Forest fire spreading trend prediction method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a forest fire spreading trend prediction method and device, equipment and a storage medium. The method comprises the steps of obtaining preprocessed landform data and meteorological data of a target region; extracting feature information related to fire spreading from the feature information, and constructing an initial feature set; evaluating the S transformation mathematical model through a feature importance evaluation method, identifying feature information having important influence on fire spreading, dividing the obtained processed data set into a training set, a verification set and a test set, and training the constructed S transformation mathematical model of fire spreading; and based on landform data and meteorological data of a to-be-detected target region, predicting and obtaining a fire spreading path and speed through the trained S transformation mathematical model. The scheme has important application prospects in the field of power grids, the early warning capability of power grid mountain fire prevention and control can be improved, the damage of fire disasters to power transmission lines is reduced, and the stability and safety of power grid operation are ensured.

Description

Technical Field

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

[0002] Wildfire damage to transmission lines may lead to power system interruptions and failures, posing huge challenges to power grid operation and maintenance. Traditional fire spread trend models mostly ignore the impact of terrain and meteorological conditions on fire spread, which is particularly prominent in areas with complex terrain such as mountainous areas and forests. Therefore, it is of great significance to develop a fire spread trend model based on terrain and meteorological conditions and apply it to the power grid field.

[0003] Currently, power grid monitoring and operation and maintenance mainly rely on manual inspections and sensor monitoring, and lack an effective fire warning system. The fire spread trend model based on terrain and meteorological conditions can combine geographic information systems (GIS) and meteorological forecasting technology to accurately predict the spread of fires, providing a scientific basis for the deployment of fire prevention resources and the formulation of emergency measures for the power grid.

[0004] In addition, with the development of intelligent power grid technology, data analysis technology is increasingly used in power grid monitoring and early warning. The fire spread model based on S transform combines the analysis of topographic and meteorological data, and integrates with the power grid monitoring system to achieve real-time monitoring and early warning of fire spread, providing protection for the safe operation of the power grid. In summary, the fire spread direction model based on terrain and meteorological conditions has important application prospects in the field of power grids. This model improves the early warning capability of power grid wildfire prevention and control, reduces the damage of fire to transmission lines, and ensures the stability and safety of power grid operation. Summary of the invention

[0005] Based on this, it is necessary to propose a method, device, equipment and storage medium for predicting the spread trend of wildfires in response to the above problems.

[0006] To achieve the above objectives, the present application provides a first aspect of a method for predicting the spread trend of wildfires, the method comprising:

[0007] Acquire topographic data of the target area, wherein the topographic data includes terrain elevation, vegetation coverage, and terrain slope;

[0008] Obtain weather data for the target area;

[0009] Preprocessing the topographic and geomorphic data and the meteorological data to obtain preprocessed topographic and geomorphic data and meteorological data;

[0010] Extracting feature information related to fire spread from the preprocessed topographic data and meteorological data, and constructing an initial feature set;

[0011] The initial feature set is evaluated by a feature importance evaluation method to identify feature information that has a significant impact on the spread of fire, and a processed data set is obtained;

[0012] Dividing the processed data set into a training set, a validation set and a test set;

[0013] Construct the S-transform mathematical model of fire spread;

[0014] The training of the S-transform mathematical model is completed through the training set, the validation set and the test set;

[0015] Based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted through the trained S-transform mathematical model.

[0016] In some embodiments, the fire spreading path and speed are predicted based on the topographic data and meteorological data of the target area to be detected by using the trained S-transform mathematical model, and then the method further includes:

[0017] A spread control strategy is determined based on the path and speed of the fire spread.

[0018] In some implementations, the preprocessing of the topographic data and the meteorological data to obtain the preprocessed topographic data and the meteorological data specifically includes:

[0019] Integrate the topographic data and meteorological data, and convert their formats;

[0020] The integrated data is subjected to noise removal, outlier and erroneous data processing, including detecting and repairing missing values ​​in the data, processing duplicate data, and converting the cleaned data into a unified format such as CSV, JSON, XML or array matrix.

