Method and related device for predicting wildfire spread and risk assessment in power transmission corridor based on satellite internet

Abstract

The present invention provides a method and related device for predicting the spread of wildfires in power transmission corridors based on satellite Internet and risk assessment, and relates to the technical field of disaster prevention and early warning of power systems. A fire risk factor database is constructed based on multimodal data of the satellite-ground fusion network model; the random forest algorithm is used to screen out key factors and determine the fire risk contribution corresponding to the key factors; based on the power wildfire hazard factors, similar features of the transmission corridor neighborhood are extracted, and the clustering algorithm is used to divide the sub-areas; a fire risk weighted scoring model is constructed to identify high, medium, and low risk areas; a wildfire spread model is constructed based on meteorological observation data, a fire risk state transfer matrix is ​​established, and a dynamic Bayesian network is combined to predict the fire spread path in real time; the risk of wildfire damage to transmission lines is assessed, and a damage risk classification map of the transmission corridor is generated. The technical effect of dynamic assessment of wildfire risk and accurate prediction of spread trend is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of disaster prevention and early warning of power systems, and in particular to a method and related device for predicting the spread of wildfires and assessing the risks of wildfires in power transmission corridors based on satellite Internet. Background Art

[0002] The frequency and intensity of wildfires are increasing. Under adverse weather conditions such as drought and wind, wildfires spread rapidly, threatening the safety of human society and the operation of infrastructure. Transmission lines are key components of the power system. Once they are attacked by wildfires, they may cause widespread power outages and damage to facilities, resulting in huge economic losses and social impacts.

[0003] Existing methods for predicting wildfire spread trends can be roughly divided into prediction methods based on physical models, statistical models, and probability models. However, such methods can only make predictions based on fixed conditions or historical data, lack dynamic response to sudden environmental changes, and are difficult to meet the needs of rapidly changing fire environments. For scenarios where the transmission corridor spans a large area and the surrounding environment is complex and changeable, the conventional use of historical fire data to predict the direction of fire spread is not accurate enough, and does not fully consider the impact of human factors such as geography and power load distribution on wildfire spread. There is a lack of key monitoring of high-risk areas and dynamic adjustment and adaptive capabilities based on real-time data, resulting in insufficient real-time and accuracy of prediction results.

[0004] Therefore, how to achieve dynamic assessment of wildfire risks and accurate prediction of their spread trends has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] In order to achieve dynamic assessment of wildfire risks and accurate prediction of spread trends, the present application provides a method and related device for predicting the spread of wildfires in power transmission corridors based on satellite Internet.

[0006] In the first aspect, the present application provides a method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet, which adopts the following technical solutions:

[0007] A method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet, comprising:

[0008] Construct a fire risk factor database based on multimodal data of satellite-ground fusion network model;

[0009] Applying a random forest algorithm to screen out key factors in the fire risk factor database and determine the fire risk contribution corresponding to the key factors;

[0010] Based on the power wildfire hazard factors, similar characteristics of the transmission corridor neighborhood are extracted, and the study area is divided into several sub-areas with similar power risk characteristics using a clustering algorithm.

[0011] Constructing a fire risk weighted scoring model, dynamically adjusting the regional division of the sub-regions based on the fire risk contribution, and identifying high, medium and low risk areas;

[0012] Based on meteorological observation data, a wildfire spread model is built, a fire risk state transfer matrix is ​​established, and a dynamic Bayesian network is used to predict the fire spread path in real time.

[0013] The fire spread path, fire intensity and equipment distribution characteristics are comprehensively considered to assess the damage risk of wildfires to transmission lines and generate a damage risk classification map for the transmission corridor.

[0014] Optionally, before the step of constructing a fire risk factor database based on multimodal data of the satellite-ground fusion network model, the step further includes:

[0015] Construct a multi-scale wildfire monitoring system based on a satellite-ground fusion network model to collect infrared thermal images, meteorological parameters, fire location and power facility status information;

[0016] The infrared thermal image, meteorological parameters, fire location and power facility status information are fused using a spatiotemporal registration algorithm to generate multimodal data.

[0017] Optionally, the step of applying the random forest algorithm to screen out key factors in the fire risk factor database and determining the fire risk contribution corresponding to the key factors includes:

[0018] Randomly select from the fire risk database Training dataset and the corresponding Datasets for prediction testing , where the factors in each training set do not completely overlap, and the prediction test set is composed of the corresponding non-training set inclusion factors;

[0019] For each training data set , use random forest to build a decision tree model and Make a prediction and count the prediction error of the corresponding data: ;

[0020] The fire risk factor in each data set is denoted as , and predict the test data set The factors in Apply perturbations and use the trained model to Make a prediction and calculate the new prediction error ;

[0021] Calculate the mean change in the prediction error of each factor as the importance of the factor :

[0022] ;

[0023] In the formula, Fire risk factor The importance of.

