Major forest fire emergency rescue channel path construction method based on mapping knowledge domain

Abstract

The invention discloses a major and extra-large forest fire emergency rescue channel path construction method based on a knowledge graph, and belongs to the field of forest fire rescue, and the method comprises the following steps: S1, multi-source heterogeneous data fusion and standardization processing; s2, performing forest fire dynamic risk assessment; s3, knowledge graph entity extraction; s4, constructing an emergency rescue channel path; and S5, calculating the feasibility score of the emergency rescue channel path. According to the major and extra forest fire emergency rescue channel path construction method based on the knowledge graph, multi-modal data semantic association and real-time risk coupling, unmanned aerial vehicle group collaborative investigation and path planning linkage, and fire spreading simulation and knowledge graph reasoning fusion are carried out; the breakthrough of the emergency rescue channel path from static planning to dynamic adaptation is realized, and the major fire rescue efficiency of the special forest is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of forest fire rescue technology, and in particular to a method for constructing emergency rescue channels and routes for major and catastrophic forest fires based on knowledge graphs. Background Technology

[0002] Forest fires are a major natural disaster threatening ecological security and human life and property, characterized by their suddenness, destructiveness, and rapid spread. In recent years, affected by global climate change, intensified drought, and human activities, major and catastrophic forest fires have occurred frequently, posing a severe challenge to traditional emergency rescue route construction technologies.

[0003] Currently, the construction of rescue channels for major forest fires mainly relies on the following traditional methods:

[0004] 1. Manual planning based on experience and static maps: Frontline commanders manually select rescue routes based on historical experience and traditional GIS maps, combined with the general fire situation. This method relies on individual cognitive levels and lacks systematic integrated analysis of dynamic risk factors such as terrain, vegetation, and weather, making it difficult to adapt to complex scenarios where the fire situation changes rapidly.

[0005] 2. Local dynamic optimization methods: Some solutions attempt to combine UAV reconnaissance or fire spread models for finite path optimization, but they mostly remain at the level of single-dimensional data application (such as relying only on remote sensing images or single-point meteorological data), lacking the ability to deeply correlate multimodal data and semantic-level risk coupling.

[0006] 3. Isolated case library and rule engine: Some systems store historical emergency rescue channel path construction cases, but lack graph-based knowledge association and dynamic reasoning mechanisms. They cannot automatically extract risk evolution patterns or core constraints of similar scenarios, resulting in poor adaptability and low iteration efficiency in emergency rescue channel path planning. Summary of the Invention

[0007] The purpose of this invention is to provide a method for constructing emergency rescue routes for major forest fires based on knowledge graphs, thereby solving the aforementioned technical problems.

[0008] To achieve the above objectives, this invention provides a method for constructing emergency rescue routes for major forest fires based on knowledge graphs, comprising the following steps:

[0009] S1. Multi-source heterogeneous data fusion and standardization processing: Integrate remote sensing, meteorological, historical fire, forest fire resources and rescue text data, and obtain feature datasets through format unification and feature extraction;

[0010] S2. Dynamic risk assessment of forest fires: Real-time prediction of fire risk using a forest fire risk prediction model trained on the feature dataset obtained in step S1.

[0011] S3. Knowledge Graph Entity Extraction: Based on the feature dataset obtained in step S1 and the real-time fire risk prediction results obtained in step S2, knowledge graph entities are extracted from multi-source heterogeneous data.

[0012] S4. Emergency rescue channel path construction: Using the knowledge graph obtained in step S3 as a constraint, the UAV reconnaissance trajectory is determined through a dual-depth Q network, and the emergency rescue channel path is generated by combining the fire observation data on the UAV reconnaissance trajectory with the A* algorithm.

[0013] S5. Verify the emergency rescue route by calculating the route feasibility score, and realize the dynamic optimization of the emergency rescue route based on the score results and the knowledge graph described in step S3 of real-time fire update.