[0021] In some implementations, constructing an S-transform mathematical model of fire spread specifically includes:

[0022] Use the S-transform machine learning algorithm in a GIS environment to build a fire spread model;

[0023] Processing the topographic data and meteorological data of the target area by using an S-transform algorithm to obtain a two-dimensional matrix of time and frequency;

[0024] Adding each column of the two-dimensional matrix related to time and frequency to obtain an energy spectrum graph related only to frequency;

[0025] The historical topographic data and their spreading trends are processed by S transformation to obtain the energy spectrum database;

[0026] The energy spectrum is analyzed by S-transform and compared with the S-transform mathematical model, and the model with the highest similarity is selected to determine the spreading trend.

[0027] In some implementations, the training of the S transform mathematical model is completed by the training set, the validation set, and the test set, specifically comprising: transforming the Fourier forward transform into:

[0028]

[0029] By adding the window function g(t) to the time series h(t), the spectrum becomes:

[0030]

[0031] For the S transform, define the window function:

[0032]

[0033] By scaling and translating the Gaussian window function, we get the formula for the S transform:

[0034]

[0035] Add the same time columns of the two-dimensional matrix after S transformation to obtain the corresponding spectrum diagram under the data.

[0036] In some implementations, determining a spread control strategy according to the path and speed of fire spread specifically includes:

[0037] Conduct risk assessments on areas that may be affected along the path of fire spread, including assessing the population density, natural resources, and infrastructure of the area to determine the potential risk level of fire spread to the area;

[0038] Divide the affected area into different emergency response areas according to the degree of potential risk;

[0039] A spread control strategy for the emergency response area is formulated based on the path and speed of the fire spread and the potential risk level; the spread control strategy includes determining the evacuation area, the blockade area, and the emergency refuge area.

[0040] In some embodiments, the meteorological data includes temperature, humidity, wind speed, and wind direction. The topographic data includes elevation, slope, and slope direction.

[0041] To achieve the above-mentioned purpose, the second aspect of the present application provides a device for predicting the spread trend of wildfires, the device comprising:

[0042] The first acquisition module is used to acquire the topographic data of the target area, wherein the topographic data includes the terrain elevation, vegetation coverage, and terrain slope;

[0043] The second acquisition module acquires the meteorological data of the target area;

[0044] A preprocessing module, used for preprocessing the topographic data and meteorological data to obtain preprocessed topographic data and meteorological data;

[0045] A feature extraction module, used to extract feature information related to fire spread from the pre-processed topographic data and meteorological data, and to construct an initial feature set;

[0046] An evaluation module, used to evaluate the initial feature set by a feature importance evaluation method, identify feature information that has a significant impact on fire spread, and obtain a processed data set;

[0047] A data set module, used to divide the processed data set into a training set, a validation set and a test set;

[0048] Model building module, used to build the S-transform mathematical model of fire spread;

[0049] A model training module, used to complete the training of the S-transform mathematical model through the training set, the validation set and the test set;

[0050] The fire spread module is used to predict the path and speed of fire spread based on the topographic data and meteorological data of the target area to be detected through the trained S-transform mathematical model.

[0051] To achieve the above-mentioned object, the third aspect of the present application 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 the following steps:

[0052] Acquire topographic data of the target area, wherein the topographic data includes terrain elevation, vegetation coverage, and terrain slope;

[0053] Obtain weather data for the target area;

[0054] Preprocessing the topographic and geomorphic data and the meteorological data to obtain preprocessed topographic and geomorphic data and meteorological data;

[0055] Extracting feature information related to fire spread from the preprocessed topographic data and meteorological data, and constructing an initial feature set;

[0056] The initial feature set is evaluated by a feature importance evaluation method to identify feature information that has a significant impact on the spread of fire, and a processed data set is obtained;

[0057] Dividing the processed data set into a training set, a validation set and a test set;

[0058] Construct the S-transform mathematical model of fire spread;

[0059] The training of the S-transform mathematical model is completed through the training set, the validation set and the test set;

[0060] Based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted through the trained S-transform mathematical model.

[0061] To achieve the above-mentioned object, the fourth aspect of the present application provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps:

[0062] Acquire topographic data of the target area, wherein the topographic data includes terrain elevation, vegetation coverage, and terrain slope;

[0063] Obtain weather data for the target area;

[0064] Preprocessing the topographic and geomorphic data and the meteorological data to obtain preprocessed topographic and geomorphic data and meteorological data;

[0065] Extracting feature information related to fire spread from the preprocessed topographic data and meteorological data, and constructing an initial feature set;

[0066] The initial feature set is evaluated by a feature importance evaluation method to identify feature information that has a significant impact on the spread of fire, and a processed data set is obtained;

[0067] Dividing the processed data set into a training set, a validation set and a test set;

[0068] Construct the S-transform mathematical model of fire spread;

[0069] The training of the S-transform mathematical model is completed through the training set, the validation set and the test set;

[0070] Based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted through the trained S-transform mathematical model.

