Disaster damage assessment device and method

Abstract

The invention relates to a disaster damage assessment device and method, and relates to the technical field of flood risk assessment, and the method comprises the steps: constructing a standardized data basis through integrating meteorological, hydrological, geographic and asset data; dynamic features such as submerging water depth and duration are extracted by adopting a gridding space correlation technology, and a multi-dimensional feature library is established in combination with disaster-bearing body attributes; linear, exponential, logarithmic, ladder and modified exponential function families are creatively applied, a refined loss rate curve is generated through clustering analysis, and the loss estimation precision is remarkably improved. Various asset losses under different water depth levels are quantified in combination with a spatial analysis algorithm, total factors of houses, agriculture, industry, business and infrastructures are covered, and flood routing and loss distribution are visually displayed through a dynamic visual platform. The method breaks through the bottleneck of the traditional technology, provides time-space accurate quantitative support for flood control decision, and effectively improves the timeliness and scientificity of disaster response.

Description

Technical Field

[0001] The present invention relates to the technical field of flood risk assessment, and in particular to a disaster damage assessment device and method. Background Art

[0002] Currently, flood risk assessment faces numerous challenges: First, accurate and real-time dynamic analysis of flood inundation range, depth, and duration is difficult to achieve. Existing methods often rely on fixed historical data and cannot be adjusted in real time based on actual flood conditions, resulting in delayed analysis results. Second, loss assessment accuracy is limited, failing to fully consider the impact of flood characteristics (such as flow velocity and arrival time) and differences in the properties of the hazard-bearing body on losses, often using crude empirical formulas. Third, the lack of intuitive and efficient visualization prevents decision makers from clearly presenting complex flood risk information, hindering decision-making efficiency and accuracy. Therefore, it is necessary to develop a disaster loss assessment device and method.

[0003] Existing technologies generally build models based on fixed historical data, which makes it difficult to dynamically adapt to changes in real-time flood conditions and cannot promptly reflect dynamic situations such as the expansion of the inundation range and changes in water depth during the development of the flood. This leads to deviations between the analysis results and the actual flood conditions, poor timeliness, and inability to provide effective support for real-time flood control decisions.

[0004] The empirical formula does not carefully consider the impact of differences in flood characteristics (such as flow velocity, arrival time, etc.) and asset attributes (such as building structure, farmland crop type, etc.) on losses, resulting in poor accuracy in loss assessment results and possible large errors, which affects the rationality of flood control and disaster reduction resource allocation.

[0005] The lack of intuitive geographic information display, most of the results are presented in the form of text reports or simple charts, which makes it difficult to clearly show the relationship between the flood inundation area and the actual geographical space, as well as the distribution of losses in each region. This makes it difficult for decision makers to understand and make decisions, and reduces decision-making efficiency. Based on this, it is necessary to study a disaster loss assessment device and method. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a disaster loss assessment device and method, which effectively solves the problem that the existing method adopts simplified empirical formulas, fails to fully couple the dynamic characteristics of floods (such as flow velocity, scouring time) with the differences in the properties of disaster-bearing bodies (such as building structure type, asset value distribution), resulting in systematic deviations in loss estimates, excessive reliance on historical data, and difficulty in capturing the dynamic characteristics of the inundation range and water depth changes during the evolution of floods in real time, resulting in the analysis results lagging behind the actual development of the disaster.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is: a disaster damage assessment device, comprising Data collection and pre-processing module: used to collect flooding data, geographic data and asset data of the disaster-prone body; and to clean, convert and standardize the collected raw data; Data analysis module: Utilizes inundation data, geographic data, and asset data in historical data to obtain inundation characteristics, geographic characteristics, and asset characteristics, respectively. Geographic characteristics include region numbers. For hazard-prone bodies, grid processing is performed on the hazard-prone bodies based on the distribution of administrative regions, and the gridded regions are numbered. Inundation characteristics include inundation depth, inundation duration, arrival time and maximum flow velocity corresponding to the area number; The asset characteristics include asset data distributed over corresponding area numbers; Loss rate assessment module: Based on the area number, cluster analysis is performed on the historical asset data and inundation data for the area number. According to the distribution characteristics of the sorted data, the corresponding loss rate function is matched to fit the loss rate curve. Based on the loss rate curve, the asset loss rate of different types of flooded and hazard-bearing objects is assessed; Disaster damage assessment module: used to store the asset loss rate generated by the inundation data acting on the corresponding area number, establish the relationship between the asset loss rate and the inundation characteristics, and calculate the direct economic losses caused by the flood to various assets based on the asset value; Visualization module: Combining the PostGIS spatial database and GeoServer map service, the Vue+OpenLayers front-end framework is used to build a visualization platform to provide intuitive and dynamic flood simulation and statistical analysis results.