[0014] Therefore, the present invention employs the above-mentioned knowledge graph-based method for constructing emergency rescue routes for major forest fires, which has the following beneficial effects:

[0015] 1. By unifying semantic modeling through knowledge graphs, we can accurately associate risk entities across the entire domain with multimodal features, thereby improving the accuracy of risk quantification;

[0016] 2. Integrate high-frequency dynamic data (UAV reconnaissance + weather updates) with map inference to achieve real-time dynamic path optimization (minute-level response to sudden fire changes);

[0017] 3. Construct a multi-dimensional composite evaluation system (fire control failure rate, traffic capacity, infrared detection accuracy, etc.) to ensure path safety redundancy and adaptability to extreme scenarios.

[0018] In summary, this invention constructs a dynamic knowledge graph by integrating multi-source remote sensing, meteorological, and textual data to achieve entity semantic association and rule-based reasoning; it optimizes the reconnaissance trajectory of UAV swarms by combining deep reinforcement learning (DRL) with fire spread simulation (FARSITE) and real-time risk coupling analysis; and it innovatively establishes a complete closed-loop mechanism covering "data fusion - risk modeling - path generation - dynamic verification - knowledge archiving," solving the problems of traditional methods such as fragmentation of multi-source heterogeneous data, lagging dynamic adaptability, insufficient security redundancy, inefficient knowledge reuse, and bottlenecks in collaborative communication. This achieves a fundamental breakthrough in emergency rescue channel paths from "passive experience response" to "proactive intelligent prediction," significantly improving the efficiency and safety of rescue operations in major and catastrophic fires.

[0019] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method for constructing emergency rescue routes for major forest fires based on knowledge graphs, as described in this invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0022] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, system, product, or server that includes a series of steps or units, not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0023] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] like Figure 1 As shown, the method for constructing emergency rescue routes for major forest fires based on knowledge graphs includes the following steps:

[0025] S1. Multi-source heterogeneous data fusion and standardization processing: Integrate remote sensing, meteorological, historical fire, forest fire resources and rescue text data, and obtain feature datasets through format unification and feature extraction;

[0026] Step S1 specifically includes the following steps:

[0027] S11. Multi-source heterogeneous data acquisition: Acquire remote sensing, meteorological, historical fire, forest fire resources and rescue text data of the target area. s The remote sensing data includes vegetation distribution data, fire distribution data, and GIS data on roads and river systems; the meteorological data includes monthly precipitation (m³). P Potential evapotranspiration (PEI), daily average relative temperature (Hu) and daily average relative humidity (Te), wind speed and wind direction;

[0028] S12. Data format unification and resampling: The collected multi-source heterogeneous data is converted to UTM coordinates and resampled to a uniform resolution after standardization to obtain a standardized dataset.

[0029] S13. Feature Extraction: Extracting vegetation features V from the standardized dataset. C Meteorological characteristics NC Forest fire characteristics I C Topographic features T f Textual remote sensing correlation features (TRS) and forest fire resource features (F) r After min-max standardization, the feature dataset F = {V} is obtained. C N C ,I C ,T f ,TRS,F r}

[0030] In step S13, vegetation feature V C Including Normalized Difference Vegetation Index (NDVI), Gross Vegetation Moisture Index (GVMI), and Combustion Load Classification (L). c and vegetation cover (Ve);

[0031] The expression for the Normalized Difference Vegetation Index (NDVI) is as follows:

[0032]

[0033] In the formula, NIR represents the near-infrared reflectance of the remote sensing satellite; R R This indicates the reflectivity of the red band of a remote sensing satellite.

[0034] The expression for the vegetation humidity index is as follows:

[0035]

[0036] In the formula, SWIR represents the shortwave infrared reflectance of the remote sensing satellite;

[0037] Combustible material load classification L c The expression is as follows:

[0038]

[0039] In the formula, l b m b and h b These represent low flammability, medium flammability, and high flammability, respectively; F l This indicates the combustible material load, measured in t / hm². 2 ;

[0040] The expression for vegetation cover (Ve) is as follows:

[0041]

[0042] In the formula, NDVI min and NDVI max These represent the minimum and maximum values ​​of the normalized vegetation index within the target area, respectively.