[0071] The embodiments of the present invention have the following beneficial effects:

[0072] The present invention obtains pre-processed topographic data and meteorological data of a target area; extracts feature information related to fire spread therefrom and constructs an initial feature set; evaluates the feature information through a feature importance evaluation method, identifies feature information that has a significant impact on fire spread, and divides the processed data set into a training set, a validation set and a test set, and trains the constructed S-transform mathematical model of fire spread; based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted through the trained S-transform mathematical model, providing a scientific basis for power grid wildfire prevention and control, improving emergency response efficiency, and reducing losses caused by fire, and can be widely used in the field of power grid transmission line wildfire prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] 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.

[0074] in:

[0075] Figure 1 A schematic diagram of a flow chart of a method for predicting a wildfire spreading trend in one embodiment;

[0076] Figure 2 It is an S-transformation model extraction diagram in one embodiment;

[0077] Figure 3 A diagram of a device for predicting the spread trend of wildfires in one embodiment;

[0078] Figure 4 is a schematic diagram of the structure of a computer device in one embodiment;

[0079] Figure 5 FIG. 4 is a schematic diagram of the structure of a computer-readable storage medium in one embodiment. DETAILED DESCRIPTION

[0080] 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.

[0081] In the embodiment of the present application, a method for predicting the spread trend of wildfire is provided. Figure 1 , Figure 11 is a flow chart of a method for predicting a wildfire spreading trend in an embodiment; the method for predicting a wildfire spreading trend includes steps S1 to S9.

[0082] Step S1, obtaining topographic data of a target area, wherein the topographic data includes terrain elevation, vegetation coverage, and terrain slope;

[0083] Step S2, obtaining meteorological data of the target area;

[0084] In some embodiments, the meteorological data includes temperature, humidity, wind speed, and wind direction. The topographic data includes elevation, slope, and slope direction.

[0085] Specifically, it includes the selection and deployment of data sources, real-time or regular data collection, data transmission and storage, etc. First, determine the granularity, coverage and spatiotemporal resolution of the data, consider the availability, accuracy and completeness of the data, select high-quality data sources, understand the data collection method, update frequency, accuracy and other information, consider the coverage and timeliness of the data, and select data sources that cover the target area and are updated in a timely manner. After determining the data source, use the GIS system to obtain the topographic data such as elevation, slope, and slope direction of the target area, and use the meteorological station to obtain meteorological information such as temperature, humidity, wind speed, and wind direction in real time. Then integrate and preprocess the collected topographic data with the meteorological data, including data format conversion, outlier detection and processing, missing value filling and other preprocessing work. Finally, store the processed topographic and meteorological data for subsequent model construction and analysis.

[0086] Step S3, preprocessing the topographic data and meteorological data to obtain preprocessed topographic data and meteorological data;

[0087] In some embodiments, the topographic data and meteorological data are preprocessed, including establishing features and normalizing data. In the feature extraction stage, feature information that has an important impact on the spread of fire is extracted from the preprocessed topographic data, including terrain elevation, slope, slope aspect, vegetation type and coverage, etc. The method is usually based on terrain analysis, remote sensing technology and geographic information system, including digital elevation model (DEM) generation based on optical and radar data, vegetation index calculation, terrain parameter calculation, etc.

[0088] In some implementations, the preprocessing of the topographic data and the meteorological data to obtain the preprocessed topographic data and the meteorological data specifically includes:

[0089] Integrate the topographic data and meteorological data, and convert their formats;

[0090] The integrated data is subjected to noise removal, outlier and erroneous data processing, including detecting and repairing missing values ​​in the data, processing duplicate data, and converting the cleaned data into a unified format such as CSV, JSON, XML or array matrix.