[0024] Optionally, the step of extracting similar features of the transmission corridor neighborhood based on the power wildfire hazard factor and dividing the study area into a number of sub-areas with similar power risk features using a clustering algorithm includes:

[0025] The transmission corridor area is evenly gridded, and the initial cluster center and cluster number are determined according to the characteristics of the power wildfire hazard factor, and the factor correlation based on the hazard factor is defined. and dispersion , thus forming a comprehensive evaluation index of the effect of cluster analysis , to determine the optimal division of transmission corridor areas value:

[0026] ;

[0027] In the formula, Indicates Data of power facilities, Indicates Cluster centers;

[0028] Calculate the distances between all power facilities , and according to Value selects the radius of the circular radiation model , calculate the number of other power facilities within the radiation range of each power facility , and set the threshold The threshold value is the total number of power facilities and the number of clusters in the transmission corridor area. The ratio of

[0029] Putting it all together and put it into the dataset At the same time, find the largest As the first cluster center , and then move it to the collection Then select Zhongyu The farthest power facility is the second cluster center. , and then move it to the collection In the same way, we can get Cluster centers are formed, and all power facilities in the transmission corridor area are allocated to the nearest cluster centers, thus forming Regions with similar power risk characteristics:

[0030] ;

[0031] In the formula, is the total number of power facilities in the region, Indicates that two power facilities are dimensional distance, x, y respectively represent the location coordinates of different power facilities in each dimension.

[0032] Optionally, the step of constructing a fire risk weighted scoring model, dynamically adjusting the regional division of the sub-regions in combination with the fire risk contribution, and identifying high, medium and low risk areas includes:

[0033] Construct a fire risk weighted scoring model;

[0034] ;

[0035] Where: According to the importance The weighted value assigned to each factor, To standardize the values ​​of the key fire risk factors in different dimensions, Provide a fire risk score for each zone;

[0036] In the sub-area, the preliminary division is analyzed Regional fire impact factors with similar power risk characteristics are assigned weights based on the risk contribution of the factors;

[0037] The fire risk score of the area is updated based on a dynamic scoring mechanism to divide the sub-area into high, medium and low risk areas by setting thresholds.

[0038] Optionally, the steps of constructing a wildfire spread model based on meteorological observation data, establishing a fire risk state transfer matrix, and predicting the fire spread path in real time in combination with a dynamic Bayesian network include:

[0039] Construct a wildfire spread model based on meteorological observation data;

[0040] ;

[0041] In the formula, The speed at which wildfires spread. It is the ideal wildfire spread speed under windless and flat slope conditions without other interference. is the wildfire radiation factor, and its value is related to the heat released by the fire. It is the speed factor brought by wildfire radiation, and its value is related to the thickness of the flame. represents the angle between the mountain fire and the ground normal, It represents the angle between the slope and the horizontal plane. and Respectively represent the wind speed and the vertical propagation speed of smoke;

[0042] Construct fire risk state transfer matrix;

[0043] ;

[0044] In the formula, Indicates that the fire spread from the area To the region The probability of fire spreading from one cluster area to all adjacent cluster areas is 1. , , It represents the three most important factors affecting fire spread: weather, fuel and terrain;

[0045] In combination with the wildfire spread model, the possibility of fire spreading from one area to an adjacent area is quantified through the fire risk state transfer matrix;

[0046] The dynamic Bayesian network is used to simulate the state evolution of the fire and predict the fire spread path in real time.

