AI-based fire fighting system optimization method, device, equipment and storage medium

Abstract

The application provides an AI-based fire-fighting system optimization method, device, equipment and storage medium, which comprises the following steps: acquiring multi-source street data covering buildings, roads and population, and carrying out fire risk assessment and building value assessment in parallel according to the multi-source street data, generating a spatialized fire risk grid map and a building value grid map, and providing a quantitative and visual double basis for core decision-making. On this basis, the method generates two sets of fire passage planning schemes with different focuses respectively for the purpose of reducing the rescue difficulty in high fire risk areas and protecting high value buildings, thereby covering the two core demands of safety and protection. Most importantly, by performing intersection operation on the two sets of schemes and removing the unacceptable high value buildings, the final output target fire optimization scheme effectively integrates the consensus reform measures under different value orientations, while automatically avoiding decisions that may cause significant losses to precious historical relics.

Description

Technical Field

[0001] This invention relates to the field of fire protection planning technology, and in particular to an AI-based method, apparatus, equipment, and storage medium for optimizing fire protection systems. Background Technology

[0002] In the field of fire safety planning, especially in historical and cultural districts or similar high-density built-up areas with dense buildings, narrow roads, and numerous historical buildings, traditional fire protection system planning methods face significant challenges. Existing technologies often rely on single-dimensional analysis; for example, risk assessments may be based solely on historical fire data and simple density indicators to guide fire station layout, or traffic capacity analysis may be primarily based on current road conditions. These methods often fail to systematically reconcile the core contradiction between "improving fire safety" and "preserving historical value." On the one hand, planning solely aimed at reducing risk may recommend unnecessary demolition of high-value historical buildings or damage to the neighborhood fabric; on the other hand, overly conservative planning may fail to substantially improve fire rescue efficiency. Furthermore, the existing scheme generation process often lacks transparent, quantitative multi-scheme comparison and comprehensive decision-making mechanisms, resulting in the final scheme potentially failing to achieve an optimal balance between safety, economy, and cultural preservation.

[0003] In summary, the problems existing in the current technology urgently need to be solved. Summary of the Invention

[0004] This invention provides an AI-based method, apparatus, equipment, and storage medium for optimizing fire protection systems, thereby addressing the shortcomings of existing technologies, improving the overall fire safety level of neighborhoods, and preserving their historical and cultural value to the greatest extent possible.

[0005] This invention provides an AI-based method for optimizing fire protection systems, comprising: Acquire multi-source street data for the street to be evaluated, including building base map data, road base map data, and population structure data; Fire risk assessment and building value assessment are performed based on the multi-source street data, generating fire risk raster maps and building value raster maps. Based on the fire risk grid map, a first fire lane planning scheme is generated, and based on the building value grid map, a second fire lane planning scheme is generated. The first and second fire lane planning schemes are evaluated by taking their intersection and removing high-value buildings, and the target fire protection optimization scheme is output.

[0006] According to the AI-based fire protection system optimization method provided by the present invention, after the step of evaluating the intersection of the first fire lane planning scheme and the second fire lane planning scheme and eliminating high-value buildings to output the target fire protection optimization scheme, the method further includes: Based on the updated road and building data after implementing the target fire protection optimization scheme, and combined with the population structure data, the evacuation cost grid is calculated to determine the new evacuation locations; Based on the evacuation cost grid, the shortest path analysis method is used to generate evacuation paths from the ends of each street to the newly added evacuation sites and the existing evacuation sites, forming an evacuation system scheme that works in conjunction with the target fire protection optimization scheme.

[0007] According to the AI-based fire protection system optimization method provided by the present invention, the step of calculating the evacuation cost grid based on the updated road and building data after executing the target fire protection optimization scheme and in combination with the population structure data specifically includes: Based on the updated road and building data, determine the initial traffic speed for each road; Based on the population structure data, calculate the age structure influence coefficient and population density influence coefficient of the service population around each road; The evacuation cost grid is calculated by substituting the initial travel speed, the age structure influence coefficient, and the population density influence coefficient into the travel time model.

[0008] According to the AI-based fire protection system optimization method provided by the present invention, the step of performing fire risk assessment and building value assessment based on the multi-source street data to generate a fire risk raster map and a building value raster map specifically includes: Fire risk assessment based on the aforementioned multi-source neighborhood data includes: Based on the aforementioned building base map data, road base map data, and population structure data, a multi-level fire risk assessment index system is constructed. The weights of each level of the fire risk assessment index system were determined using the analytic hierarchy process (AHP). The evaluation index values ​​in the multi-source street data are converted into spatial raster data, and weighted and superimposed according to the corresponding weights to generate the fire risk raster map. Building valuation based on the aforementioned multi-source neighborhood data includes: Based on the aforementioned building base map data, and according to the building's cultural relic protection level and physical quality information, a building value grid representing the building's comprehensive value is calculated.

[0009] According to the AI-based fire protection system optimization method provided by the present invention, the step of generating a first fire lane planning scheme based on the fire risk grid map and generating a second fire lane planning scheme based on the building value grid map specifically includes: Based on the building value grid and the spatial proximity relationship between the building and the existing road, a building demolition cost grid is generated through cost distance analysis. Based on the road base map data, the initial toll cost grid of the existing road network and the target toll cost grid of the planned road network are calculated respectively, and the difference between the two is calculated as the road improvement potential grid. A fire road feasibility grid is generated by performing a composite calculation based on the building demolition cost grid and the road improvement potential grid, which represents the comprehensive feasibility of setting a specific road section as a fire road. Spatial overlay analysis is performed on the fire risk grid map and the fire road feasibility grid to generate a first fire lane planning scheme with the goal of prioritizing the improvement of road connectivity in high fire risk areas. By performing spatial overlay analysis on the building value raster map and the fire road feasibility raster, a second fire access planning scheme with the goal of prioritizing the protection of high-value buildings and improving the accessibility of roads around high-value buildings is generated.

[0010] According to the AI-based fire protection system optimization method provided by the present invention, the existing road network traffic cost grid is generated based on vector data reflecting the current road width and spatial morphology obtained from open-source geographic information databases or field surveys. The planned road network traffic cost grid is generated based on ideal road network vector data that meets the traffic standards for fire trucks, pre-defined according to fire protection design specifications and the existing spatial layout.

[0011] According to the AI-based fire protection system optimization method provided by the present invention, the step of evaluating the intersection of the first fire lane planning scheme and the second fire lane planning scheme and eliminating high-value buildings to output the target fire protection optimization scheme specifically includes: Extract the first set of buildings to be demolished and the first set of roads to be renovated from the first fire lane planning scheme; Extract the second set of buildings to be demolished and the second set of roads to be renovated from the second fire lane planning scheme; Perform a spatial intersection operation between the first set of buildings to be demolished and the second set of buildings to be demolished to obtain a preliminary intersection of the buildings to be demolished. In the initial set of buildings to be demolished, buildings with a value higher than a preset threshold are removed to generate the final set of buildings to be demolished. Perform a spatial intersection operation between the first road modification set and the second road modification set to generate the final road modification set; Based on the final set of buildings to be demolished and the final set of road modifications, a target fire protection optimization plan containing specific spatial locations and engineering measures is generated.