[0008] Furthermore, the loss rate function includes a linear function, an exponential function, a logarithmic function, a step function and a modified exponential function; wherein the linear function is used for the linear loss relationship within the low water depth range; the exponential function is used for the rapidly rising loss rate within the high water depth range; the piecewise function is used for the nonlinear change in different water depth intervals; the logarithmic function is used for the scenario where the loss rate increases slowly with the water depth; the step function is used for the scenario where the loss rate suddenly changes within the water depth range, and the modified exponential function is used to evaluate the impact of the submergence duration t.

[0009] Furthermore, according to the water depth levels defined in the loss rate curve, flood data in different water depth ranges are integrated into the same result set. At the same time, the geometric merging algorithm Union is used to merge the geometric shapes that meet the conditions to generate flood inundation areas of different water depth levels.

[0010] Furthermore, it also includes a spatial analysis module, which includes a geometric merging algorithm, a geometric intersection algorithm, a geometric intersection calculation, a geometric area calculation, a geometric length calculation, and a geometric centroid calculation; it performs spatial topology checks on grid layers, land use data, and disaster-prone body distribution data to ensure that there are no topological errors such as unclosed, gaps, overlaps, and self-intersections.

[0011] Furthermore, based on the spatial analysis module, the number of point-type disaster-prone areas, the area of ​​surface-type disaster-prone features, and the length of line-type disaster-prone features are extracted; the number of point-type disaster-prone areas is used to obtain the number of disaster-prone features that meet the spatial intersection conditions, and the ratio of the number of disaster-prone feature units to the number of intersection parts is calculated to quantify the impact of floods on disaster-prone features; the area of ​​surface-type disaster-prone features is extracted using the geometric area algorithm Area, and the ratio of the area of ​​the disaster-prone feature unit to the area of ​​the intersection part is calculated to quantify the impact of floods on disaster-prone features; the length of line-type disaster-prone features is obtained by extracting the length of the disaster-prone feature unit using the geometric length algorithm Length, and the ratio of the length of the disaster-prone feature unit to the length of the intersection part is calculated to quantify the impact of floods on disaster-prone features.

[0012] Furthermore, based on the spatial analysis module, the inundation layer is superimposed with the administrative district layer, the residential map layer, and the cultivated land surface layer to obtain the flooded administrative district area, flooded residential area, and flooded cultivated land area corresponding to different flooding depth levels in different scenarios, and displayed in the visualization module.

[0013] Furthermore, asset data include household property, household housing, agricultural losses, industrial losses, industrial output value, commercial assets, commercial main income, railways, provincial roads and above, and roads below provincial roads; for each type of asset, the loss rate under different flooding characteristics is compiled separately.

[0014] A disaster damage assessment method comprises the following steps: Step 1: Collect inundation data, geographic data, and asset data of the hazard-prone body; and clean, convert, and standardize the collected raw data; Step 2: derive flood characteristics, geographic characteristics, and asset characteristics respectively from the inundation data, geographic data, and asset data in the historical data; The geographical feature is the regional number, and the specific processing method is as follows: for the disaster-prone body, the disaster-prone body is gridded based on the distribution of administrative regions, and the gridded areas are numbered; Inundation characteristics include inundation depth, inundation duration, arrival time and maximum flow velocity corresponding to the area number; The asset characteristics include asset data distributed over corresponding area numbers; Step 3: Perform cluster analysis on the historical asset data and flooding data for each area number to generate a flooding dataset and an asset loss rate dataset based on the area number. Based on the loss rate function set for the corresponding asset, the loss rate function is trained using the flooding dataset and the asset loss rate dataset to obtain the asset loss rate caused by the flooding data for each area number. This results in an asset loss rate assessment model. Step 4: Extract and process the real-time inundation data, geographic data, and asset data in the same manner as in Steps 1 and 2 to obtain inundation characteristics and asset characteristics under the corresponding area number. Input the inundation characteristics and asset characteristics into the asset loss rate assessment model to obtain the losses of various assets. Step 5: Add up the losses of various assets to obtain the total asset losses.