[0047] Optionally, the step of comprehensively evaluating the risk of wildfire damage to power transmission lines and generating a damage risk classification map for power transmission corridors by comprehensively considering the fire spread path, fire intensity and equipment distribution characteristics includes:

[0048] Integrate the fire spread prediction path to identify fire affected areas and key equipment;

[0049] The affected area of ​​wildfires is divided into three types: high-temperature combustion area, charge accumulation area, and smoke particle area. By analyzing the historical data of line tripping caused by wildfires and considering the wildfire spread conditions and line breakdown probability, the tripping probability of transmission lines under wildfire conditions is obtained:

[0050] ;

[0051] In the formula, and are the space gap voltage and space gap breakdown voltage under wildfire conditions, It is the sum of the spatial gap breakdown voltages of the three regions: high temperature combustion region, charge accumulation region, and smoke particle region. is the phase voltage of the transmission line operation;

[0052] The impact of line tripping probability and fault propagation on power grid operation is considered, and the problems caused by fire are simulated based on power grid flow analysis, and a damage assessment model is constructed;

[0053] Among them, the construction method of the comprehensive probability model of transmission line wildfire tripping is:

[0054] ;

[0055] Where: is the maximum spreading speed of the fire point along the line connecting the police tower and the fire point, This is the wildfire data update cycle. is the distance between the fire point and the alarm tower;

[0056] The damage assessment model is constructed as follows:

[0057] ;

[0058] Where: is the risk value of wildfire damage to transmission lines, For the The probability of a line tripping due to a wildfire, For the The power grid damage value caused by tripping of the line, are load loss and voltage fluctuation respectively. are weight coefficients. Adjusting these two parameters can control the damage model according to different actual conditions.

[0059] Generate a transmission corridor damage risk classification map based on the output results of the damage assessment model.

[0060] In a second aspect, the present application provides a transmission corridor wildfire spread prediction and risk assessment system based on satellite Internet, and the transmission corridor wildfire spread prediction and risk assessment system based on satellite Internet includes:

[0061] Data construction module, used to construct a fire risk factor database based on multimodal data of satellite-ground fusion network mode;

[0062] A contribution module, for applying a random forest algorithm to screen out key factors in the fire risk factor database and determine the fire risk contribution corresponding to the key factors;

[0063] The sub-region module is used to extract similar features of the transmission corridor neighborhood based on the power wildfire hazard factor, and use a clustering algorithm to divide the study area into several sub-regions with similar power risk characteristics;

[0064] A regional division module is used to construct a fire risk weighted scoring model, dynamically adjust the regional division of the sub-regions based on the fire risk contribution, and identify high, medium and low risk areas;

[0065] The path prediction module is used to build a wildfire spread model based on meteorological observation data, establish a fire risk state transfer matrix, and combine the dynamic Bayesian network to predict the fire spread path in real time;

[0066] The classification map module is used to comprehensively consider the fire spread path, fire intensity and equipment distribution characteristics, assess the damage risk of wildfires to transmission lines, and generate a damage risk classification map for the transmission corridor.

[0067] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, wherein the processor executes the method described above when running computer instructions stored in the memory.

[0068] In a fourth aspect, the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a computer, enable the computer to execute the method described above.

[0069] In summary, the present application includes the following beneficial technical effects:

[0070] This application builds a fire risk factor database based on multimodal data of the satellite-ground fusion network model; uses the random forest algorithm to screen out key factors and determine the fire risk contribution corresponding to the key factors; based on the power wildfire hazard factors, extracts similar features of the transmission corridor neighborhood, and uses a clustering algorithm to divide sub-areas; builds a fire risk weighted scoring model to identify high, medium, and low risk areas; builds a wildfire spread model based on meteorological observation data, establishes a fire risk state transfer matrix, and combines a dynamic Bayesian network to predict the fire spread path in real time; evaluates the risk of wildfire damage to transmission lines, and generates a damage risk classification map for the transmission corridor. The technical effect of dynamic assessment of wildfire risk and accurate prediction of spread trends is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application;

[0072] Figure 2 It is a flowchart of the first embodiment of the method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet in the present application;

[0073] Figure 3 This is a dynamic Bayesian network topology diagram of the transmission corridor wildfire spread prediction and risk assessment method based on satellite Internet in this application. Figure 3 (a) is the initial network of the dynamic Bayesian network topology, Figure 3 (b) is a schematic diagram of the transfer network;

[0074] Figure 4 It is a structural block diagram of the first embodiment of the transmission corridor wildfire spread prediction and risk assessment system based on satellite Internet in this application. DETAILED DESCRIPTION

[0075] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below through the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0076] Reference Figure 1 , Figure 1 A schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0077] like Figure 1 As shown, the computer device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0078] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0079] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a satellite Internet-based transmission corridor wildfire spread prediction and risk assessment program.

[0080] exist Figure 1In the computer device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the present application can be set in the computer device, and the computer device calls the satellite Internet-based power transmission corridor wildfire spread prediction and risk assessment program stored in the memory 1005 through the processor 1001, and executes the satellite Internet-based power transmission corridor wildfire spread prediction and risk assessment method provided in the embodiment of the present application.