[0012] The present invention also provides an AI-based fire protection system optimization device, comprising: The data acquisition module is used to acquire multi-source street data of the street to be evaluated, including building base map data, road base map data and population structure data; The raster generation module is used to perform fire risk assessment and building value assessment based on the multi-source street data, and generate fire risk raster map and building value raster map. The scheme generation module is used to generate a first fire lane planning scheme based on the fire risk grid map, and to generate a second fire lane planning scheme based on the building value grid map. The scheme optimization module is used to evaluate the intersection of the first fire lane planning scheme and the second fire lane planning scheme and to eliminate high-value buildings, and output the target fire protection optimization scheme.

[0013] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the AI-based fire protection system optimization method described above.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the AI-based fire protection system optimization method as described above.

[0015] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the AI-based fire protection system optimization method as described above.

[0016] This invention provides an AI-based fire protection system optimization method, apparatus, equipment, and storage medium. First, it acquires multi-source street data covering buildings, roads, and population, and then conducts parallel fire risk and building value assessments to generate spatial fire risk and building value raster maps, providing quantitative and visual evidence for core decision-making. Based on this, the method generates two fire lane planning schemes with different focuses, aiming to reduce the difficulty of rescue in high-fire-risk areas and protect high-value buildings, thus covering the two core demands of safety and protection. Most importantly, by performing an intersection operation on the two schemes and eliminating unacceptable high-value buildings, the final target fire protection optimization scheme effectively integrates consensus-based renovation measures under different value orientations, while automatically avoiding decisions that could cause significant damage to precious historical heritage. This process achieves an automated, rule-based multi-objective trade-off and scheme fusion, resulting in a final scheme that significantly improves the overall fire safety level of the street while maximizing the preservation of its historical and cultural value. It provides a scientific, practical, and operable refined fire protection planning solution for complex scenarios such as high-density historical and cultural districts. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 This is one of the flowcharts illustrating the AI-based fire protection system optimization method provided by the present invention; Figure 2 This is the second flowchart of the AI-based fire protection system optimization method provided by the present invention; Figure 3 This invention provides an evacuation path generation diagram based on the AI-based fire protection system optimization method. Figure 4 This is a fire risk assessment diagram of the AI-based fire protection system optimization method provided by the present invention; Figure 5 This is a grid diagram of building demolition costs for the AI-based fire protection system optimization method provided by this invention. Figure 6 This invention provides a road improvement potential grid map for the AI-based fire protection system optimization method. Figure 7 This is a diagram of a high-risk protection priority rescue system scheme based on the AI-based fire protection system optimization method provided by the present invention; Figure 8 This is a diagram of a high-value protection priority rescue system scheme based on the AI-based fire protection system optimization method provided by the present invention; Figure 9 This is a diagram of the target fire protection optimization scheme of the AI-based fire protection system optimization method provided by the present invention; Figure 10 This is a schematic diagram of the structure of the AI-based fire protection system optimization device provided by the present invention; Figure 11 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0020] To address the problems in existing technologies, this invention proposes an AI-based fire protection system optimization method that improves the overall fire safety level of a neighborhood while preserving its historical and cultural value to the greatest extent possible. The following describes this AI-based fire protection system optimization method, such as... Figure 1 and Figure 2 As shown, including but not limited to the following steps: Step 110: Obtain multi-source street data for the street to be evaluated. The multi-source street data includes building base map data, road base map data, and population structure data.

[0021] The purpose of this step is to prepare standardized and unified basic data for subsequent analysis. During implementation, the system automatically or semi-automatically collects and integrates multi-source data from the target street area through various channels.

[0022] Specifically: Architectural base map data: Vector polygon data of all buildings within the block area are obtained from publicly available GIS databases or open-source map platforms (such as OpenStreetMap) of urban surveying and mapping departments. Each building data record contains necessary attribute information, such as: unique identifier, building outline geometry, number of floors, structural type (e.g., brick-wood, brick-concrete, frame), main function (e.g., residential, commercial, cultural relic), and key protection level information (e.g., "National Key Cultural Relics Protection Unit", "Historical Building", "Traditional Style Building", etc.). If the publicly available data lacks building quality information, it can be supplemented by connecting to existing building survey databases or setting up expert scoring interfaces.

[0023] Road base map data: Obtain street network vector line data from open-source GIS data or planning departments. Each road data point must include a width attribute (unit: meters), which should be the effective width for fire trucks to pass. For streets with irregular widths, the minimum effective width of the bottleneck section can be used. The data should also distinguish between different road classes (e.g., municipal roads, internal streets).

[0024] Population structure data: Obtained from street or community statistical yearbooks and census micro-data. The data must include at least the total population within a block or more granular statistical unit (such as a census tract), as well as the population distribution for each age group (e.g., 0-14 years, 15-64 years, 65 years and above). This data will be linked to spatial units for subsequent risk assessment and evacuation analysis.

[0025] Before entering the analysis process, all acquired raw data must undergo coordinate system reforming, data cleaning (such as handling missing values ​​and correcting obvious errors), and format standardization to ensure data consistency.

[0026] Step 120: Based on the multi-source street data, conduct fire risk assessment and building value assessment respectively, and generate fire risk raster map and building value raster map.

[0027] This step is the core analysis phase, designed to transform multi-source data into two key types of raster charts that inform decision-making.

[0028] The specific implementation is as follows: Fire risk assessment and fire risk grid generation: Constructing an indicator system: Based on domain knowledge, a three-layer fire risk assessment system is constructed, comprising an objective layer, a criterion layer, and an indicator layer. For example, the criterion layer may include "regional environmental hazard," "weaknesses in fire protection facilities," and "deficiencies in safety management." The indicator layer corresponds to specific quantifiable data, such as "building density," "average fire resistance rating of buildings," "road network density," "ratio of blind spots in fire station service areas," and "percentage of elderly population."

[0029] Determining indicator weights: The Analytic Hierarchy Process (AHP) was used to determine the relative importance weights of each criterion and indicator. Experts in fire safety and urban planning were invited to conduct pairwise comparisons and scoring of the indicators, constructing a judgment matrix. After calculation and consistency testing (CR < 0.1), a scientific weighting system was obtained.

[0030] Rasterization Calculation: In the GIS platform, the data corresponding to each indicator (such as building area density, road line density, and population distribution point density) is uniformly converted into a raster layer with the same resolution (e.g., 2m x 2m) and spatial range using tools such as kernel density analysis and reclassification. The value of each cell represents the score of that indicator at that location. Finally, using a raster calculator, a weighted summation is performed according to the formula Fire Risk Value = Σ(Indicator Raster i × Weight i) to generate a continuous fire risk raster map. The higher the cell value, the higher the overall fire risk at that location.

[0031] Building valuation and building value raster generation: Assessment Model Establishment: The assessment of architectural value primarily considers its historical and cultural value. A simplified quantitative model is established: Architectural Value Score = Cultural Relics Protection Level Score × Weight W1 + Architectural Condition Score × Weight W2. The Cultural Relics Protection Level Score is directly assigned based on the national, provincial, municipal, or district-level protection level (e.g., 100, 80, 60, 40); the Architectural Condition Score can be assessed by experts based on structural safety, architectural integrity, etc. (0-100 points). W1 and W2 are preset weights (e.g., W1=0.7, W2=0.3), emphasizing the dominant role of the cultural relics protection level.

[0032] Raster map generation: The calculated "building value score" for each building is assigned a corresponding building outline surface attribute. Then, using the GIS "surface to raster" tool, this attribute field is converted into a raster layer with the same resolution as the fire risk raster, thus obtaining the building value raster map. Areas without buildings can be set to 0 or empty values.

