initial wildfire spread simulation data, determining that the adjusted nth target CFD model needs to be optimized.

[0110] In an embodiment of the present application, by comparing the nth historical wildfire spread trend with the nth initial wildfire spread simulation data (and the adjusted nth initial wildfire spread simulation data), it is determined whether the nth target CFD model needs to be optimized, which can improve the model accuracy.

[0111] It is understandable that improving the accuracy of the model: by comparing the historical wildfire spread trend with the initial wildfire spread trend simulation data, the effectiveness of the model under specific conditions can be verified. If the initial wildfire spread trend simulation data contains the historical wildfire spread trend, it means that the model is accurate under these conditions and no further optimization is required. If the initial wildfire spread trend simulation data does not contain the historical wildfire spread trend, that is, there are significant differences, it means that the model has deficiencies or errors in some aspects. At this time, the model needs to be optimized to improve the accuracy of the wildfire spread trend simulation data it generates.

[0112] In a feasible implementation method, the above-mentioned embodiment simulates the nth target CFD model to generate the nth initial wildfire spread simulation data, including: using different types of combustion models and radiation models to simulate the nth target CFD model to generate wildfire spread simulation data corresponding to different types of combustion models and radiation models; using the wildfire spread simulation data corresponding to different types of combustion models and radiation models as the nth initial wildfire spread simulation data.

[0113] The above-mentioned embodiment simulates the adjusted nth target CFD model to generate adjusted nth initial wildfire spread simulation data, including: using different types of combustion models and radiation models to simulate the adjusted nth target CFD model to generate wildfire spread simulation data corresponding to different types of combustion models and radiation models; using the wildfire spread simulation data corresponding to different types of combustion models and radiation models as the nth initial wildfire spread simulation data.

[0114] In an embodiment of the present application, by simulating the nth target CFD model (or the adjusted nth target CFD model) using different types of combustion models and radiation models, and generating corresponding wildfire spread simulation data as the nth initial wildfire spread simulation data, the comprehensiveness and accuracy of the simulation data can be improved.

[0115] It is understandable that the comprehensiveness and accuracy of simulation data should be improved: the wildfire spread trend process involves a variety of combustion mechanisms, such as pyrolysis, combustion and burnout of combustibles. Different types of combustion models can be used to simulate these different combustion mechanisms, thereby more comprehensively reflecting the actual and detailed process of the wildfire spread trend. In the wildfire spread trend process, heat transfer through radiation is an important link. The use of different types of radiation models can simulate the radiation transfer process of heat under different conditions, thereby improving the accuracy of wildfire spread trend simulation data in terms of radiation heat transfer.

[0116] In a second aspect, the present application provides a device for predicting wildfire spread trends.

[0117] See also Figure 2 , is a schematic diagram of a wildfire spread trend prediction device in an embodiment of the present application, the device 210 includes:

[0118] An acquisition module 211 is used to acquire a plurality of historical wildfire data within a preset area and a historical wildfire spreading trend corresponding to the plurality of historical wildfire data;

[0119] A generating module 212, for generating historical wildfire spread trend simulation data corresponding to each historical wildfire data, using a preset CFD model according to each historical wildfire data and each historical wildfire spread trend;

[0120] A construction module 213 is used to construct an initial machine learning model;

[0121] The training prediction module 214 is used to input multiple historical wildfire data and multiple historical wildfire spread trend simulation data into the initial machine learning model in sequence for training to obtain a target machine learning model, which is used to predict the details of the wildfire spread trend in a preset area.

[0122] In the embodiment of the present application, the relevant contents of the acquisition module 211, the generation module 212, the construction module 213 and the training prediction module 214 can be referred to in Figure 1 The contents in the illustrated embodiment will not be described in detail here.

[0123] It should be noted that the device 210 of the present application also includes some other modules. It can be understood that the method of the present application and the device 210 have a one-to-one correspondence. Therefore, some other modules of the device 210 of the present application are the contents corresponding to the method of the present application in the above-mentioned embodiment.

[0124] In an embodiment of the present application, by obtaining multiple historical wildfire data within a preset area and the historical wildfire spread trends corresponding to the multiple historical wildfire data, and then using a preset CFD model, according to each historical wildfire data and each historical wildfire spread trend, historical wildfire spread trend simulation data corresponding to each historical wildfire data is generated, and then an initial machine learning model is constructed, and finally, multiple historical wildfire data and multiple historical wildfire spread trend simulation data are sequentially input into the initial machine learning model for training to obtain a target machine learning model, which is used to predict the detailed wildfire spread trend within a preset area; that is, by using a CFD model and a machine learning model to perform a combined prediction of the wildfire spread trend, the wildfire spread trend can be accurately analyzed to predict wildfire high-incidence areas and key areas, and then accurately detect wildfires and quickly extinguish them, thereby effectively solving the technical problems existing in the prior art.

[0125] In a third aspect, the present application further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes a method for predicting wildfire spread trends in the above method embodiment.

[0126] In a fourth aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes a method for predicting a wildfire spread trend in the above method embodiment.

[0127] Figure 3 The internal structure diagram of a computer device in some embodiments is shown. The computer device may be a terminal, a server, or a gateway. Figure 3 As shown, the computer device includes a processor, a memory and a network interface connected through a system bus.