[0081] The present application embodiment provides a method for predicting the spread of wildfires and assessing the risk of wildfires in power transmission corridors based on satellite Internet. Figure 2 , Figure 2 This is a flow chart of the first embodiment of the method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet in this application.

[0082] In this embodiment, the method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet includes the following steps:

[0083] Step S10: Construct a fire risk factor database based on the multimodal data of the satellite-ground fusion network model.

[0084] It should be noted that before the step of constructing a fire risk factor database based on multimodal data of a satellite-ground fusion network model, it also includes: constructing a multi-scale wildfire monitoring system based on a satellite-ground fusion network model to collect infrared thermal maps, meteorological parameters, fire point locations and power facility status information; using a spatiotemporal registration algorithm to fuse infrared thermal maps, meteorological parameters, fire point locations and power facility status information to generate multimodal data.

[0085] In the specific implementation, the fire risk factors in the transmission corridor environment are analyzed, and a multi-scale wildfire monitoring system architecture based on the satellite-ground fusion network model is constructed; in the satellite-ground fusion network model, multimodal data such as infrared thermal images, meteorological parameters, fire point locations and real-time status information of power facilities are collected, and the data is transmitted to the data processing center in real time through satellite Internet for fusion processing; the observation data of different sensors are integrated using the spatiotemporal registration data fusion algorithm to ensure spatiotemporal consistency, and a fire risk factor database is constructed to provide multi-scale data support for dynamic monitoring and spread prediction of wildfires.

[0086] Step S20: Apply the random forest algorithm to screen out key factors in the fire risk factor database and determine the fire risk contribution corresponding to the key factors.

[0087] It can be understood that the steps of applying the random forest algorithm to screen out key factors in the fire risk factor database and determine the fire risk contribution corresponding to the key factors include:

[0088] Random sampling from the fire risk database Training dataset and the corresponding Datasets for prediction testing , where the factors in each training set do not completely overlap, and the prediction test set is composed of the corresponding non-training set inclusion factors;

[0089] For each training data set , use random forest to build a decision tree model and Make a prediction and count the prediction error of the corresponding data: ;

[0090] The fire risk factor in each data set is denoted as , and predict the test data set The factors in Apply perturbations and use the trained model to Make a prediction and calculate the new prediction error ;

[0091] Calculate the mean change in the prediction error of each factor as the importance of the factor :

[0092] ;

[0093] In the formula, Fire risk factor The importance of.

[0094] In the specific implementation, the change in the prediction error after the disturbance can reflect the factor The contribution to the model performance.

[0095] It should be noted that the greater the change in the prediction error, the greater the impact of the factor on the model prediction performance and the higher its importance. Then, according to the calculated factor importance, all factors are sorted and low-contribution factors are eliminated to finally generate an optimized risk factor set.

[0096] Step S30: Based on the power wildfire hazard factor, similar features of the transmission corridor neighborhood are extracted, and a clustering algorithm is used to divide the study area into several sub-areas with similar power risk characteristics.

[0097] It should be noted that based on the power wildfire hazard factor, similar characteristics of the transmission corridor neighborhood are extracted, and the clustering algorithm is used to divide the study area into several sub-areas with similar power risk characteristics, including: uniformly gridding the transmission corridor area, determining the initial cluster center and the number of clusters according to the characteristics of the power wildfire hazard factor, and defining the factor correlation degree based on the hazard factor. and dispersion , thus forming a comprehensive evaluation index of the effect of cluster analysis , to determine the optimal division of transmission corridor areas value:

[0098] ;

[0099] In the formula, Indicates Data of power facilities, Indicates Cluster centers;

[0100] Calculate the distances between all power facilities , and according to Value selects the radius of the circular radiation model , calculate the number of other power facilities within the radiation range of each power facility , and set the threshold The threshold value is the total number of power facilities and the number of clusters in the transmission corridor area. The ratio of

[0101] Putting it all together and put it into the dataset At the same time, find the largest As the first cluster center , and then move it to the collection Then select Zhongyu The farthest power facility is the second cluster center. , and then move it to the collection In the same way, we can get Cluster centers are formed, and all power facilities in the transmission corridor area are allocated to the nearest cluster centers, thus forming Regions with similar power risk characteristics:

[0102] ;

[0103] In the formula, is the total number of power facilities in the region, Indicates that two power facilities are dimensional distance, x, y respectively represent the location coordinates of different power facilities in each dimension.