[0033] Step 130: Based on the fire risk grid map, generate a first fire lane planning scheme, and based on the building value grid map, generate a second fire lane planning scheme.

[0034] This step aims to generate alternative fire lane optimization schemes based on different priority objectives.

[0035] The specific implementation is as follows: Preliminary analysis – Generating a “Fire Road Feasibility Grid”: To quantify where it is more "cost-effective" to open or widen fire lanes, a key intermediate grid needs to be calculated first: the fire lane feasibility grid. Its calculation involves two sub-grids: Building Demolition Cost Raster: Based on the building value raster generated in step 120, the cost distance tool in GIS is used for calculation. The edge of the existing road is used as the "source," and the building value raster is used as the "cost" required to "pass through" the cell. After calculation, the closer the area is to the road and the lower the building value, the lower its "demolition cost" value, which means that the economic and cultural costs of clearing a path for fire trucks at that location are smaller.

[0036] Road Improvement Potential Grid: First, based on the requirements for fire lane width and turning radius in fire protection codes (such as the "Code for Fire Protection Design of Buildings" GB50016), an ideal "planned road network" is designed (e.g., virtually widening all streets and alleys requiring modification to 4 meters). Then, the time cost (traffic cost grid) required for vehicles to pass through each cell of the existing road network and each cell of the planned road network is calculated separately. The difference between the two cost grids is the "road improvement potential grid," with higher values ​​indicating greater traffic efficiency improvements after opening up or widening the area.

[0037] Finally, the fire road feasibility grid is calculated as: Road improvement potential grid / Building demolition cost grid. A higher grid value indicates a higher "cost-effectiveness" for road modifications at that location.

[0038] Two planning schemes are generated: First Fire Lane Planning Scheme (High-Risk Protection Priority): In GIS, the "Fire Risk Raster Map" and the "Fire Road Feasibility Raster" are overlaid and analyzed. The system sets a threshold and automatically identifies pixel clusters with both "high fire risk" and "high feasibility of modification." In these areas, based on the direction of the planned road network, suggested new or widened fire lane segments (vector lines) are automatically generated and associated with the outlines of low-value buildings to be demolished, forming the first scheme.

[0039] Second Fire Lane Planning Scheme (Prioritizing High-Value Building Protection): This scheme overlays and analyzes the "Building Value Grid Map" and the "Fire Lane Feasibility Grid Map." The system identifies pixel clusters with both "high building value" and "high renovation feasibility," i.e., areas around high-value buildings that are easy to renovate. In these areas, based on the planned road network, it automatically generates suggested new or widened fire lane segments to ensure that high-value buildings can be quickly accessed in the event of a fire. Simultaneously, it associates these with low-value buildings that need to be demolished, forming a second scheme.

[0040] Step 140: Perform an evaluation process of finding the intersection and removing high-value buildings between the first fire lane planning scheme and the second fire lane planning scheme, and output the target fire protection optimization scheme.

[0041] This step involves automatically integrating and optimizing the two alternative solutions to arrive at the final executable solution.

[0042] The specific implementation is as follows: Scheme Element Extraction: Two sets of key elements are extracted from the first scheme and the second scheme respectively: {Set of buildings to be demolished A, Set of roads to be renovated A} and {Set of buildings to be demolished B, Set of roads to be renovated B}.

[0043] Intersection processing: Intersection of Buildings to be Demolished: In GIS, spatial intersection operations are performed on sets A and B to obtain a preliminary intersection of buildings to be demolished, C. These buildings are those that both plans agree should be demolished, representing the highest level of consensus.

[0044] Intersection of Roads to be Modified: Similarly, spatial intersection operations are performed on road set A and road set B to obtain the final road set D to be modified. These road segments are the most core paths recommended for modification by both schemes, representing the most important paths for improving safety and protection value.

[0045] High-value buildings are removed: To ensure the acceptability of the plan, a building value threshold is set (e.g., a value score ≥ 70 points, corresponding to a municipal-level or above cultural relic protection unit or a well-preserved historical building). The system automatically checks each building in the initial intersection of buildings to be demolished, C. If its value score is higher than this threshold, it is removed from the demolition list, forming the final set of buildings to be demolished, E.

[0046] Solution Generation and Output: The system integrates the final set of roads to be modified (D) and the final set of buildings to be demolished (E) to generate the final target fire safety optimization plan. This plan is output in clear drawing and list format. The drawings indicate the specific locations and widths of the recommended widened or added fire lanes, as well as the building numbers and outlines of the buildings to be demolished. The lists detail the start and end points, length, and modification methods of each modified road, and the number, current status, and a brief value assessment of each building to be demolished. This plan represents the optimal solution that balances fire risk reduction, high-value building protection, and the cost-effectiveness of modification.

[0047] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0048] As a further optional embodiment, after the step of performing the intersection and high-value building elimination evaluation process on the first fire lane planning scheme and the second fire lane planning scheme to output the target fire protection optimization scheme, the method further includes: Based on the updated road and building data after implementing the target fire protection optimization scheme, and combined with the population structure data, the evacuation cost grid is calculated to determine the new evacuation locations; Based on the evacuation cost grid, the shortest path analysis method is used to generate evacuation paths from the ends of each street to the newly added evacuation sites and the existing evacuation sites, forming an evacuation system scheme that works in conjunction with the target fire protection optimization scheme.

[0049] like Figure 3 As shown, after completing step 140 and outputting the target fire protection optimization scheme, the method further includes a step of generating an evacuation system scheme, thereby forming a complete fire protection system solution from "rescue route optimization" to "personnel evacuation planning". This extended step is implemented as follows: Based on the updated road and building data after implementing the target fire protection optimization scheme, and in conjunction with the population structure data, an evacuation cost grid is calculated to determine new evacuation locations.

[0050] The purpose of this step is to plan an efficient personnel evacuation system based on the optimized physical space. Specific implementation includes: Data Environment Update: First, create a "post-planning scenario" data environment in the GIS. Merge the final set of roads to be modified (output from step 140) with the original road data to generate updated road network data, which includes all proposed new or widened fire lanes. Simultaneously, remove the final set of buildings to be demolished from the original building data to generate updated building outline data. This step ensures that the evacuation analysis is conducted within the future spatial layout of the neighborhood after fire safety modifications.

[0051] Evacuation cost grid calculation: Basic traffic speed assignment: Based on the width and grade of each road (or road segment) in the updated road network, and with reference to the study of pedestrian walking speed, an initial traffic speed is assigned to it (e.g., for roads with a width > 4m, the speed is set to 1.2m / s; for roads with a width of 2-4m, the speed is set to 1.0m / s; for roads with a width < 2m, the speed is set to 0.8m / s).

[0052] Population impact coefficient calculation: Spatial correlation analysis is performed based on population structure data.

[0053] Age structure impact coefficient (K_age): The age composition of the population within a certain service radius (e.g., 50 meters) of each road is statistically analyzed. The formula K_age = [(P_children proportion × 0.7) + (P_young and middle-aged proportion × 1.2) + (P_elderly proportion × 0.6)] / 0.95 is used for normalization. This coefficient reflects the impact of the mixing of different age groups on the overall evacuation speed.