[0015] Furthermore, the loss rate function includes linear function, exponential function, logarithmic function, step function and modified exponential function; using the selected function form, combined with the sorted flooding data set and asset loss rate data set, a statistical analysis method is used to fit the loss rate curve, and by adjusting the parameters of the function, the error between the fitting curve and the actual data points is minimized, and finally the loss rate curve equation for each type of asset is obtained.

[0016] Furthermore, the statistical analysis method is least squares method or maximum likelihood estimation.

[0017] The beneficial effects of the above technical solution are as follows: The present invention provides a method for accurately and dynamically analyzing flood impacts and efficiently assessing losses. Through multi-source data fusion and dynamic analysis, it achieves highly accurate quantitative assessment of flood losses. First, inundation, geographic, and asset data are collected and preprocessed to ensure data quality. Second, geographic features are established based on grid coding of administrative regions, and dynamic features such as inundation depth and duration are extracted. These features are then correlated with asset data to form a multidimensional feature library. Cluster analysis is then used to construct a regionalized inundation-loss rate dataset. Loss rate curves are fitted using linear, exponential, logarithmic, step, and modified exponential functions to establish a refined assessment model. Finally, feature extraction is performed on real-time data and input into the model to calculate and summarize losses for various asset types, providing precise spatial and temporal quantitative support for flood control decision-making. Refined loss calculation formulas are developed for different asset types, comprehensively considering the impact of flood characteristics (such as flow velocity and arrival time) and hazard-bearing object properties (such as building structure and land use type) on losses. This approach also implements intuitive and visual flood risk map impact analysis and loss assessment technology, meeting the requirements for efficient, accurate, and reliable flood control and disaster reduction decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Flow chart for the implementation of the present invention; Figure 2Provide a logical block diagram for loss assessment; Figure 3 Provide indicators for flood impact analysis and loss assessment; Figure 4 Schematic diagram of GIS expression of statistical indicators; Figure 5 It is a logic analysis flow chart of the present invention; Figure 6 This is a system diagram of the present invention. DETAILED DESCRIPTION

[0019] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1. This embodiment aims to provide a disaster damage assessment device, which mainly uses a flood impact analysis and loss assessment model. The existing flood risk assessment has the following problems: First, the dynamic response capability is weak. The existing model relies too much on historical data and is difficult to capture the dynamic characteristics of the inundation range and water depth changes during the evolution of the flood in real time, resulting in the analysis results lagging behind the actual development of the disaster; second, the assessment accuracy is insufficient. The traditional method uses a simplified empirical formula and does not fully couple the flood dynamic characteristics (such as flow velocity, flushing time) with the differences in the attributes of the disaster-bearing body (such as building structure type, asset value distribution), resulting in systematic deviations in loss estimation; third, the decision support efficiency is low. There is a lack of visualization means to associate complex risk information with geographic space, making it difficult to intuitively present the flood evolution trajectory, the distribution of high-risk areas and the spatiotemporal characteristics of losses, which restricts the timeliness and accuracy of flood control decisions.

[0020] A disaster damage assessment device includes a data acquisition and preprocessing module for collecting inundation data, geographic data, and asset data of disaster-prone objects; and cleaning, converting, and standardizing the collected raw data. Specifically, the device integrates real-time rainfall data from meteorological departments, water level and flow monitoring data from hydrological stations, topographic data (such as altitude and slope) from geographic information systems, census data, and various asset data (including building information, land use types, and industrial and agricultural production data) to provide a comprehensive and accurate data foundation for flood risk assessment. The device also cleans, converts, and standardizes the collected raw data. For example, geospatial data in different formats is uniformly converted into a coordinate system and data format that meets system requirements. Missing value interpolation, outlier detection, and correction are performed on meteorological and hydrological data to ensure data quality and meet the needs of subsequent analysis and calculation.