[0091] Specifically, firstly, outliers and missing values ​​are processed to ensure the quality and consistency of the data; then the collected topographic and meteorological data are integrated and converted to a format so that they can be used by the model, and the integrated data are de-noised, outliers, and erroneous data are processed, including detecting and repairing missing values ​​in the data and processing duplicate data, and the cleaned data is converted to a unified format such as CSV, JSON, XML or array matrix so that it can be used by the model. If there is a special structure, it is converted to array and matrix format. During the format conversion process, the data needs to be standardized or normalized to eliminate the dimensional differences between different features and improve the prediction accuracy. For example, different coordinate systems are used when collecting topographic and meteorological data, and it is necessary to use geographic information system (GIS) tools to convert the coordinate systems so that the two are in the same coordinate system. During the format conversion process, the data is standardized to eliminate the dimensional differences between different features.

[0092] Step S4, extracting feature information related to fire spread from the preprocessed topographic data and meteorological data, and constructing an initial feature set;

[0093] Specifically, after the data are unified and processed, a feature set is established, and a series of features related to fire spread are selected from the collected topographic and meteorological data to construct the initial feature set.

[0094] Step S5, evaluating the initial feature set by a feature importance evaluation method, identifying feature information that has a significant impact on the spread of fire, and obtaining a processed data set;

[0095] Specifically, the feature importance evaluation method is used to evaluate the initial feature set and identify the features that have an important impact on the spread of fire. The feature importance evaluation methods include information gain, variance analysis, correlation coefficient analysis, and feature importance based on tree models. Based on the feature importance evaluation results, the features that have an important impact on the spread of fire are selected, and single variable feature selection methods (such as variance selection method), model-based feature selection methods (such as recursive feature elimination method), embedding-based feature selection methods (such as Lasso regression), etc. are used. For the original data that cannot be used directly, the PCA, linear discriminant analysis (LDA), and independent component analysis (ICA) feature extraction methods are used to extract representative features from them, and finally multiple features are combined.

[0096] Step S6, dividing the processed data set into a training set, a validation set and a test set;

[0097] Specifically, the processed dataset is divided into training set, validation set and test set in preparation for model training, tuning and evaluation.

[0098] Step S7, constructing an S-transform mathematical model of fire spread;

[0099] Specifically, the collected topographic and meteorological data are used to construct an S-transform mathematical model of fire spread, including data input, S-transform processing, energy spectrum calculation, model fitting and other steps.

[0100] In some implementations, constructing an S-transform mathematical model of fire spread specifically includes:

[0101] Use the S-transform machine learning algorithm in a GIS environment to build a fire spread model;

[0102] Processing the topographic data and meteorological data of the target area by using an S-transform algorithm to obtain a two-dimensional matrix of time and frequency;

[0103] Adding each column of the two-dimensional matrix related to time and frequency to obtain an energy spectrum graph related only to frequency;

[0104] The historical topographic data and their spreading trends are processed by S transformation to obtain the energy spectrum database;

[0105] The energy spectrum is analyzed by S-transform and compared with the S-transform mathematical model, and the model with the highest similarity is selected to determine the spreading trend.

[0106] Specifically, the S-transform machine learning algorithm is used in the GIS environment to build a fire spread model. The data analysis tools provided by the GIS software are used to verify and evaluate the model to ensure the accuracy and reliability of the model. Finally, the topographic and meteorological data of the target area are input into the GIS, and the S-transform machine learning algorithm is used to predict the fire spread path and speed. The steps to build a fire spread model using the S-transform machine learning algorithm are as follows:

[0107] Problem definition, first clearly define the problem, that is, the specific goals of the fire spread model and the problems to be solved, including predicting the fire spread path, assessing the risk of fire spread, etc. According to the definition of the problem, the topographic data and meteorological data are input into the S transform algorithm to obtain a two-dimensional matrix about time and frequency. Each column of the S transform matrix is ​​added to obtain an energy spectrum graph related only to frequency. The historical topographic data and its spread trend are processed by S transform to obtain an energy spectrum database, which provides a decision basis for subsequent samples. Input the sample data, use S transform to analyze its energy spectrum graph and compare it with the model, select the model with the highest similarity to determine the spread trend, and thus guide emergency response and fire control work.