[0054] Population density impact coefficient: Through kernel density analysis, the estimated population distribution within buildings (estimated using building area and per capita area) is diffused to the entire block space, generating a population density raster. The traffic speed at each point on the road will be reduced according to the population density of its location using a preset speed-density relationship function.

[0055] Raster Generation: The updated road network is converted into a raster, where the value of each cell is determined by the adjusted traffic speed of its road segment (initial speed × K_age × density reduction factor). The travel time (cost) from each cell to the nearest evacuation exit (or location) is then calculated: travel time = cell size / adjusted traffic speed. These values ​​are then summed to generate the evacuation cost raster for the entire area. The value of each cell in this raster represents the theoretical minimum time cost required to evacuate from that point to a safe area.

[0056] New evacuation sites identified: Potential open space identification: In the updated building outline data, identify all open spaces without building coverage (such as plazas, open spaces, and green spaces). At the same time, consider the plots of buildings that will eventually be demolished as potential new open spaces that can be obtained through redevelopment.

[0057] Site selection analysis and screening: Buffer and overlay analyses are performed on these potential spaces to exclude areas that do not meet minimum safety distance requirements (e.g., distance from high-risk buildings) or have unfavorable terrain conditions. For the remaining candidate sites, their effective area, shape regularity, and connectivity to main roads are evaluated.

[0058] Service coverage verification: Existing emergency shelters and newly selected candidate sites are analyzed together using a network analysis. The service area coverage is calculated based on a 5-minute walk radius. If blind spots still exist, the locations of candidate sites are iteratively adjusted or additional sites are considered until the vast majority of the area (e.g., over 95%) is covered. The final list of new evacuation sites and their specific locations are then determined.

[0059] Based on the evacuation cost grid, the shortest path analysis method is used to generate evacuation paths from the ends of each street to the newly added evacuation sites and the existing evacuation sites, forming an evacuation system scheme that works in conjunction with the target fire protection optimization scheme.

[0060] This step aims to plan specific optimal evacuation routes for each micro-unit within the block.

[0061] Evacuation route generation: All road intersections, dead ends, and entrances to important public buildings within the block will be designated as evacuation starting points.

[0062] Set the entrances of all emergency shelters (including existing ones and those newly identified in step 150) as evacuation endpoints (convergence points).

[0063] In GIS, shortest path algorithms (such as Dijkstra's algorithm) are used, with evacuation cost grids as the traffic resistance surface, to calculate the path with the minimum cumulative cost from each evacuation starting point to the nearest evacuation endpoint. This path is the theoretically optimal evacuation path.

[0064] Solution integration and output: The system automatically integrates all generated optimal evacuation routes to form a networked evacuation route plan.

[0065] This evacuation route plan, together with the layout of newly added evacuation sites, will form a complete evacuation system plan.

[0066] Ultimately, the target fire safety optimization plan (fire rescue and transportation system) and the evacuation system plan were output together. These two plans are based on the same spatial data and are interconnected. Fire lanes form the main evacuation routes, and evacuation sites rely on the spaces freed up after the renovation or areas with optimized accessibility. This achieves integrated and coordinated planning of "rescue" and "evacuation," providing a complete, spatialized solution for historical and cultural districts, from external emergency response to the safe evacuation of internal personnel.

[0067] By implementing the above optional steps, the method of the present invention not only optimizes the "entry" passage for fire rescue, but also systematically plans the "evacuation" path and safety space for personnel, significantly improving the overall resilience of complex historical districts in the face of fire.

[0068] As a further optional embodiment, the step of calculating the evacuation cost grid based on the updated road and building data after implementing the target fire protection optimization scheme, combined with the population structure data, specifically includes: Based on the updated road and building data, determine the initial traffic speed for each road; Based on the population structure data, calculate the age structure influence coefficient and population density influence coefficient of the service population around each road; The evacuation cost grid is calculated by substituting the initial travel speed, the age structure influence coefficient, and the population density influence coefficient into the travel time model.

[0069] As a further optional embodiment, in step 150, the step of "calculating the evacuation cost grid based on the updated road and building data after implementing the target fire protection optimization scheme, and in combination with the population structure data" is specifically implemented as follows, aiming to accurately quantify the time cost in the evacuation process through multi-factor fusion: Determine the initial passage speed: The system reads the updated road data, which integrates all roads to be newly built or widened in the target fire safety optimization plan.

[0070] Based on the physical properties of each road or road segment, an initial travel speed is assigned according to a preset mapping rule. This mapping rule is based on factors such as road width and road surface material. For example, paved roads with a width of 4 meters or more have an initial speed of 1.4 meters per second; alleyways with a width between 2 and 4 meters have an initial speed of 1.0 meter per second; and narrow passages or staircases with a width of less than 2 meters have an initial speed of 0.7 meters per second. This initial speed reflects the walking ability of people in an ideal, uncongested, single-group environment.

[0071] Calculate the population impact coefficient: Age structure influence coefficient (K_age): The system is based on demographic data, which is typically linked to building or statistical units. First, through spatial correlation (such as proximity allocation or service area analysis), the source of the "impact population" served by each road is determined.

[0072] Then, the percentage of people in each age group (e.g., 0-14 years old (children), 15-64 years old (young adults), 65 years old and above (elderly)) in the affected population is calculated.

[0073] Based on the typical speed characteristics of different age groups during emergency evacuation, a speed weighting factor is set for each age group (e.g., 0.8 for children, 1.2 for young adults, and 0.7 for the elderly).

[0074] The age structure impact coefficient is calculated using the following model: K_age = Σ (population proportion of each age group × speed weighting factor for that age group). This coefficient comprehensively reflects the objective impact of the age composition of the road service population on the overall evacuation speed, and its value is usually between 0.7 and 1.2.

[0075] Population density impact coefficient (K_density): The system uses information such as building outline, number of floors and per capita area to estimate the population in a building, and generates a continuous raster map of population density distribution through spatial allocation models (such as kernel density estimation).

[0076] For each road segment or sampling point in the road network, the system extracts the population density value of its location.

[0077] Based on classic population flow theories or empirical models (e.g., velocity-density relationship functions, such as V=V_free*(1 -a*ρ^b), where V is the actual velocity, V_free is the free-flow velocity, ρ is the density, and a and b are parameters), population density values ​​are converted into corresponding velocity reduction coefficients, i.e., the population density influence coefficient K_density. This coefficient is close to 1.0 when the population is sparse and decreases as the density increases, simulating the slowing effect of congestion on evacuation speed.

[0078] Calculate the evacuation cost grid: The system substitutes the initial travel speed, age structure influence coefficient K_age, and population density influence coefficient K_density of each road into the travel time model.

[0079] A typical travel time model is: Road segment travel time = Road segment length / (Initial travel speed × K_age × K_density). This model comprehensively considers the combined effects of road physical conditions, population age structure, and real-time distribution density on evacuation efficiency.

[0080] The system uses this model to calculate the travel time cost of each basic unit in the road network, and based on this, generates an evacuation cost grid covering the entire block through rasterization or network analysis techniques. The value of each cell in this grid represents the minimum theoretical time cost required to evacuate from that location to the nearest safe exit, providing a high-precision spatial cost benchmark for subsequent evacuation route planning and site selection.

[0081] Through the implementation of this embodiment, the calculation of evacuation cost grid is no longer based on a single road attribute, but deeply integrates demographic sociology characteristics and micro-spatial distribution, significantly improving the scientific nature and simulation accuracy of the evacuation analysis model, making the final generated evacuation system scheme closer to the real scenario, and effective and reliable.