[0043] Meteorological characteristics N C Including Standardized Precipitation Evapotranspiration Index (SPEI), Humidity-Temperature Index (HTI), and Wind Speed ​​(W). S and wind direction θ d The standardized precipitation evapotranspiration index is expressed as follows:

[0044]

[0045] In the formula, σ P-PET Historical monthly precipitation (m) P and the standard deviation of potential evapotranspiration (PEI);

[0046] The Humidity-Temperature Index (HTI) expression is as follows:

[0047] HTI=0.6×(1-Te / 100)+0.4×(Hu / 35);

[0048] Forest fire characteristics I C Including forest fire density D t Fire radiation power FPR and the shortest straight-line distance d between the real-time fire point and the road entity. Fire-Road Forest fire density D t The expression is as follows:

[0049]

[0050] In the formula, ΔN i S represents the number of new fire points added in the target area within the time period Δt; target Indicates the area of ​​the target region;

[0051] Terrain features T f Including slope, and the expression for slope Sl is as follows:

[0052]

[0053] In the formula, and These represent the rates of elevation change in the x and y directions, respectively, in degrees (°).

[0054] The shortest straight-line distance d between the real-time fire point and the road entity Fire-Road The expression is as follows:

[0055]

[0056] In the formula, (x Fire ,y Fire (x) represents the UTM coordinates of the fire point; Road ,y Road () represents the UTM coordinates of a road node;

[0057] Characteristics of forest fire fighting resources F r Including response radius r i and maximum water supply distance D max-water ;

[0058] The steps for obtaining the text remote sensing association feature TRS are as follows:

[0059] The first step is to use the BERT-base model to analyze the text data T on forest fire prevention resources and rescue. s Perform text semantic encoding to obtain text vectors:

[0060] Vec(T s ) = BERT(s tokenized );

[0061] In the formula, Vec(T) s ) represents the semantic vector of forest fire fighting resources and rescue text data; BERT(s tokenized This indicates the use of the BERT-based model for semantic encoding.

[0062] Step 2: Remote sensing image feature extraction: A CNN network is used to extract infrared image features (R) from the remote sensing images. i Output image features:

[0063] Feat(r i ) = CNN(i n );

[0064] In the formula, Feat(R) i ) represents infrared image feature R i Feature vectors; CNN(i n ) represents the pixel value i normalized to [0,1]. n Feature extraction is performed using a CNN network;

[0065] Step 3: Extracting Text Remote Sensing Relationship Features (TRS):

[0066]

[0067] In the formula, Sim(T) s ,R i ) represents the cosine similarity between the text vector and the image feature; Sim(·) represents the cosine similarity function.

[0068] S2. Dynamic risk assessment of forest fires: Real-time prediction of fire risk using a forest fire risk prediction model trained on the feature dataset obtained in step S1.

[0069] Step S2 specifically includes the following steps:

[0070] S21. Construct a forest fire risk prediction model based on the XGBoost model:

[0071] S22. Determine the optimal initialization hyperparameters of the forest fire risk prediction model using the TPE algorithm within the Optuna framework;

[0072] S23. Divide the feature dataset F into a training set and a validation set. First, input the training set into the forest fire risk prediction model with the optimal initial hyperparameters for training until the area under the validation curve (AUC) of the validation set reaches the set value. Then, determine that the forest fire risk prediction model has converged and stop training.

[0073] S24. Collect real-time multi-source heterogeneous data of the target area and input it into the trained forest fire risk prediction model. Output the real-time predicted fire risk value (risk) and classify the fire risk value (risk) into a fire risk level (F). l ;

[0074] S3. Knowledge Graph Entity Extraction: Based on the feature dataset obtained in step S1 and the real-time fire risk prediction results obtained in step S2, knowledge graph entities are extracted from multi-source heterogeneous data.

[0075] Step S3 specifically includes the following steps:

[0076] S31. Extract knowledge graph entities: Divide the target area into grids and extract real-time fire risk prediction results and vegetation features V. C Meteorological characteristics N C Forest fire characteristics I C Topographic features T f And the text remote sensing correlation feature (TRS) is associated with the corresponding grid.