[0104] Step S40: construct a fire risk weighted scoring model, dynamically adjust the regional division of sub-regions based on the fire risk contribution, and identify high, medium and low risk areas.

[0105] It is understood that the steps of constructing a fire risk weighted scoring model, dynamically adjusting the regional division of sub-regions in combination with fire risk contribution, and identifying high, medium, and low risk regions include: constructing a fire risk weighted scoring model;

[0106] ;

[0107] Where: According to the importance The weighted value assigned to each factor, To standardize the values ​​of the key fire risk factors in different dimensions, Score the fire risk for each zone; analyze the initial zone in the sub-zones The fire impact factors of regions with similar power risk characteristics are calculated, and weights are assigned based on the risk contribution of the factors. The fire risk scores of the regions are updated based on a dynamic scoring mechanism to divide the sub-regions into high, medium, and low risk areas by setting thresholds.

[0108] Step S50: Build a wildfire spread model based on meteorological observation data, establish a fire risk state transfer matrix, and use a dynamic Bayesian network to predict the fire spread path in real time.

[0109] It should be noted that the wildfire spread model is built based on meteorological observation data, and the possibility of fire spreading from one area to an adjacent area is quantified through the fire risk state transfer matrix; combined with the dynamic Bayesian network, the state evolution of the fire at different time steps is simulated, and the fire spread path is predicted in real time; according to the updated results of the dynamic risk score, the state transfer matrix is ​​adaptively adjusted to ensure that the fire spread prediction can accurately reflect the current risk status. The dynamic Bayesian network topology in this embodiment is shown in the figure below: Figure 3 As shown, Figure 3 (a) Initial network representing the dynamic Bayesian network topology; Figure 3 (b) Representation of the transition network of the dynamic Bayesian network.

[0110] In the specific implementation, the method of establishing the fire spread prediction model based on satellite meteorological observation and data-driven transmission is as follows:

[0111] ;

[0112] In the formula, The speed at which wildfires spread. It is the ideal wildfire spread speed under windless and flat slope conditions without other interference. is the wildfire radiation factor, and its value is related to the heat released by the fire. It is the speed factor brought by wildfire radiation, and its value is related to the thickness of the flame. represents the angle between the mountain fire and the ground normal, It represents the angle between the slope and the horizontal plane. and They represent the wind speed and the vertical propagation speed of smoke respectively.

[0113] Furthermore, the fire risk state transfer matrix is ​​constructed as follows:

[0114] ;

[0115] Where: Indicates that the fire spread from the area To the region The probability of fire spreading from one cluster area to all adjacent cluster areas is 1. , , Represents the three most important fire propagation influencing factors of meteorology, fuel, and terrain. This set of propagation influencing factors is obtained based on the optimal factor set screened in S2.

[0116] Step S60: Comprehensively consider the fire spread path, fire intensity and equipment distribution characteristics, assess the risk of wildfire damage to transmission lines, and generate a damage risk classification map for the transmission corridor.

[0117] It should be noted that the steps of comprehensively evaluating the damage risk of wildfires to transmission lines and generating a damage risk classification map for transmission corridors by comprehensively considering the fire spread path, fire intensity and equipment distribution characteristics include: comprehensively predicting the fire spread path to identify the fire-affected area and key equipment; dividing the wildfire-affected area into three types of areas: high-temperature combustion area, charge accumulation area and smoke particle area; and analyzing the historical data of line tripping caused by wildfires, considering the wildfire spread conditions and line breakdown probability, and obtaining the transmission line tripping probability under wildfire conditions:

[0118] ;

[0119] In the formula, and are the space gap voltage and space gap breakdown voltage under wildfire conditions, It is the sum of the spatial gap breakdown voltages of the three regions: high temperature combustion region, charge accumulation region, and smoke particle region. The phase voltage of the transmission line operation; the impact of line trip probability and fault propagation on the operation of the power grid is comprehensively considered, and the problems caused by the fire are simulated based on the power grid flow analysis, and a damage assessment model is constructed;

[0120] Among them, the construction method of the comprehensive probability model of transmission line wildfire tripping is:

[0121] ;

[0122] Where: is the maximum spreading speed of the fire point along the line connecting the police tower and the fire point, This is the wildfire data update cycle. is the distance between the fire point and the alarm tower;

[0123] The damage assessment model is constructed as follows:

[0124] ;

[0125] Where: is the risk value of wildfire damage to transmission lines, For the The probability of a line tripping due to a wildfire, For the The power grid damage value caused by tripping of the line, are load loss and voltage fluctuation respectively. are weight coefficients. Adjusting these two parameters can control the damage model according to different actual conditions. A transmission corridor damage risk grading map is generated based on the output results of the damage assessment model.