[0082] As a further optional embodiment, the step of performing fire risk assessment and building value assessment based on the multi-source street data to generate fire risk raster maps and building value raster maps specifically includes: Fire risk assessment based on the aforementioned multi-source neighborhood data includes: Based on the aforementioned building base map data, road base map data, and population structure data, a multi-level fire risk assessment index system is constructed. The weights of each level of the fire risk assessment index system were determined using the analytic hierarchy process (AHP). The evaluation index values ​​in the multi-source street data are converted into spatial raster data, and weighted and superimposed according to the corresponding weights to generate the fire risk raster map. Building valuation based on the aforementioned multi-source neighborhood data includes: Based on the aforementioned building base map data, and according to the building's cultural relic protection level and physical quality information, a building value grid representing the building's comprehensive value is calculated.

[0083] like Figure 4 As shown, in step 120, the specific technical implementation of the step of "conducting fire risk assessment and building value assessment based on the multi-source street data, and generating fire risk raster map and building value raster map" is as follows, aiming to transform multi-source heterogeneous data into a unified and computable spatial decision layer through a structured quantitative model: A. Fire risk assessment and fire risk grid generation Construct a multi-level fire risk assessment indicator system: Based on fire science, fire safety regulations, and the characteristics of historical and cultural blocks, the system constructs a three-level evaluation indicator system that includes a target layer, a criterion layer, and an indicator layer.

[0084] The target layer is "fire risk".

[0085] The criteria layer is used to summarize the main sources of risk, and typically includes: Hazardous factors: reflect the potential conditions in the area itself that could cause or make fires more likely to occur.

[0086] Vulnerability of disaster-bearing structures: reflects the sensitivity and vulnerability of buildings, populations, etc., to fire damage.

[0087] Weakness in disaster prevention and mitigation capabilities: This reflects the shortcomings in the region's ability to prevent, resist, and respond to fires.

[0088] The indicator layer consists of directly measurable, specific parameters that correspond one-to-one with multi-source street data. For example: Indicators that may be classified as “hazardous factors” include: building density (derived from building base maps), estimated electrical load density (derived from building function and area), and density of key fire source units (such as the distribution of catering establishments).

[0089] Indicators that may be categorized as “vulnerability of disaster-bearing structures” include: the proportion of timber-framed buildings (derived from the structural attributes of building base maps), population density (derived from population structure data), and the proportion of elderly population (derived from population structure data).

[0090] Indicators categorized under "weakness in disaster prevention and mitigation capabilities" may include: insufficient effective width of fire lanes (derived from comparison of road base maps and standards), area of ​​blind spots in fire station services (derived from road network analysis), and the proportion of areas with poor accessibility to emergency shelters.

[0091] The weights of the indicators are determined using the analytic hierarchy process (AHP). The system has a built-in or connected AHP calculation module. During implementation, experts in fields such as fire safety, urban planning, and heritage protection are invited to conduct pairwise importance comparisons of various indicators under the same criteria to form a judgment matrix.

[0092] The system calculates the largest eigenvalue and the corresponding eigenvector for each judgment matrix. The normalized eigenvector becomes the weight of the index at that layer relative to the criteria at the upper layer. Simultaneously, the system performs a consistency check (calculating the consistency ratio CR). If CR < 0.1, the judgment matrix is ​​considered reasonable and the weights are valid; otherwise, the system prompts the expert to readjust the judgment.

[0093] Ultimately, the system integrates the weights of each layer and calculates the comprehensive weight of each bottom-level indicator relative to the overall target (fire risk) through multiplication.

[0094] Rasterization calculation generates a fire risk raster map: Rasterization of indicator data: The system uses GIS tools to uniformly convert the source data (vector polygons, vector lines, point densities) of all indicator layers into raster data with the same geographical range, coordinate system, and spatial resolution (e.g., 0.5m × 0.5m). For example, building surface data is converted into building density raster by calculating surface density; road line data is converted into road network density raster by line density analysis; and population point data is converted into population distribution raster by kernel density estimation.

[0095] Weighted overlay analysis: In a GIS raster computing environment, the system calculates fire risk value (per pixel) according to the formula: Fire Risk Value (per pixel) = Σ (i-th indicator raster value × i-th indicator's comprehensive weight). This operation calculates independently for each pixel, ultimately generating a continuous fire risk raster map where pixel values ​​represent the level of fire risk. The higher the value, the greater the overall fire risk at that spatial location.

[0096] B. Building valuation and generation of building value raster maps Quantitative Assessment Model: Architectural value assessment focuses on its irreplaceable historical and cultural value. The system implements a quantitative assessment model: Architectural Value Score = f(Cultural Relic Protection Level Score, Architectural Quality Score). Specifically, a linear weighted model can be used: Value Score = α × S_protection + β × S_condition. Where S_protection is a standard score directly assigned based on the statutory protection level at the national, provincial, municipal, or district level (e.g., National Protection = 100, Provincial Protection = 85, Municipal Protection = 70, Historical Building = 55, Traditional Style Building = 40, General Building = 10). S_condition is an assessment score (0-100 points) based on experts' evaluation of the building's preservation status (e.g., structural safety, stylistic integrity, material authenticity). α and β are preset weighting coefficients, usually α > β, to highlight the decisive role of the cultural relic protection level.

[0097] Generating a building value raster: The system iterates through each building in the base map data and calculates its value score based on its attributes. Then, using the GIS "polygon to raster" tool, this score is used as a conversion field to convert the building vector polygon data into raster data. In the conversion settings, the spatial extent and resolution of the output raster are specified to be completely consistent with the fire risk raster map. For cases where the same cell is covered by multiple buildings (such as building edges), the maximum value method or area weighting method can be used for assignment. The final generated building value raster map uses cell values ​​that represent the historical and cultural value of the building at that location; the higher the value, the higher the priority it should be for protection in fire safety renovations.

[0098] Through the implementation of this embodiment, both fire risk assessment and building value assessment have achieved the transformation from qualitative description to quantitative spatial expression. The two generated raster maps provide accurate, intuitive and calculable spatial basis for subsequent multi-objective optimization decisions.

[0099] As a further optional embodiment, the step of generating a first fire lane planning scheme based on the fire risk grid map and a second fire lane planning scheme based on the building value grid map specifically includes: Based on the building value grid and the spatial proximity relationship between the building and the existing road, a building demolition cost grid is generated through cost distance analysis. Based on the road base map data, the initial toll cost grid of the existing road network and the target toll cost grid of the planned road network are calculated respectively, and the difference between the two is calculated as the road improvement potential grid. A fire road feasibility grid is generated by performing a composite calculation based on the building demolition cost grid and the road improvement potential grid, which represents the comprehensive feasibility of setting a specific road section as a fire road. Spatial overlay analysis is performed on the fire risk grid map and the fire road feasibility grid to generate a first fire lane planning scheme with the goal of prioritizing the improvement of road connectivity in high fire risk areas. By performing spatial overlay analysis on the building value raster map and the fire road feasibility raster, a second fire access planning scheme with the goal of prioritizing the protection of high-value buildings and improving the accessibility of roads around high-value buildings is generated.