[0128] The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement each step in the above method embodiment. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can implement each step in the above method embodiment. Those skilled in the art can understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0129] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0130] Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synch li nk) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0131] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0132] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for predicting the spread trend of wildfires, characterized in that: The method comprises: Obtaining multiple historical wildfire data within a preset area and historical wildfire spreading trends corresponding to the multiple historical wildfire data; Using a preset CFD model, based on each historical wildfire data and each historical wildfire spread trend, generate historical wildfire spread trend simulation data corresponding to each historical wildfire data; Build an initial machine learning model; Multiple historical wildfire data and multiple historical wildfire spread trend simulation data are sequentially input into the initial machine learning model for training to obtain a target machine learning model, which is used to predict the details of the wildfire spread trend in the preset area.

2. The method according to claim 1, characterized in that The preset CFD model is used to generate historical wildfire spread trend simulation data corresponding to each historical wildfire data according to each historical wildfire data and each historical wildfire spread trend, including: According to the nth historical wildfire data, the model parameters of the preset CFD model are adjusted to obtain the nth target CFD model; Simulate the nth target CFD model to generate nth initial wildfire spread simulation data; Determine whether the nth target CFD model needs to be optimized based on the nth initial wildfire spread simulation data and the nth historical wildfire spread trend; When it is determined that there is no need to optimize the nth target CFD model, the nth initial wildfire spread simulation data is used as the historical wildfire spread trend simulation data corresponding to the nth historical wildfire data; Among them, n takes integers greater than 0 in sequence until n is equal to the total number of historical wildfire data.

3. The method according to claim 2, characterized in that The method of adjusting the model parameters of the preset CFD model according to the nth historical wildfire data to obtain the nth target CFD model includes: Constructing a preset geometric model of the preset area; Constructing the preset CFD model of the preset geometric model, and setting initial conditions and boundary conditions of the preset CFD model; According to the nth historical wildfire data, the model parameters of the preset geometric model are adjusted to obtain the nth target geometric model; According to the nth target geometric model, the model parameters of the preset CFD model are adjusted to obtain the nth target CFD model.

4. The method according to claim 3, characterized in that: The step of constructing a preset geometric model of the preset area range includes: Acquire regional data of the preset regional range; The preset geometric model is constructed according to the area data.

5. The method according to claim 3, characterized in that: The method further comprises: When it is determined that the nth target CFD model needs to be optimized, adjusting the model parameters and boundary conditions of the nth target CFD model to obtain an adjusted nth target CFD model; Simulating the adjusted nth target CFD model to generate adjusted nth initial wildfire spread simulation data; Determine whether the adjusted nth target CFD model needs to be optimized based on the adjusted nth initial wildfire spread simulation data and the nth historical wildfire spread trend; When it is determined that there is no need to optimize the adjusted nth target CFD model, the adjusted nth initial wildfire spread simulation data is used as the historical wildfire spread trend simulation data corresponding to the nth historical wildfire data; When it is determined that the adjusted nth target CFD model needs to be optimized, the adjustment and simulation are continued until optimization is no longer necessary.

6. The method according to claim 5, characterized in that The adjusting the model parameters and boundary conditions of the nth target CFD model to obtain the adjusted nth target CFD model includes: Using an optimization algorithm, searching for the best combination parameters between each parameter of the model parameters of the nth target CFD model and each condition in the boundary conditions; According to the optimal combination parameters, the model parameters and boundary conditions of the nth target CFD model are adjusted to obtain the adjusted nth target CFD model.

7. The method according to claim 2 or 5, characterized in that: The determining whether to optimize the nth target CFD model according to the nth initial wildfire spread simulation data and the nth historical wildfire spread trend includes: In the case where the nth historical wildfire spread trend situation belongs to the nth initial wildfire spread simulation data, it is determined that there is no need to optimize the nth target CFD model; In the case where the nth historical wildfire spread trend situation does not belong to the nth initial wildfire spread simulation data, it is determined that the nth target CFD model needs to be optimized; According to the adjusted nth initial wildfire spread simulation data and the nth historical wildfire spread trend, it is determined whether the adjusted nth target CFD model needs to be optimized, including: In the case where the nth historical wildfire spread trend situation belongs to the adjusted nth initial wildfire spread simulation data, it is determined that there is no need to optimize the adjusted nth target CFD model; When the nth historical wildfire spread trend situation does not belong to the adjusted nth initial wildfire spread simulation data, it is determined that the adjusted nth target CFD model needs to be optimized.

8. The method according to claim 2 or 5, characterized in that: The simulating the nth target CFD model to generate the nth initial wildfire spread simulation data includes: Use different types of combustion models and radiation models to simulate the nth target CFD model and generate wildfire spread simulation data corresponding to different types of combustion models and radiation models; The wildfire spread simulation data corresponding to different types of combustion models and radiation models are used as the nth initial wildfire spread simulation data; The adjusted nth target CFD model is simulated to generate adjusted nth initial wildfire spread simulation data, including: Using different types of combustion models and radiation models, simulating the adjusted nth target CFD model, and generating wildfire spread simulation data corresponding to different types of combustion models and radiation models; The wildfire spread simulation data corresponding to different types of combustion models and radiation models are used as the nth initial wildfire spread simulation data.

9. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 8.

10. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the method according to any one of claims 1 to 8.

Images