[0108] Step S8, completing the training of the S-transform mathematical model through the training set, the validation set and the test set;

[0109] In some implementations, the training of the S transform mathematical model is completed by the training set, the validation set, and the test set, specifically comprising: transforming the Fourier forward transform into:

[0110]

[0111] By adding the window function g(t) to the time series h(t), the spectrum becomes:

[0112]

[0113] For the S transform, define the window function:

[0114]

[0115] By scaling and translating the Gaussian window function, we get the formula for the S transform:

[0116]

[0117] Add the same time columns of the two-dimensional matrix after S transformation to obtain the corresponding spectrum diagram under the data.

[0118] Specifically, see Figure 2 , which is the S-transformation model extraction diagram in the embodiment of the present invention, that is, the spectrum diagram under the corresponding data extracted after model training. By comparing and analyzing the spectrum diagram with the spectrum library, the spread trend of the wildfire can be further determined.

[0119] Step S9, based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted by the trained S-transform mathematical model;

[0120] In some embodiments, the S-transform learning model is optimized and adjusted using existing data to improve its prediction ability, including selecting a suitable network structure, optimizing model parameters, and improving model algorithms, etc., by increasing or decreasing the connection relationship between nodes, adjusting the probability distribution parameters of nodes, or improving the learning algorithm of the model, improving the accuracy and robustness of the model, and predicting and analyzing the topography and meteorological data of the target area, and then interpreting and evaluating the results of the model output using error analysis, quantitative evaluation, etc. Finally, the rationality and feasibility of the prediction results are analyzed and visualized by drawing fire spread path maps, speed maps, area maps, etc. At the same time, data statistics tables, model evaluation reports, etc. are prepared to comprehensively analyze and summarize the prediction results.

[0121] In some embodiments, based on the verification results, the S-transform machine learning algorithm is optimized and adjusted to improve its fit and prediction performance, predict the path and speed of fire spread, interpret and evaluate the results, analyze the rationality and feasibility of the prediction results, and visualize the prediction results in the form of charts after the evaluation is completed.

[0122] In some embodiments, the path and speed of fire spread are predicted by the trained S-transform mathematical model based on the topographic data and meteorological data of the target area to be detected, and then the method further includes:

[0123] A spread control strategy is determined based on the path and speed of the fire spread.

[0124] Specifically, the output results of the fire spread model are interpreted, including the predicted path and speed, and risk assessment is conducted on areas that may be affected along the fire spread path. Based on the risk assessment results, the affected areas are divided into different emergency response areas and different spread control strategies are formulated.

[0125] In some implementations, determining a spread control strategy according to the path and speed of fire spread specifically includes:

[0126] Conduct risk assessments on areas that may be affected along the path of fire spread, including assessing the population density, natural resources, and infrastructure of the area to determine the potential risk level of fire spread to the area;

[0127] Divide the affected area into different emergency response areas according to the degree of potential risk;

[0128] A spread control strategy for the emergency response area is formulated based on the path and speed of the fire spread and the potential risk level; the spread control strategy includes determining the evacuation area, the blockade area, and the emergency refuge area.

[0129] In some embodiments, the output results of the fire spread model are first interpreted as the basis for formulating a spread control strategy, and a risk assessment is performed on the areas that may be affected on the fire spread path, including assessing the population density, natural resources, infrastructure, etc. of the area to determine the potential risk level of the fire spread to the area, and the affected area is divided into different emergency response areas. According to the prediction results and risk assessment of the fire spread model, different spread control strategies are formulated, including determining evacuation areas, blockade areas, emergency refuge areas, etc. Finally, according to the formulated spread control strategy, emergency response measures are taken in a timely manner and early warnings are issued.

[0130] By adopting the technical solution of this embodiment, a method for predicting the spread trend of wildfires is disclosed. The present invention provides an efficient and reliable scientific basis for fire spread prediction and emergency response. By combining topography and meteorological conditions, the S transformation model of the present invention can accurately predict the spread path and speed of the fire, so that relevant departments and institutions can take emergency response measures in time, quickly mobilize rescue forces, minimize the losses caused by the fire, and improve the efficiency of emergency response. At the same time, by formulating a scientific and reasonable spread control strategy, the damage of wildfires to power grid transmission lines is minimized, the risk of line failures and power outages is reduced, and the stable operation of the power grid is guaranteed, which helps to improve the disaster resistance of the power grid, reduce the impact of wildfires on the power grid, and make important contributions to the sustainable development of the power industry and social stability.