[0100] In step 130, the step of "generating a first fire lane planning scheme based on the fire risk grid map and generating a second fire lane planning scheme based on the building value grid map" specifically includes a series of key spatial analysis and decision-making models, aiming to quantify the transformation potential and generate goal-oriented differentiated solutions, as follows: like Figure 5 , Figure 6 As shown, the first step: key auxiliary grid calculation Generate a building demolition cost grid: The system uses the architectural value grid generated in the aforementioned steps as its basic input. The higher the grid value, the greater the historical and cultural value of the building, and the higher the cost of demolition.

[0101] The system utilizes the cost distance analysis tool in GIS spatial analysis. The edge line of the existing road network is set as the "source," which is the starting point for cost distance calculation (the area with a cost of 0).

[0102] Using the architectural value grid as the "cost surface," each cell in the space has its corresponding "travel cost," the value of which is equal to the architectural value of that cell.

[0103] After performing cost distance analysis, a building demolition cost grid is generated. The value of each cell in this grid represents the minimum cumulative "building value cost" required to reach that cell from the nearest road "source". Therefore, areas closer to the road and with lower building value have smaller "demolition cost" values, meaning that the overall cost of making room for fire lanes is relatively low at these locations.

[0104] Generate a grid of road enhancement potential: The system performs a dual analysis based on road base map data. First, based on the actual width and alignment of the existing road network, and according to a preset speed-width relationship model, it calculates and generates an initial traffic cost grid for the existing road network. This grid reflects the theoretical time cost for fire trucks or rescue forces to travel to each location in the current road network.

[0105] Secondly, based on the requirements for fire truck access roads in national fire protection technical specifications (such as the "Code for Fire Protection Design of Buildings"), and under the constraints of the existing space, an ideal road network that meets the minimum width, turning radius, and network connectivity should be planned (as the target).

[0106] Based on this ideal road network plan, the same model is used to calculate and generate the target toll cost grid for the planned road network.

[0107] The system calculates the difference between two cost grids: Road Improvement Potential Grid = Current Traffic Cost Grid - Target Traffic Cost Grid. Areas with a positive and larger grid value represent areas with greater potential for traffic efficiency improvement through modifications such as widening, straightening, and connectivity.

[0108] Generate a fire road feasibility grid: The system combines the results of the two analyses mentioned above to calculate the fire road feasibility grid. The calculation formula is usually: Feasibility Grid = Road Improvement Potential Grid / Building Demolition Cost Grid.

[0109] The core logic of this operation is to conduct a "benefit-cost" ratio analysis. The road improvement potential represents the "benefit" of the renovation (improved traffic efficiency), while the building demolition cost represents the "cost" of the renovation. Therefore, the higher the feasibility grid value of an area, the higher the "cost-effectiveness" of carrying out fire road renovations there, meaning that a relatively small loss of cultural value can be exchanged for a significant improvement in fire rescue efficiency.

[0110] like Figure 7 , Figure 8 As shown, the second step is to generate planning schemes based on different value orientations. Generate the first fire lane planning scheme (high-risk protection takes priority): The system performs spatial overlay analysis (such as grid multiplication or weighted overlay) on the fire risk grid map and the fire road feasibility grid.

[0111] This analysis aims to identify spatial overlap areas with high fire risk and high feasibility for renovation. The system automatically extracts these high-priority areas by setting thresholds or using the natural breakpoint method.

[0112] Within these areas, the system automatically generates suggested new or widened fire lane routes (vector lines) based on a pre-set ideal road network and fine-tunes according to constraints such as terrain and property rights. It also identifies the set of buildings that need to be demolished to implement these route modifications. The resulting spatial modification plan constitutes the first fire lane planning scheme, whose core objective is to prioritize reducing the difficulty of rescue operations in high-fire-risk areas.

[0113] Generate a second fire lane planning scheme (prioritizing the protection of high-value buildings): The system performs spatial overlay analysis of building value raster maps and fire road feasibility raster maps.

[0114] This analysis aims to identify spatial areas with high architectural value and high feasibility for surrounding redevelopment. Specifically, it identifies buildings and their adjacent areas that are inherently valuable and require special protection, while also having a relatively easy-to-optimize road network environment.

[0115] Within these areas, the system also generates or optimizes connecting routes based on an ideal road network plan, aiming to ensure that every high-value building has at least one reliable rescue route that meets fire safety requirements and can be easily reached. The generated route lines and the corresponding low-value buildings to be demolished together constitute the second fire access planning scheme, the core objective of which is to prioritize the fire accessibility of high-value historical and cultural heritage sites.

[0116] Through the implementation of this embodiment, the generation of the two planning schemes is no longer based on subjective experience or a single indicator, but is derived through the analysis of spatial data and calculation of multi-objective models. This makes the schemes have a clear optimization orientation ("ensuring safety" or "preserving heritage"), and are based on the quantitative evaluation of "benefit-cost" analysis, providing a scientific and clear alternative basis for subsequent comprehensive decision-making.

[0117] As a further optional embodiment, the existing road network toll cost grid is generated based on vector data reflecting the current road width and spatial morphology obtained from open-source geographic information databases or field surveys; The planned road network traffic cost grid is generated based on ideal road network vector data that meets the traffic standards for fire trucks, pre-defined according to fire protection design specifications and the existing spatial layout.

[0118] As a further optional embodiment, the existing road network access cost grid and the planned road network access cost grid involved in the calculation of the road improvement potential grid are generated in the following specific way, aiming to provide an accurate and comparable basic cost surface for the analysis of improvement potential: Generation of the current road network toll cost grid: Data Acquisition and Processing: First, acquire high-precision road network vector data of the target street area from open-source geographic information databases (such as OpenStreetMap) or through on-site laser scanning and total station mapping. This data should accurately reflect the current road centerline orientation, effective traffic width, and key spatial morphological features (such as turning radius, slope, steps, etc.). For irregular alleyways, their minimum effective width needs to be recorded through measurement and digitization.

[0119] Traffic Cost Modeling: Based on the aforementioned vector data, a traffic cost model is established for each road segment in the road network. The core of the cost is travel time. The system assigns a baseline travel speed to each road segment based on its width attribute and the passage capacity standards for fire trucks or rescue equipment (e.g., road segments with a width ≥ 4.0 meters can accommodate large fire trucks, with a speed of 16.7 m / s; those with a width between 2.0 and 4.0 meters can accommodate small fire trucks or motorcycles, with a speed of 10 m / s; those with a width < 2.0 meters only allow portable pumps or pedestrians, with a speed of 4 m / s or 0.2 m / s). For road segments with steep slopes, steps, or other special terrain features, the speed needs to be further adjusted based on the empirical model.

[0120] Rasterization calculation: The vector road network, assigned traffic speeds, is converted into raster data using GIS spatial interpolation or cost allocation tools. The travel time cost of each raster cell is calculated using the following formula: Time cost = Cell size / Traffic speed at the cell's location. The resulting continuous surface is the current road network traffic cost raster, which quantifies the time cost required for rescue forces to traverse each location within a block under existing road conditions.

[0121] Generation of the planned road network toll cost grid: Ideal Road Network Design: The planned road network is not arbitrarily set, but rather digitally designed by planners within the constraints of the existing spatial layout, based on fire protection technical standards such as the "Code for Fire Protection Design of Buildings" (GB50016) regarding minimum width (usually not less than 4 meters), turning radius, turning area settings, and distance from building exterior walls. This process includes: identifying road sections in the existing road network that do not meet the standards and virtually "widening" them to the standard width; planning new connecting fire lanes in densely built-up areas based on potential demolition corridors; and ensuring that the entire block forms a ring-shaped fire road network or meets the evacuation requirements of two different directions. Ultimately, this results in an "ideal" road network vector data layer that meets the requirements of the standards.