[0077] S32. Entity extraction result verification: Using ArcGIS's spatial connectivity tool, verify the consistency of the spatial range and spatial consistency of the entity knowledge graph to obtain the entity set of the knowledge graph;

[0078] S33. Triple Relationship Extraction: Based on the knowledge graph entity set, semantic relationships between entities are extracted through spatial analysis and business rules to form subject-relationship-object triples, resulting in a knowledge graph triple set. The knowledge graph triple set includes spatial location relationships and interaction relationships between entities, as well as supporting or restricting relationships between entities on the construction of emergency rescue channel paths.

[0079] S34. Based on the knowledge graph triple set, construct and store a structured knowledge graph to form a semantic network that supports path construction and reasoning.

[0080] S4. Emergency rescue channel path construction: Using the knowledge graph obtained in step S3 as a constraint, the UAV reconnaissance trajectory is determined through a dual-depth Q network, and the emergency rescue channel path is generated by combining the fire observation data on the UAV reconnaissance trajectory with the A* algorithm.

[0081] Step S4 specifically includes the following steps:

[0082] S41. Assign the toll cost attribute to the corresponding grid in the target area:

[0083] C(i,j)=road cost +0.3ω1×risk+ω2×Sl+ω3×ρ v ;

[0084] In the formula, C(i,j) represents the comprehensive travel cost from grid i to grid j; road cost ω1 represents the cost of opening the channel; ω1, ω2, and ω3 all represent weighting coefficients, and ω1 + ω2 + ω3 = 1; risk represents the fire risk value of the current grid; ρ v ρ represents the vegetation density of the current grid. v =Ve×k veg k veg Indicates the vegetation type coefficient;

[0085] S42. Determining the trajectory of a drone swarm based on a dual-depth Q-network;

[0086] S421. Define the state space of a dual-depth Q-network as s = (x UVA ,y UVA ,risk,d Fire-Road ,comm); where, (x UVA ,y UVA ) represents the coordinates of the drone; omm represents the communication connection status; motion space s = (East, South, West, North, Hover), where East, South, West, North, and Hover represent east, south, west, north, and hover, respectively;

[0087] Integrating the FRZI risk index (FRZI) with the path cost reward function r:

[0088] r = ω α ·FRZI-ω β ·d Fire-Road -ω γ ·collision-ω δ ·comm loss -ω ∈ ·P block ;

[0089] in,

[0090]

[0091] P block =ω W ×wind factor +ω s ×slope factor +ω f ×fuel factor ;

[0092] In the formula, ω α ω β ω γ ω δ and ω ∈ Both represent weighting coefficients, and ω α +ω β +ω γ +ω δ +ω ∈ =1; collision represents the drone collision penalty; comm loss Indicates a penalty for communication interruption; P block ω represents the probability of a fire going out of control within the grid. W ω s and ω f Both represent weighting coefficients, ω W +ω s +ω f =1, wind factor This represents the wind speed factor between [0,1], and W s_min and W s_max These represent the minimum reference wind speed and the maximum reference wind speed, respectively; slope factor Represents the slope factor between [0,1]; fuel factor L represents the flammability factor in the range [0,1]. i,j The path length is represented by w0, w1, w2, and w3, which are all weight coefficients, and w0 + w1 + w2 + w3 = 1.

[0093] S422, Training the dual-depth Q-network defined in step S421 until the number of training steps reaches the set number, customized training, and outputting the dual-depth Q-network;

[0094] S423. Input the real-time state space s into a dual-depth Q-network to obtain the optimal reconnaissance trajectory of the UAV swarm and the corresponding fire observation data.