[0131] In the embodiment of the present application, a device for predicting the spread trend of wildfire is provided. Figure 3 , Figure 3 This is a diagram of a device for predicting wildfire spread trends in an embodiment, which includes a first acquisition module 201, a second acquisition module 202, a preprocessing module 203, a feature extraction module 204, an evaluation module 205, a data set module 206, a model building module 207, a model training module 208 and a fire spread module 209.

[0132] The first acquisition module 201 is configured to acquire topographic data of a target area, wherein the topographic data includes topographic elevation, vegetation coverage, and topographic slope;

[0133] The second acquisition module 202 is configured to acquire meteorological data of a target area;

[0134] Specifically, the first acquisition module 201 and the second acquisition module 202 collect topographic and meteorological data of the target area to provide necessary basic data support for subsequent model construction and predictive analysis, and obtain topographic and meteorological data through various sensors, monitoring equipment or other data sources, including terrain elevation, slope, slope direction, temperature, humidity, wind speed, wind direction and other information.

[0135] The preprocessing module 203 is configured to preprocess the topographic data and the meteorological data to obtain the preprocessed topographic data and the meteorological data;

[0136] A feature extraction module 204 is configured to extract feature information related to fire spread from the pre-processed topographic data and meteorological data, and construct an initial feature set;

[0137] An evaluation module 205 is configured to evaluate the initial feature set by a feature importance evaluation method, identify feature information that has a significant impact on fire spread, and obtain a processed data set;

[0138] The data set module 206 is configured to divide the processed data set into a training set, a validation set and a test set;

[0139] A model building module 207, configured to build an S-transform mathematical model of fire spread;

[0140] Specifically, the processed topographic and meteorological data are combined with the S-transform mathematical modeling technology to construct an S-transform mathematical model of fire spread that can accurately predict the fire spread path and speed. The model construction includes input data, S-transform processing, energy spectrum analysis, etc. Finally, the model construction module 207 outputs the constructed S-transform mathematical model of fire spread as the input of the prediction analysis module and the control strategy module.

[0141] A model training module 208, configured to complete the training of the S-transform mathematical model through the training set, the validation set and the test set;

[0142] The fire spreading module 209 is configured to predict the path and speed of fire spreading based on the topographic data and meteorological data of the target area to be detected through the trained S-transform mathematical model.

[0143] Specifically, based on the trained S-transform mathematical model, the input topographic and meteorological data are analyzed and predicted to predict the spread path, speed and scope of the fire.

[0144] In some embodiments, the wildfire spread trend prediction device further includes a control strategy module 210 configured to determine a spread control strategy according to the path and speed of the fire spread.

[0145] Specifically, according to the predicted fire spread path, speed and scope, appropriate emergency response measures are taken, including resource allocation, personnel evacuation, fire line blockade and other control strategies to minimize the spread and impact of the fire. At the same time, the control strategy module dynamically adjusts and optimizes the control strategy according to the real-time situation of fire spread to improve the efficiency and effectiveness of fire response.

[0146] In some implementations, the preprocessing module 203 is further configured to integrate the topographic data and meteorological data and perform format conversion;

[0147] The integrated data is subjected to noise removal, outlier and erroneous data processing, including detecting and repairing missing values ​​in the data, processing duplicate data, and converting the cleaned data into a unified format such as CSV, JSON, XML or array matrix.

[0148] In some embodiments, the model building module 207 is further configured to build a fire spread model using an S-transform machine learning algorithm in a GIS environment;

[0149] Processing the topographic data and meteorological data of the target area by using an S-transform algorithm to obtain a two-dimensional matrix of time and frequency;

[0150] Adding each column of the two-dimensional matrix related to time and frequency to obtain an energy spectrum graph related only to frequency;

[0151] The historical topographic data and their spreading trends are processed by S transformation to obtain the energy spectrum database;

[0152] The energy spectrum is analyzed by S-transform and compared with the S-transform mathematical model, and the model with the highest similarity is selected to determine the spreading trend.

[0153] In some embodiments, the model training module 208 is further configured to transform the Fourier transform into:

[0154]

[0155] By adding the window function g(t) to the time series h(t), the spectrum becomes:

[0156]

[0157] For the S transform, define the window function:

[0158]

[0159] By scaling and translating the Gaussian window function, we get the formula for the S transform:

[0160]

[0161] Add the same time columns of the two-dimensional matrix after S transformation to obtain the corresponding spectrum diagram under the data.