[0126] It should be noted that in this embodiment, based on the real-time fusion and coordination of multimodal data in the satellite-ground fusion network mode, a fire risk factor database is constructed; key factors in the database are screened, and the contribution of each key factor to the fire risk is calculated; the similar characteristics of the transmission corridor neighborhood based on power factors are studied, and the preliminary division of the study area is completed based on the power wildfire hazard factor; a quantitative scoring model is constructed to achieve dynamic adjustment and refined division of the region according to the real-time risk score; a wildfire spread model driven by meteorological observation data is constructed, and a fire risk state transfer matrix is ​​established to achieve real-time prediction of the fire spread path; a comprehensive analysis of the fire spread path, fire intensity and equipment distribution characteristics, a quantitative assessment of the damage risk of wildfires to transmission lines, and a damage risk classification map of the transmission corridor.

[0127] This embodiment builds a fire risk factor database based on multimodal data of the satellite-ground fusion network model; uses the random forest algorithm to screen out key factors and determine the fire risk contribution corresponding to the key factors; based on the power wildfire hidden danger factors, extracts similar features of the transmission corridor neighborhood, and uses a clustering algorithm to divide the sub-areas; builds a fire risk weighted scoring model to identify high, medium, and low risk areas; drives the construction of a wildfire spread model based on meteorological observation data, establishes a fire risk state transfer matrix, and combines a dynamic Bayesian network to predict the fire spread path in real time; evaluates the risk of wildfire damage to transmission lines, and generates a damage risk classification map for the transmission corridor. The technical effect of dynamic assessment of wildfire risk and accurate prediction of spread trends is achieved.

[0128] In addition, an embodiment of the present application also proposes a computer-readable storage medium, on which is stored a program for predicting the spread of wildfires and risk assessment in transmission corridors based on satellite Internet. When the program for predicting the spread of wildfires and risk assessment in transmission corridors based on satellite Internet is executed by a processor, the steps of the method for predicting the spread of wildfires and risk assessment in transmission corridors based on satellite Internet as described above are implemented.

[0129] Reference Figure 4 , Figure 4 This is a structural block diagram of the first embodiment of the satellite Internet-based transmission corridor wildfire spread prediction and risk assessment system of the present application.

[0130] like Figure 4 As shown, the transmission corridor wildfire spread prediction and risk assessment system based on satellite Internet proposed in the embodiment of the present application includes:

[0131] A data construction module 10 is used to construct a fire risk factor database based on multimodal data of a satellite-ground fusion network model;

[0132] The contribution module 20 is used to apply the random forest algorithm to screen out key factors in the fire risk factor database and determine the fire risk contribution corresponding to the key factors;

[0133] The sub-region module 30 is used to extract similar features of the transmission corridor neighborhood based on the power wildfire hazard factor, and divide the study area into several sub-regions with similar power risk characteristics using a clustering algorithm;

[0134] A regional division module 40 is used to construct a fire risk weighted scoring model, dynamically adjust the regional division of sub-regions based on the fire risk contribution, and identify high, medium and low risk areas;

[0135] The path prediction module 50 is used to drive the construction of a wildfire spread model based on meteorological observation data, establish a fire risk state transfer matrix, and combine the dynamic Bayesian network to predict the fire spread path in real time;

[0136] The classification diagram module 60 is used to comprehensively consider the fire spread path, fire intensity and equipment distribution characteristics, evaluate the damage risk of wildfires to transmission lines, and generate a damage risk classification diagram for the transmission corridor.

[0137] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present application. In specific applications, technicians in this field can make settings as needed, and the present application does not impose any limitation on this.

[0138] This embodiment builds a fire risk factor database based on multimodal data of the satellite-ground fusion network model; uses the random forest algorithm to screen out key factors and determine the fire risk contribution corresponding to the key factors; based on the power wildfire hidden danger factors, extracts similar features of the transmission corridor neighborhood, and uses a clustering algorithm to divide the sub-areas; builds a fire risk weighted scoring model to identify high, medium, and low risk areas; drives the construction of a wildfire spread model based on meteorological observation data, establishes a fire risk state transfer matrix, and combines a dynamic Bayesian network to predict the fire spread path in real time; evaluates the risk of wildfire damage to transmission lines, and generates a damage risk classification map for the transmission corridor. The technical effect of dynamic assessment of wildfire risk and accurate prediction of spread trends is achieved.