[0122] Traffic cost assignment and raster generation: For this "ideal" planned road network vector data, traffic cost modeling rules (the same speed-width correspondence) are used to assign values ​​to it, identical to those used for the existing road network. That is, all roads in the pre-defined planned road network are assumed to meet fire truck passage requirements, thus assigning them a standard fire truck passage speed (e.g., 16.7 m / s for a 4-meter-wide road). Subsequently, the same rasterization process is used to calculate and generate the planned road network traffic cost raster. This raster reflects the theoretical time cost for rescue forces to travel after the street network is ideally modified to fully comply with fire safety regulations.

[0123] Through the implementation of this embodiment, the traffic costs under both the current and planned road network conditions can be spatially expressed and compared under the same set of quantitative standards. The difference between the two (i.e., "road improvement potential") can objectively and accurately measure the improvement in fire rescue efficiency that can be achieved by modifying each spatial location, laying a solid quantitative analysis foundation for subsequent feasibility analysis and scheme generation.

[0124] As a further optional embodiment, the step of performing an evaluation process of taking the intersection of the first fire lane planning scheme and the second fire lane planning scheme and eliminating high-value buildings to output the target fire protection optimization scheme specifically includes: Extract the first set of buildings to be demolished and the first set of roads to be renovated from the first fire lane planning scheme; Extract the second set of buildings to be demolished and the second set of roads to be renovated from the second fire lane planning scheme; Perform a spatial intersection operation between the first set of buildings to be demolished and the second set of buildings to be demolished to obtain a preliminary intersection of the buildings to be demolished. In the initial set of buildings to be demolished, buildings with a value higher than a preset threshold are removed to generate the final set of buildings to be demolished. Perform a spatial intersection operation between the first road modification set and the second road modification set to generate the final road modification set; Based on the final set of buildings to be demolished and the final set of road modifications, a target fire protection optimization plan containing specific spatial locations and engineering measures is generated.

[0125] like Figure 9 As shown, as a further optional embodiment, in step 140, the step of "evaluating the intersection of the first fire lane planning scheme and the second fire lane planning scheme and eliminating high-value buildings to output the target fire protection optimization scheme" is specifically implemented through a series of structured spatial data operations and logical judgments to merge the two differentiated schemes into a final scheme with strong consensus and feasibility. The specific process is as follows: Extraction of solution elements: The system analyzes the first fire lane planning scheme (prioritizing high-risk protection) and the second fire lane planning scheme (prioritizing the protection of high-value buildings). These two schemes are essentially spatialized engineering suggestion datasets.

[0126] From the first scheme, two key spatial element sets are extracted: the first set of buildings to be demolished (A1, which is the outline of the buildings to be demolished according to the scheme) and the first set of road modifications (R1, which is the road segment to be newly built or widened).

[0127] Similarly, from the second scheme, the corresponding second set of buildings to be demolished (A2) and the second set of road renovations (R2) are extracted.

[0128] Consensus screening and safety filtering of buildings to be demolished: Spatial intersection operation: The system performs an "intersection" spatial analysis on the first set of buildings to be demolished (A1) and the second set of buildings to be demolished (A2) in GIS. The result of this operation is the preliminary intersection of the buildings to be demolished (A_intersect), which consists of buildings that appear in both plans. These buildings are those that can be sacrificed under the "common recognition" of both protection orientations (safety preservation and heritage preservation), and have the highest degree of consensus on demolition.

[0129] High-value building protective removal: To ensure the plan does not affect core protected objects, the system accesses the building value raster map or associated attribute table to obtain the building value score for each building in A_intersect. The system sets a preset threshold (e.g., a score ≥ 70 corresponds to a municipal-level or higher cultural relic protection unit or an extremely well-preserved historical building). The system automatically iterates through A_intersect, removing all buildings with a building value score higher than the threshold from the list of buildings to be demolished. This step is a mandatory application of protection rules.

[0130] After elimination, the remaining buildings constitute the final set of buildings to be demolished (A_final). This set simultaneously meets the dual criteria of "need for renovation" and "acceptability of value".

[0131] Consensus building on road improvement measures: The system performs "intersection" or "sameness" spatial analysis on the first road modification set (R1) and the second road modification set (R2) in GIS. This operation generates the final road modification set (R_final).

[0132] R_final contains the road modification segments that are suggested in both options. These segments represent the "greatest common denominator" under different value orientations, and are the most critical basic road network optimization measures that are considered essential for improving overall fire-fighting effectiveness, regardless of whether safety or protection is emphasized.

[0133] Target solution generation and structured output: The system integrates the final set of buildings to be demolished (A_final) with the final set of road modifications (R_final).

[0134] Based on these two core elements, the system automatically generates the final target fire protection optimization plan. This plan is output in the form of structured data objects and visual drawings, specifically including: List of renovation projects: Clearly list the start and end points, current width, planned width, renovation type (widening / new construction), length and estimated workload of each road to be renovated; list the number, address, current building area, building value score and a brief description of the reasons for demolition for each building to be demolished.

[0135] Specialized drawings: On the street map, all road modification segments in R_final are highlighted (different colors can be used to distinguish between new construction and widening), and the outlines of all buildings to be demolished in A_final are clearly delineated.

[0136] Solution Report Summary: Automatically generates a solution summary, explaining that the solution is based on the fusion of the "high-risk priority" and "high-value priority" approaches and has undergone protective filtering. It also summarizes the main engineering workload, the expected increase in fire protection coverage, and the status of the high-value buildings that have been protected.

[0137] Through the implementation of this embodiment, the final fire protection optimization scheme is no longer a simple compromise between two schemes. Instead, through rigorous spatial logic calculations and preset protection rules, a feasible implementation scheme is formed that combines high consensus (intersection) and strict protection (elimination of high value), scientifically balancing the core contradiction between "ensuring safety" and "preserving heritage" in the fire protection renovation of historical and cultural blocks.

[0138] The following describes the AI-based fire protection system optimization device provided by the present invention, such as... Figure 10 As shown, the AI-based fire protection system optimization device described below and the AI-based fire protection system optimization method described above can be referred to in correspondence.

[0139] An AI-based fire protection system optimization device includes: The data acquisition module 1010 is used to acquire multi-source street data of the street to be evaluated, including building base map data, road base map data and population structure data; The raster generation module 1020 is used to perform fire risk assessment and building value assessment based on the multi-source street data, and generate fire risk raster map and building value raster map. The scheme generation module 1030 is used to generate a first fire lane planning scheme based on the fire risk grid map, and to generate a second fire lane planning scheme based on the building value grid map. The scheme optimization module 1040 is used to evaluate the intersection of the first fire lane planning scheme and the second fire lane planning scheme and to eliminate high-value buildings, and output the target fire protection optimization scheme.