[0095] S43. Generate emergency rescue route: Based on the fire observation data corresponding to the trajectory obtained in step S423 and the road entities in the knowledge graph, combined with the FARSITE model to predict the risk of the fire spreading out of control across the road, the A* algorithm is used to generate the initial emergency rescue route:

[0096] Cost(n) = g(n) + h(n) + λ·P block (n);

[0097] in,

[0098]

[0099] h(n) = |x n -x goal |+|y n -y goal |;

[0100] In the formula, Cost(n) represents the cost function for establishing the initial emergency rescue route; g(n) represents the actual cost from the starting point to node n; k represents the number of grids traversed from the starting point to node n; and road cost (i) represents the cost of opening a channel within grid i; risk cost (i) represents the fire risk cost of grid i; h(n) represents the heuristic function; (x n ,y n (x) represents the UTM coordinates of node n; goal ,y goal ) represents the endpoint coordinates; λ represents the weighting coefficient for the risk of fire spiraling out of control; P block (n) represents the probability of fire getting out of control at node n;

[0101] S44. By combining the road entity connectivity in the knowledge graph, interrupted road sections are eliminated, and the points of passage are fitted with Bézier curves to achieve smooth path processing, thereby obtaining the final emergency rescue channel path.

[0102] S5. Verify the emergency rescue route by calculating the route feasibility score, and dynamically optimize the emergency rescue route based on the score results and the knowledge graph described in step S3 of real-time fire update.

[0103] Step S5 specifically includes the following steps:

[0104] S51. Route Feasibility Verification:

[0105] F S =F ars +R us +A s +R tr +D vrs ;

[0106] in,

[0107]

[0108]

[0109] In the formula, F S Indicates the total score for path feasibility; F ars R us A s R tr and D vrs These represent the scores for fire risk avoidance rate, passage utilization rate, traffic capacity adaptability, real-time risk matching, and drone verification rate, respectively; d min d represents the minimum distance from the constructed path to the boundary of the fire zone; min,max and d min,min L represents the maximum safe distance and minimum reasonable distance to the target area, respectively; road and L total L represents the length of the native soil zone (free of combustibles) in the constructed path and the total length of the constructed path; qual This indicates the length of roads in the constructed path that allow forest fire fighting resources to pass through; This represents the average fire risk value along the constructed path; r target Indicates the target fire risk tolerance value; r min and r max L represents the global minimum and maximum values ​​of the fire risk value for the target area, respectively; verify This indicates the verified path length for drone aerial photography;

[0110] S52, Determine the total score for path feasibility (F) S If the value is not less than the threshold set by experts, output the final emergency rescue route; otherwise, output the route based on the total feasibility score F. S Based on real-time changes in the fire situation, return to step S3 and dynamically update and optimize the final emergency rescue route using the knowledge graph.

[0111] Simulation Experiment

[0112] Experimental scenario setting

[0113] Target area: Select a typical alpine forest area (approximately 100 km²). 2 It includes complex terrain (slope 0° to 45°), diverse vegetation (60% coniferous forest, 30% broad-leaved forest, and 10% shrub forest) and a hierarchical road network (2 main roads and 15 rural roads).

[0114] Fire Scenario: A crown fire scenario simulating a major fire. The initial fire point is located in the center of the area, and the fire spreads dynamically with wind speed (3-12 m / s) and slope. The fire situation is updated every 10 minutes.

[0115] The emergency rescue routes for the above experimental scenario were constructed using traditional method 1 (manual construction of emergency rescue routes based on static GIS (relying on historical experience and not considering dynamic fire conditions), traditional method 2 (local dynamic construction of emergency rescue routes based on single UAV reconnaissance (without knowledge graph semantic association)), and the method described in this invention, respectively.

[0116] Table 1 Comparison of Core Indicators

[0117]