[0139] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present application. In practical applications, technicians in this field can select part or all of it according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0140] In addition, for technical details not described in detail in this embodiment, please refer to the method for predicting the spread of wildfires and risk assessment in power transmission corridors based on satellite Internet provided in any embodiment of the present application, which will not be repeated here.

[0141] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0142] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0143] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0144] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet, characterized in that: include: Construct a fire risk factor database based on multimodal data of satellite-ground fusion network model; Applying a random forest algorithm to screen out key factors in the fire risk factor database and determine the fire risk contribution corresponding to the key factors; Based on the power wildfire hazard factors, similar characteristics of the transmission corridor neighborhood are extracted, and the study area is divided into several sub-areas with similar power risk characteristics using a clustering algorithm. Constructing a fire risk weighted scoring model, dynamically adjusting the regional division of the sub-regions based on the fire risk contribution, and identifying high, medium and low risk areas; Based on meteorological observation data, a wildfire spread model is built, a fire risk state transfer matrix is ​​established, and a dynamic Bayesian network is used to predict the fire spread path in real time. Comprehensively evaluate the fire spread path, fire intensity and equipment distribution characteristics to assess the risk of wildfire damage to transmission lines and generate a damage risk classification map for the transmission corridor; The steps of constructing a fire risk weighted scoring model, dynamically adjusting the regional division of the sub-regions in combination with the fire risk contribution, and identifying high, medium and low risk areas include: Construct a fire risk weighted scoring model; ； Where: According to the importance The weighted value assigned to each factor, To standardize the values ​​of the key fire risk factors in different dimensions, Provide a fire risk score for each zone; In the sub-area, the preliminary division is analyzed Regional fire impact factors with similar power risk characteristics are assigned weights based on the risk contribution of the factors; updating the fire risk score of the area based on a dynamic scoring mechanism to divide the sub-area into high, medium and low risk areas by setting thresholds; The steps of building a wildfire spread model based on meteorological observation data, establishing a fire risk state transfer matrix, and predicting the fire spread path in real time by combining a dynamic Bayesian network include: Construct a wildfire spread model based on meteorological observation data; ； In the formula, The speed at which wildfires spread. It is the ideal wildfire spread speed under windless and flat slope conditions without other interference. is the wildfire radiation factor, and its value is related to the heat released by the fire. It is the speed factor brought by wildfire radiation, and its value is related to the thickness of the flame. represents the angle between the mountain fire and the ground normal, It represents the angle between the slope and the horizontal plane. and Respectively represent the wind speed and the vertical propagation speed of smoke; Construct fire risk state transfer matrix; ； In the formula, Indicates that the fire spread from the area To the region The probability of fire spreading from one cluster area to all adjacent cluster areas is 1. , , It represents the three most important factors affecting fire spread: weather, fuel and terrain; In combination with the wildfire spread model, the possibility of fire spreading from one area to an adjacent area is quantified through the fire risk state transfer matrix; The dynamic Bayesian network is used to simulate the state evolution of the fire and predict the fire spread path in real time.

2. According to claim 1, a method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet is characterized in that: Before the step of constructing a fire risk factor database based on the multimodal data of the satellite-ground fusion network model, the method further includes: Construct a multi-scale wildfire monitoring system based on a satellite-ground fusion network model to collect infrared thermal images, meteorological parameters, fire location and power facility status information; The infrared thermal image, meteorological parameters, fire location and power facility status information are fused using a spatiotemporal registration algorithm to generate multimodal data.

3. According to claim 1, a method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet is characterized in that: The step of applying the random forest algorithm to screen out key factors in the fire risk factor database and determining the fire risk contribution corresponding to the key factors comprises: Randomly select from the fire risk factor database Training dataset and the corresponding Datasets for prediction testing , where the factors in each training set do not completely overlap, and the prediction test set is composed of the corresponding non-training set inclusion factors; For each training data set , use random forest to build a decision tree model and Make a prediction and count the prediction error of the corresponding data: ; The fire risk factor in each data set is denoted as , and predict the test data set The factors in Apply perturbations and use the trained model to Make a prediction and calculate the new prediction error ; Calculate the mean change in the prediction error of each factor as the importance of the factor : ； In the formula, Fire risk factor The importance of.