[0140] Figure 11 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 11As shown, the electronic device may include: a processor 1110, a communications interface 1120, a memory 1130, and a communication bus 1140, wherein the processor 1110, the communications interface 1120, and the memory 1130 communicate with each other via the communication bus 1140. The processor 1110 can call logical instructions in the memory 1130 to execute an AI-based fire protection system optimization method, which includes: Acquire multi-source street data for the street to be evaluated, including building base map data, road base map data, and population structure data; Fire risk assessment and building value assessment are performed based on the multi-source street data, generating fire risk raster maps and building value raster maps. Based on the fire risk grid map, a first fire lane planning scheme is generated, and based on the building value grid map, a second fire lane planning scheme is generated. The first and second fire lane planning schemes are evaluated by taking their intersection and removing high-value buildings, and the target fire protection optimization scheme is output.

[0141] Furthermore, the logical instructions in the aforementioned memory 1130 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0142] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the AI-based fire protection system optimization method provided by the above methods, the method comprising: Acquire multi-source street data for the street to be evaluated, including building base map data, road base map data, and population structure data; Fire risk assessment and building value assessment are performed based on the multi-source street data, generating fire risk raster maps and building value raster maps. Based on the fire risk grid map, a first fire lane planning scheme is generated, and based on the building value grid map, a second fire lane planning scheme is generated. The first and second fire lane planning schemes are evaluated by taking their intersection and removing high-value buildings, and the target fire protection optimization scheme is output.

[0143] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the AI-based fire protection system optimization method provided by the above methods, the method comprising: Acquire multi-source street data for the street to be evaluated, including building base map data, road base map data, and population structure data; Fire risk assessment and building value assessment are performed based on the multi-source street data, generating fire risk raster maps and building value raster maps. Based on the fire risk grid map, a first fire lane planning scheme is generated, and based on the building value grid map, a second fire lane planning scheme is generated. The first and second fire lane planning schemes are evaluated by taking their intersection and removing high-value buildings, and the target fire protection optimization scheme is output.

[0144] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0145] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0146] 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An AI-based method for optimizing fire protection systems, characterized in that, include: Acquire multi-source street data for the street to be evaluated, including building base map data, road base map data, and population structure data; Fire risk assessment and building value assessment are performed based on the multi-source street data, generating fire risk raster maps and building value raster maps. Based on the fire risk grid map, a first fire lane planning scheme is generated, and based on the building value grid map, a second fire lane planning scheme is generated. The first and second fire lane planning schemes are evaluated by taking their intersection and removing high-value buildings, and the target fire protection optimization scheme is output.

2. The AI-based fire protection system optimization method according to claim 1, characterized in that, After the step of evaluating the intersection of the first fire lane planning scheme and the second fire lane planning scheme and eliminating high-value buildings to output the target fire protection optimization scheme, the following steps are also included: Based on the updated road and building data after implementing the target fire protection optimization scheme, and combined with the population structure data, the evacuation cost grid is calculated to determine the new evacuation locations; Based on the evacuation cost grid, the shortest path analysis method is used to generate evacuation paths from the ends of each street to the newly added evacuation sites and the existing evacuation sites, forming an evacuation system scheme that works in conjunction with the target fire protection optimization scheme.

3. The AI-based fire protection system optimization method according to claim 2, characterized in that, The step of calculating the evacuation cost grid based on the updated road and building data after implementing the target fire protection optimization scheme, combined with the population structure data, specifically includes: Based on the updated road and building data, determine the initial traffic speed for each road; Based on the population structure data, calculate the age structure influence coefficient and population density influence coefficient of the service population around each road; The evacuation cost grid is calculated by substituting the initial travel speed, the age structure influence coefficient, and the population density influence coefficient into the travel time model.

4. The AI-based fire protection system optimization method according to claim 1, characterized in that, The step of performing fire risk assessment and building value assessment based on the multi-source street data to generate fire risk raster maps and building value raster maps specifically includes: Fire risk assessment based on the aforementioned multi-source neighborhood data includes: Based on the aforementioned building base map data, road base map data, and population structure data, a multi-level fire risk assessment index system is constructed. The weights of each level of the fire risk assessment index system were determined using the analytic hierarchy process (AHP). The evaluation index values ​​in the multi-source street data are converted into spatial raster data, and weighted and superimposed according to the corresponding weights to generate the fire risk raster map. Building valuation based on the aforementioned multi-source neighborhood data includes: Based on the aforementioned building base map data, and according to the building's cultural relic protection level and physical quality information, a building value grid representing the building's comprehensive value is calculated.

5. The AI-based fire protection system optimization method according to claim 4, characterized in that, The step of generating a first fire lane planning scheme based on the fire risk grid map and a second fire lane planning scheme based on the building value grid map specifically includes: Based on the building value grid and the spatial proximity relationship between the building and the existing road, a building demolition cost grid is generated through cost distance analysis. Based on the road base map data, the initial toll cost grid of the existing road network and the target toll cost grid of the planned road network are calculated respectively, and the difference between the two is calculated as the road improvement potential grid. A fire road feasibility grid is generated by performing a composite calculation based on the building demolition cost grid and the road improvement potential grid, which represents the comprehensive feasibility of setting a specific road section as a fire road. Spatial overlay analysis is performed on the fire risk grid map and the fire road feasibility grid to generate a first fire lane planning scheme with the goal of prioritizing the improvement of road connectivity in high fire risk areas. By performing spatial overlay analysis on the building value raster map and the fire road feasibility raster, a second fire access planning scheme with the goal of prioritizing the protection of high-value buildings and improving the accessibility of roads around high-value buildings is generated.

6. The AI-based fire protection system optimization method according to claim 5, characterized in that, The current road network toll cost grid is generated based on vector data reflecting the current road width and spatial morphology obtained from open-source geographic information databases or field surveys. The planned road network traffic cost grid is generated based on ideal road network vector data that meets the traffic standards for fire trucks, pre-defined according to fire protection design specifications and the existing spatial layout.

7. The AI-based fire protection system optimization method according to claim 1, characterized in that, The step of evaluating the intersection of the first and second fire lane planning schemes and eliminating high-value buildings to output the target fire safety optimization scheme specifically includes: Extract the first set of buildings to be demolished and the first set of roads to be renovated from the first fire lane planning scheme; Extract the second set of buildings to be demolished and the second set of roads to be renovated from the second fire lane planning scheme; Perform a spatial intersection operation between the first set of buildings to be demolished and the second set of buildings to be demolished to obtain a preliminary intersection of the buildings to be demolished. In the initial set of buildings to be demolished, buildings with a value higher than a preset threshold are removed to generate the final set of buildings to be demolished. Perform a spatial intersection operation between the first road modification set and the second road modification set to generate the final road modification set; Based on the final set of buildings to be demolished and the final set of road modifications, a target fire protection optimization plan containing specific spatial locations and engineering measures is generated.

8. An AI-based fire protection system optimization device, characterized in that, include: The data acquisition module is used to acquire multi-source street data of the street to be evaluated, including building base map data, road base map data and population structure data; The raster generation module is used to perform fire risk assessment and building value assessment based on the multi-source street data, and generate fire risk raster map and building value raster map. The scheme generation module is used to generate a first fire lane planning scheme based on the fire risk grid map, and to generate a second fire lane planning scheme based on the building value grid map. The scheme optimization module is used to evaluate the intersection of the first fire lane planning scheme and the second fire lane planning scheme and to eliminate high-value buildings, and output the target fire protection optimization scheme.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the AI-based fire protection system optimization method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the AI-based fire protection system optimization method as described in any one of claims 1 to 7.

Images