[0021] Data analysis module: Utilizes the inundation data, geographic data and asset data in the historical data to obtain inundation characteristics, geographic characteristics and asset characteristics respectively.

[0022] The geographical features include regional numbers. The disaster-prone bodies are gridded based on the distribution of administrative regions, and the gridded areas are numbered. According to the "regional code", the spatial association between socioeconomic statistical data and administrative division and land use data is established through spatial distribution to ensure the correlation of socioeconomic data in spatial distribution.

[0023] Inundation characteristics include flood depth, duration, arrival time, and maximum flow velocity, corresponding to the area number. By deeply integrating real-time meteorological, hydrological, and geospatial data with efficient vector-raster hybrid computing, this system enables real-time dynamic analysis of flood depth, duration, arrival time, and maximum flow velocity. Unlike existing static analysis methods, this system can promptly update analysis results as the flood develops, providing a timely basis for flood control decisions.

[0024] By superimposing flood analysis grids, administrative boundaries, land use and other data, spatial data with grid, administrative unit, disaster-prone body or land type information and their topological relationships are analyzed and generated; through aggregation statistics, based on map data such as administrative boundaries and land use, socioeconomic statistics, calculated or historical (measured or investigated) flood inundation element values ​​and their spatial distribution data, multi-indicator flood impact results based on disaster-prone body type, inundation characteristics and administrative districts, and their corresponding spatial distribution data are obtained.

[0025] Asset characteristics include asset data distributed in corresponding area numbers; asset data include household property, household housing, agricultural losses, industrial losses, industrial output value, commercial assets, commercial main income, railways, provincial roads and above, and roads below provincial roads; for each type of asset, the loss rate under different inundation characteristics is sorted out separately.

[0026] Loss rate assessment module: Based on the area number, cluster analysis is performed on the historical asset data and inundation data of the area number. According to the distribution characteristics of the sorted data, the corresponding loss rate function is matched to fit the loss rate curve. Based on the loss rate curve, the asset loss rate of different types of flooded and disaster-prone objects is evaluated.

[0027] In specific implementation, the loss rate function includes linear function, exponential function, logarithmic function, step function and modified exponential function; among them, the linear function is used for the linear loss relationship in the low water depth range; the exponential function is used for the rapidly rising loss rate in the high water depth range; the piecewise function is used for nonlinear changes in different water depth intervals; the logarithmic function is used for the scenario where the loss rate increases slowly with water depth; the step function is used for the scenario where the loss rate suddenly changes within the water depth range, and the modified exponential function is used to evaluate the impact of the submergence duration t.

[0028] Specific implementation linear function: linear loss relationship for low water depth range.

[0029] ;in : No. Assets in the Loss rate at grade water depth; : No. Level water depth value; 、 : No. Parameters of the linear function of the asset class.

[0030] Exponential function: used for rapid ascent loss rates in high water depth ranges.

[0031] ; : No. The exponential decay coefficient of the asset class.

[0032] Piecewise function: used for nonlinear changes in different water depth ranges.

[0033] ; 、 : Segmentation threshold (divided according to actual water depth data); 、 、 : No. The parameters of the quadratic polynomial for the asset class.

[0034] Logarithmic function: used for scenarios where the loss rate increases slowly with water depth.

[0035] ; 、 : No. Parameters of the logarithmic function of the asset class; : Natural logarithm function.

[0036] Step function: used for scenarios where the loss rate changes suddenly within the water depth range.

[0037] ; 、 : water depth threshold; 、 ,..., : Fixed loss rate corresponding to the water depth range.

[0038] Modified exponential function: taking into account flooding duration impact.

[0039] ; Duration of flooding (hours); : Correction coefficient for submergence duration.

[0040] Disaster loss assessment module: used to store the asset loss rate generated by the inundation data acting on the corresponding area number, establish the relationship between the asset loss rate and the inundation characteristics, and calculate the direct economic losses caused by the flood to various assets in combination with the asset value.