[0162] In some embodiments, the control strategy module 210 is further configured to perform risk assessment on areas that may be affected along the path of fire spread, including assessing the population density, natural resource conditions, and infrastructure of the area to determine the potential risk level of fire spread to the area;

[0163] Divide the affected area into different emergency response areas according to the degree of potential risk;

[0164] A spread control strategy for the emergency response area is formulated based on the path and speed of the fire spread and the potential risk level; the spread control strategy includes determining the evacuation area, the blockade area, and the emergency refuge area.

[0165] In some embodiments, the second acquisition module 202 is further configured to determine that the meteorological data includes temperature, humidity, wind speed, and wind direction. The topographic data includes elevation, slope, and slope direction.

[0166] For other details of implementing the above technical solution in each module in the wildfire spread trend prediction device, please refer to the description of the wildfire spread trend prediction method provided above, which will not be repeated here.

[0167] In an embodiment of the present application, a computer device is provided. Figure 4 , Figure 4 3 is a schematic diagram of the structure of a computer device in an embodiment. The device includes a memory 301 and a processor 302. The memory 301 stores a computer program. When the computer program is executed by the processor 302, the processor 302 performs the following steps:

[0168] Acquire topographic data of the target area, wherein the topographic data includes terrain elevation, vegetation coverage, and terrain slope;

[0169] Obtain weather data for the target area;

[0170] Preprocessing the topographic and geomorphic data and the meteorological data to obtain preprocessed topographic and geomorphic data and meteorological data;

[0171] Extracting feature information related to fire spread from the preprocessed topographic data and meteorological data, and constructing an initial feature set;

[0172] The initial feature set is evaluated by a feature importance evaluation method to identify feature information that has a significant impact on the spread of fire, and a processed data set is obtained;

[0173] Dividing the processed data set into a training set, a validation set and a test set;

[0174] Construct the S-transform mathematical model of fire spread;

[0175] The training of the S-transform mathematical model is completed through the training set, the validation set and the test set;

[0176] Based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted through the trained S-transform mathematical model.

[0177] Among them, the processor 302 can also be called a CPU (Central Processing Unit), and the processor 302 may be an integrated circuit chip with signal processing capabilities; the processor 302 can also be a general-purpose processor, DSP (Digital Signal Process), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, among which the general-purpose processor can be a microprocessor or the processor 302 can also be any conventional processor, etc.

[0178] In an embodiment of the present application, a computer readable storage medium is provided. Figure 5 , Figure 5 The structure diagram of a computer-readable storage medium in an embodiment is a schematic diagram, on which a readable computer program 401 is stored; wherein the computer program 401 may be stored in the above storage medium in the form of a software product, and includes a number of instructions for causing a computer device (which may be a personal computer, a server machine, or a network device, etc.) or a processor to perform the following steps:

[0179] Acquire topographic data of the target area, wherein the topographic data includes terrain elevation, vegetation coverage, and terrain slope;

[0180] Obtain weather data for the target area;

[0181] Preprocessing the topographic and geomorphic data and the meteorological data to obtain preprocessed topographic and geomorphic data and meteorological data;

[0182] Extracting feature information related to fire spread from the preprocessed topographic data and meteorological data, and constructing an initial feature set;

[0183] The initial feature set is evaluated by a feature importance evaluation method to identify feature information that has a significant impact on the spread of fire, and a processed data set is obtained;

[0184] Dividing the processed data set into a training set, a validation set and a test set;

[0185] Construct the S-transform mathematical model of fire spread;

[0186] The training of the S-transform mathematical model is completed through the training set, the validation set and the test set;

[0187] Based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted through the trained S-transform mathematical model.

[0188] The aforementioned storage media include: U disk, mobile hard disk, magnetic disk or CD, ROM (Read-Only Memory), RAM (Random Access Memory) and other media that can store program code, or terminal devices such as computers, service machines, mobile phones, and tablets.

[0189] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and 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. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can 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 many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0190] 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.