[0118] As shown in Table 1, the method described in this invention responds to sudden changes in fire intensity 47% faster than the traditional method 2, and compared to the traditional method 1, it can reduce the risk of fire getting out of control by 72%, shorten the passage length by 30% (compared to the traditional method 1) while ensuring safety, and increase the coverage of high-risk areas by 31.4%, thus balancing rescue efficiency and risk control, thereby proving the effectiveness of this invention.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for constructing emergency rescue routes for major forest fires based on knowledge graphs, characterized by: Includes the following steps: S1. Multi-source heterogeneous data fusion and standardization processing: Integrate remote sensing, meteorological, historical fire, forest fire resources and rescue text data, and obtain feature datasets through format unification and feature extraction; S2. Dynamic risk assessment of forest fires: Real-time prediction of fire risk using a forest fire risk prediction model trained on the feature dataset obtained in step S1. S3. Knowledge Graph Entity Extraction: Based on the feature dataset obtained in step S1 and the real-time fire risk prediction results obtained in step S2, knowledge graph entities are extracted from multi-source heterogeneous data. S4. Emergency rescue channel path construction: Using the knowledge graph obtained in step S3 as a constraint, the UAV reconnaissance trajectory is determined through a dual-depth Q network, and the emergency rescue channel path is generated by combining the fire observation data on the UAV reconnaissance trajectory with the A* algorithm. S5. Verify the emergency rescue route by calculating the route feasibility score, and realize the dynamic optimization of the emergency rescue route based on the score results and the knowledge graph described in step S3 of real-time fire update.

2. The method for constructing emergency rescue routes for major forest fires based on knowledge graphs according to claim 1, characterized in that: Step S1 specifically includes the following steps: S11. Multi-source heterogeneous data acquisition: Acquire remote sensing, meteorological, historical fire, forest fire resources and rescue text data of the target area. s The remote sensing data includes vegetation distribution data, fire distribution data, and GIS data of roads and river systems; the meteorological data includes monthly precipitation (m³). P Potential evapotranspiration (PEI), daily average relative temperature (Hu), daily average relative humidity (Te), wind speed, and wind direction; S12. Data format unification and resampling: The collected multi-source heterogeneous data is converted to UTM coordinates and resampled to a uniform resolution after standardization to obtain a standardized dataset. S13. Feature Extraction: Extracting vegetation features V from the standardized dataset. C Meteorological characteristics N C Forest fire characteristics I C Topographic features T f Textual remote sensing correlation features (TRS) and forest fire resource features (F) r After min-max standardization, the feature dataset F = {V} is obtained. C N C ,I C ,T f ,TRS,F r } 3. The method for constructing emergency rescue routes for major forest fires based on knowledge graphs according to claim 2, characterized in that: In step S13, vegetation feature V C Including Normalized Difference Vegetation Index (NDVI), Gross Vegetation Moisture Index (GVMI), and Combustion Load Classification (L). c and vegetation cover (Ve); The expression for the Normalized Difference Vegetation Index (NDVI) is as follows: In the formula, NIR represents the near-infrared reflectance of the remote sensing satellite; R R This indicates the reflectivity of the red band of a remote sensing satellite. The expression for the vegetation humidity index is as follows: In the formula, SWIR represents the shortwave infrared reflectance of the remote sensing satellite; Combustible material load classification L c The expression is as follows: In the formula, l b m b and h b These represent low flammability, medium flammability, and high flammability, respectively; F l This indicates the combustible material load, measured in t / hm². 2 ; The expression for vegetation cover (Ve) is as follows: In the formula, NDVI min and NDVI max These represent the minimum and maximum values ​​of the normalized vegetation index within the target area, respectively. Meteorological characteristics N C Including Standardized Precipitation Evapotranspiration Index (SPEI), Humidity-Temperature Index (HTI), and Wind Speed ​​(W). s and wind direction θ d The standardized precipitation evapotranspiration index is expressed as follows: In the formula, σ P-PET Historical monthly precipitation (m) P and the standard deviation of potential evapotranspiration (PEI); The Humidity-Temperature Index (HTI) expression is as follows: HTI=0.6×(1-Te / 100)+0.4×(Hu / 35); Forest fire characteristics I C Including forest fire density D t Fire radiation power FPR and the shortest straight-line distance d between the real-time fire point and the road entity. Fire-Road Forest fire density D t The expression is as follows: In the formula, ΔN i S represents the number of new fire points added in the target area within the time period Δt; target Indicates the area of ​​the target region; Terrain features T f Including slope, and the expression for slope Sl is as follows: In the formula, and These represent the rates of elevation change in the x and y directions, respectively, in degrees (°). The shortest straight-line distance d between the real-time fire point and the road entity Fire-Roda The expression is as follows: In the formula, (x Fire ,y Fire (x) represents the UTM coordinates of the fire point; Road ,y Road () represents the UTM coordinates of a road node; Characteristics of forest fire fighting resources F r Including response radius r i and maximum water supply distance D max-water ; The steps for obtaining the text remote sensing association feature TRS are as follows: The first step is to use the BERT-base model to analyze the text data T on forest fire prevention resources and rescue. s Perform text semantic encoding to obtain text vectors: Vec(T s )=BERT(s tokenized ); In the formula, Vec(T) s ) represents the semantic vector of forest fire fighting resources and rescue text data; BERT(s tokenized This indicates the use of the BERT-based model for semantic encoding. Step 2: Remote sensing image feature extraction: A CNN network is used to extract infrared image features (R) from the remote sensing images. i Output image features: Feat(R i )=CNN(i n ); In the formula, Feat(R) i ) represents infrared image feature R i Feature vectors; CNN(i n ) represents the pixel value i normalized to [0,1]. n Feature extraction is performed using a CNN network; Step 3: Extract Text Remote Sensing Relationship Features (TRS) In the formula, Sim(T) s ,R i ) represents the cosine similarity between the text vector and the image feature; Sim(·) represents the cosine similarity function.