4. According to claim 1, a method for predicting and assessing wildfire spread in power transmission corridors based on satellite Internet is characterized in that: The steps of extracting similar features of the transmission corridor neighborhood based on the power wildfire hazard factor and using a clustering algorithm to divide the study area into a number of sub-areas with similar power risk features include: The transmission corridor area is evenly gridded, and the initial cluster center and cluster number are determined according to the characteristics of the power wildfire hazard factor, and the factor correlation based on the hazard factor is defined. and dispersion , thus forming a comprehensive evaluation index of the effect of cluster analysis , to determine the optimal division of the transmission corridor area value: ； In the formula, Indicates Data of power facilities, Indicates Cluster centers; Calculate the distances between all power facilities , and according to k Value selects the radius of the circular radiation model , calculate the number of other power facilities within the radiation range of each power facility , and set the threshold The threshold value is the total number of power facilities and the number of clusters in the transmission corridor area. k The ratio of Putting it all together and put it into the dataset At the same time, find the largest As the first cluster center , and then move it to the collection Then select Zhongyu The farthest power facility is the second cluster center. , and then move it to the collection In the same way, we can get k Cluster centers are formed, and all power facilities in the transmission corridor area are allocated to the nearest cluster centers, thus forming k Regions with similar power risk characteristics: ； In the formula, is the total number of power facilities in the region, Indicates that two power facilities are dimensional distance, x, y respectively represent the location coordinates of different power facilities in each dimension.

5. According to the satellite internet-based transmission corridor wildfire spread prediction and risk assessment method of claim 1, it is characterized in that: The steps of comprehensively considering the fire spread path, fire intensity and equipment distribution characteristics, assessing the damage risk of wildfires to transmission lines, and generating a damage risk classification map for transmission corridors include: Synthesize fire spread paths to identify fire impact areas and critical equipment; The affected area of ​​wildfires is divided into three types: high-temperature combustion area, charge accumulation area, and smoke particle area. The historical data of line tripping caused by wildfires is analyzed, and the wildfire spread conditions and line breakdown probability are considered to obtain the transmission line tripping probability under wildfire conditions: ； In the formula, and are the space gap voltage and space gap breakdown voltage under wildfire conditions, It is the sum of the spatial gap breakdown voltages of the three regions: high temperature combustion region, charge accumulation region, and smoke particle region. is the phase voltage of the transmission line operation; The impact of line tripping probability and fault propagation on power grid operation is considered, and the problems caused by fire are simulated based on power grid flow analysis, and a damage assessment model is constructed; Among them, the construction method of the comprehensive probability model of transmission line wildfire tripping is: ； Where: is the maximum spreading speed of the fire point along the line connecting the alarm tower and the fire point, This is the wildfire data update cycle. is the distance between the fire point and the alarm tower; The damage assessment model is constructed as follows: ； Where: is the risk value of wildfire damage to transmission lines, For the The probability of a line tripping due to a wildfire, For the The power grid damage value caused by tripping of the line, are load loss and voltage fluctuation respectively. are weight coefficients. Adjusting these two parameters can control the damage model according to different actual conditions. Generate a transmission corridor damage risk classification map based on the output results of the damage assessment model.

6. A transmission corridor wildfire spread prediction and risk assessment system based on satellite Internet, characterized in that: Executing the method according to claim 1, the satellite internet-based power transmission corridor wildfire spread prediction and risk assessment system comprises: Data construction module, used to construct a fire risk factor database based on multimodal data of satellite-ground fusion network mode; A contribution module, for applying a random forest algorithm to screen out key factors in the fire risk factor database and determine the fire risk contribution corresponding to the key factors; The sub-region module is used to extract similar features of the transmission corridor neighborhood based on the power wildfire hazard factor, and use a clustering algorithm to divide the study area into several sub-regions with similar power risk characteristics; A regional division module is used to construct a fire risk weighted scoring model, dynamically adjust the regional division of the sub-regions based on the fire risk contribution, and identify high, medium and low risk areas; The path prediction module is used to build a wildfire spread model based on meteorological observation data, establish a fire risk state transfer matrix, and combine the dynamic Bayesian network to predict the fire spread path in real time; The classification map module is used to comprehensively consider the fire spread path, fire intensity and equipment distribution characteristics, assess the damage risk of wildfires to transmission lines, and generate a damage risk classification map for the transmission corridor.

7. A computer device, characterized in that: The device comprises: a memory and a processor, wherein the processor executes the method according to any one of claims 1 to 5 when running computer instructions stored in the memory.

8. A computer-readable storage medium, characterized in that: The method comprises instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 5.

Images