[0191] The above-described 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 includes: Acquire topographic data of the target area, wherein the topographic data includes terrain elevation, vegetation coverage, and terrain slope; Obtain weather data for the target area; Preprocessing the topographic and geomorphic data and the meteorological data to obtain preprocessed topographic and geomorphic data and meteorological data; Extracting feature information related to fire spread from the preprocessed topographic data and meteorological data, and constructing an initial feature set; The initial feature set is evaluated by a feature importance evaluation method to identify feature information that has a significant impact on the spread of fire, and a processed data set is obtained; Dividing the processed data set into a training set, a validation set and a test set; Construct the S-transform mathematical model of fire spread; The training of the S-transform mathematical model is completed through the training set, the validation set and the test set; Based on the topographic data and meteorological data of the target area to be detected, the path and speed of fire spread are predicted through the trained S-transform mathematical model.

2. The method for predicting the spread trend of wildfire according to claim 1, characterized in that: The method further comprises: predicting the path and speed of fire spread based on the topographic data and meteorological data of the target area to be detected by using the trained S-transform mathematical model; and then: A spread control strategy is determined based on the path and speed of the fire spread.

3. The method for predicting the spread trend of wildfire according to claim 2, characterized in that: The preprocessing of the topographic data and the meteorological data to obtain the preprocessed topographic data and the meteorological data specifically includes: Integrate the topographic data and meteorological data, and convert their formats; The integrated data is subjected to noise removal, outlier and erroneous data processing, including detecting and repairing missing values ​​in the data, processing duplicate data, and converting the cleaned data into a unified format such as CSV, JSON, XML or array matrix.

4. The method for predicting the spread trend of wildfire according to claim 3, characterized in that: The construction of the S-transform mathematical model of fire spread specifically includes: Use the S-transform machine learning algorithm in a GIS environment to build a fire spread model; Processing the topographic data and meteorological data of the target area by using an S-transform algorithm to obtain a two-dimensional matrix of time and frequency; Adding each column of the two-dimensional matrix related to time and frequency to obtain an energy spectrum graph related only to frequency; The historical topographic data and their spreading trends are processed by S transformation to obtain the energy spectrum database; The energy spectrum is analyzed by S-transform and compared with the S-transform mathematical model, and the model with the highest similarity is selected to determine the spreading trend.

5. The method for predicting the spread trend of wildfire according to claim 4, characterized in that: The training of the S transform mathematical model is completed by using the training set, the validation set and the test set, specifically including: The Fourier transform is: By adding the window function g(t) to the time series h(t), the spectrum becomes: For the S transform, define the window function: By scaling and translating the Gaussian window function, we get the formula for the S transform: Add the same time columns of the two-dimensional matrix after S transformation to obtain the corresponding spectrum diagram under the data.

6. The method for predicting the spread trend of wildfires according to claim 5, characterized in that: Determining the fire spread control strategy according to the fire spread path and speed specifically includes: Conduct risk assessments on areas that may be affected along the path of fire spread, including assessing the population density, natural resources, and infrastructure of the area to determine the potential risk level of fire spread to the area; Divide the affected area into different emergency response areas according to the degree of potential risk; A spread control strategy for the emergency response area is formulated based on the path and speed of the fire spread and the potential risk level; the spread control strategy includes determining the evacuation area, the blockade area, and the emergency refuge area.

7. The method for predicting the spread trend of wildfires according to claim 6, characterized in that: The meteorological data include temperature, humidity, wind speed and wind direction. The topographic data include elevation, slope and slope direction.

8. A device for predicting the spread of wildfires, characterized in that: The device comprises: The first acquisition module is used to acquire the topographic data of the target area, wherein the topographic data includes the terrain elevation, vegetation coverage, and terrain slope; The second acquisition module acquires the meteorological data of the target area; A preprocessing module, used for preprocessing the topographic data and meteorological data to obtain preprocessed topographic data and meteorological data; A feature extraction module, used to extract feature information related to fire spread from the pre-processed topographic data and meteorological data, and to construct an initial feature set; An evaluation module, used to evaluate the initial feature set by a feature importance evaluation method, identify feature information that has a significant impact on fire spread, and obtain a processed data set; A data set module, used to divide the processed data set into a training set, a validation set and a test set; Model building module, used to build the S-transform mathematical model of fire spread; A model training module, used to complete the training of the S-transform mathematical model through the training set, the validation set and the test set; The fire spread module is used to predict the path and speed of fire spread based on the topographic data and meteorological data of the target area to be detected through the trained S-transform mathematical model.

9. A computer device 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 steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

Images