4. The method for constructing emergency rescue routes for major forest fires based on knowledge graphs according to claim 3, characterized in that: Step S2 specifically includes the following steps: S21. Construct a forest fire risk prediction model based on the XGBoost model: S22. Determine the optimal initialization hyperparameters of the forest fire risk prediction model using the TPE algorithm within the Optuna framework; S23. Divide the feature dataset F into a training set and a validation set. First, input the training set into the forest fire risk prediction model with the optimal initial hyperparameters for training until the area under the validation curve (AUC) of the validation set reaches the set value. Then, determine that the forest fire risk prediction model has converged and stop training. S24. Collect real-time multi-source heterogeneous data of the target area and input it into the trained forest fire risk prediction model. Output the real-time predicted fire risk value (risk) and classify the fire risk value (risk) into a fire risk level (F). l。 5. The method for constructing emergency rescue routes for major forest fires based on knowledge graphs according to claim 4, characterized in that: Step S3 specifically includes the following steps: S31. Extract knowledge graph entities: Divide the target area into grids and extract real-time fire risk prediction results and vegetation features V. C Meteorological characteristics N C Forest fire characteristics I C Topographic features T f And the text remote sensing correlation feature (TRS) is associated with the corresponding grid. S32. Entity extraction result verification: Using ArcGIS's spatial connectivity tool, verify the consistency of the spatial range and spatial consistency of the entity knowledge graph to obtain the entity set of the knowledge graph; S33. Triple Relationship Extraction: Based on the knowledge graph entity set, semantic relationships between entities are extracted through spatial analysis and business rules to form subject-relationship-object triples, resulting in a knowledge graph triple set. The knowledge graph triple set includes spatial location relationships and interaction relationships between entities, as well as supporting or restricting relationships between entities on the construction of emergency rescue channel paths. S34. Based on the knowledge graph triple set, construct and store a structured knowledge graph to form a semantic network that supports reasoning for constructing emergency rescue channel paths.

6. The method for constructing emergency rescue routes for major forest fires based on knowledge graphs according to claim 5, characterized in that: Step S4 specifically includes the following steps: S41. Assign the toll cost attribute to the corresponding grid in the target area: C(i,j)=road cost +0.3ω1×risk+ω2×Sl+ω3×ρ v ; In the formula, C(i,j) represents the comprehensive travel cost from grid i to grid j; road cost ω1 represents the cost of opening the channel; ω1, ω2, and ω3 all represent weighting coefficients, and ω1 + ω2 + ω3 = 1; risk represents the fire risk value of the current grid; ρ v ρ represents the vegetation density of the current grid. v =Ve×k veg k veg Indicates the vegetation type coefficient; S42. Determining the trajectory of a drone swarm based on a dual-depth Q-network; S421. Define the state space of a dual-depth Q-network as s = (x UVA ,y UVA ,risk,d Fire-Road ,comm); where, (x UVA ,y UVA ) represents the coordinates of the drone; omm represents the communication connection status; motion space s = (East, South, West, North, Hover), where East, South, West, North, and Hover represent east, south, west, north, and hover, respectively; Integrating the risk index FRZI and the path cost reward function r: r=ω α ·FRZI-ω β ·d Fire-Road -oh γ ·collision-ω δ ·comm loss -oh ∈ ·P block ; in, P block =ω W ×wind factor +oh s ×slope factor +oh f ×fuel factor ; In the formula, ω α ω β ω γ ω δ and ω ∈ Both represent weighting coefficients, and ω α +ω β +ω γ +ω δ +ω ∈ =1; collision represents the drone collision penalty; comm loss Indicates a penalty for communication interruption; P block ω represents the probability of a fire going out of control within the grid. W ω s and ω f Both represent weighting coefficients, ω W +ω s +ω f =1, wind factor This represents the wind speed factor between [0,1], and W s_min and W s_max These represent the minimum reference wind speed and the maximum reference wind speed, respectively; slope factor Represents the slope factor between [0,1]; fuel factor L represents the flammability factor in the range [0,1]. i,j The path length is represented by w0, w1, w2, and w3, which are all weight coefficients, and w0 + w1 + w2 + w3 = 1. S422, Training the dual-depth Q-network defined in step S421 until the number of training steps reaches the set number, customized training, and outputting the dual-depth Q-network; S423. Input the real-time state space s into a dual-depth Q-network to obtain the optimal reconnaissance trajectory of the UAV swarm and the corresponding fire observation data. S43. Generate emergency rescue route: Based on the fire observation data corresponding to the optimal reconnaissance trajectory obtained in step S423 and the road entities in the knowledge graph, combined with the FARSITE model to predict the risk of fire spreading out of control across the road, the A* algorithm is used to generate the initial emergency rescue route: Cost(n)=g(n)+h(n)+λ·P block (n); in, h(n)=|x n -x goal |+|y n -y goal |; In the formula, Cost(n) represents the cost function for constructing the initial emergency rescue route; g(n) represents the actual cost from the starting point to node n; k represents the number of grids traversed from the starting point to node n; and road cost (i) represents the road infrastructure cost within grid i; risk cost (i) represents the fire risk cost of grid i; h(n) represents the heuristic function; (x n ,y n (x) represents the UTM coordinates of node n; goal ,y goal ) represents the endpoint coordinates; λ represents the weighting coefficient for the risk of fire spiraling out of control; P block (n) represents the probability of fire getting out of control at node n; S44. By combining the road entity connectivity in the knowledge graph, interrupted road sections are eliminated, and the points of passage are fitted with Bézier curves to achieve smooth path processing, thereby obtaining the final emergency rescue channel path.

7. The method for constructing emergency rescue routes for major forest fires based on knowledge graphs according to claim 6, characterized in that: Step S5 specifically includes the following steps: S51. Route Feasibility Verification: F S =F ars +R us +A s +R tr +D vrs ; in, In the formula, F S Indicates the total score for path feasibility; F ars R us A s R tr and D vrs These represent the scores for fire risk avoidance rate, road utilization rate, traffic capacity adaptability, real-time risk matching, and drone verification rate, respectively; d min d represents the minimum distance from the constructed path to the boundary of the fire zone; min,max and d min,min L represents the maximum safe distance and minimum reasonable distance to the target area, respectively; road and L total L represents the length of the native soil zone within the constructed path and the total length of the constructed path. qual This indicates the length of the constructed path that satisfies the requirement for forest fire fighting resources to pass through; This represents the average fire risk value along the constructed path; r target Indicates the target fire risk tolerance value; r min and r max L represents the global minimum and maximum values ​​of the fire risk value for the target area, respectively; verify This indicates the verified path length for drone aerial photography; S52, Determine the total score for path feasibility (F) S If the value is not less than the threshold set by experts, output the final emergency rescue route; otherwise, output the route based on the total feasibility score F. S Based on real-time changes in the fire situation, return to step S3 and dynamically update and optimize the final emergency rescue route using the knowledge graph.

Images