[0041] The flood loss assessment and analysis is conducted based on the superposition analysis of the flood-affected disaster-bearing bodies. The loss indicators involved include residential housing losses, household property losses, farmland losses, industrial assets and output value losses, commercial assets and main business losses, and road and railway losses. The calculation method is as follows: Housing loss value: Σ(flooded area * building proportion * building unit price * net asset value ratio * loss rate); Household property loss value: Σ(flooded area * building ratio / per capita housing * per capita income * income ratio * loss rate); Agricultural loss value: Σ(flooded area / cultivated land area* agricultural loss / multiple cropping index* loss rate* farmland correction coefficient); Industrial asset loss value: Σ((fixed assets * asset depreciation rate + current assets) * (inundated industrial land area / total industrial land area) * loss rate); Commercial asset loss value: Σ((fixed assets * asset depreciation rate + current assets) * (flooded commercial land area / total commercial land area) * loss rate); Industrial output loss value: Σ(regional industrial output value * conversion coefficient * (inundated industrial land area / total industrial land area) * loss rate); Loss of main business income of the commercial sector: Σ(regional main business income * conversion coefficient * (flooded area of ​​commercial land / total area of ​​commercial land) * loss rate); Road loss value: Σ(affected length * repair cost * loss rate); Railway loss value: Σ(affected length * repair cost * loss rate).

[0042] Visualization Module: Combining the PostGIS spatial database with GeoServer map services, a visualization platform was built using the Vue+OpenLayers front-end framework to provide intuitive and dynamic flood simulation and statistical analysis results. In practice, dynamic inundation simulation was implemented using WebGL technology, with WMTS services published through GeoServer to dynamically load MVT slice data. Web Workers were used to asynchronously analyze gridded water depth data, generating heat maps and isosurfaces to visually display flood inundation conditions.

[0043] According to the water depth levels defined in the loss rate curve, flood data in different water depth ranges are integrated into the same result set. At the same time, the geometric merging algorithm Union is used to merge the geometric shapes that meet the conditions to generate flood inundation areas of different water depth levels.

[0044] This embodiment establishes a sophisticated loss calculation formula for different asset types, comprehensively considering the impact of flood characteristics (such as flow velocity and arrival time) and disaster-prone object properties (such as building structure and land use type) on losses. It also adopts asynchronous queue processing concurrent computing and gRPC efficient communication technology to ensure the accuracy and timeliness of loss assessment results, overcoming the low estimation accuracy problem of existing technologies.

[0045] Based on map data such as administrative boundaries and land use, socioeconomic statistics, calculated or historical (measured or surveyed) flood inundation factor values ​​and their spatial distribution data, on the basis of flood impact analysis, the losses of different types of flooded disaster-prone objects are further evaluated based on parameters such as loss rate curves. According to the relationship with the administrative regions or spatial aggregation statistics, the final output is multi-indicator flood loss results by disaster-prone object type, inundation characteristics, and administrative region, as well as their corresponding spatial distribution.

[0046] This embodiment also includes a spatial analysis module, which includes a geometric merging algorithm, a geometric intersection algorithm, a geometric intersection calculation, a geometric area calculation, a geometric length calculation, and a geometric centroid calculation; a spatial topology check is performed on the gridded layers, land use data, and disaster-prone body distribution data to ensure that there are no topological errors such as unclosed, gaps, overlaps, and self-intersections.

[0047] Geometric intersection algorithms accurately identify and extract the overlapping portions of two geometric regions, generating new geometric shapes. For example, when calculating the overlap between a flooded area and a specific administrative division, point-by-point comparison and line-surface intersection operations are used to determine the intersection boundary, forming a new polygon representing the flooded area within that administrative division. This provides a precise geometric basis for subsequent statistical analysis of inundated areas and other indicators.

[0048] Implementation Principle: The coordinates of two geometric figures are compared point by point to determine their shared boundary. For polygons, a line-surface intersection algorithm is used to calculate the intersection of each edge with the face of the other geometric figure. These intersections are then connected sequentially to form a new polygon, representing the intersection of the two geometric figures. The calculation process takes into account the accuracy of the coordinate points and the topological relationships of the geometric figures to ensure the accuracy of the intersection result.

[0049] The geometry merging algorithm can merge multiple adjacent or overlapping geometric regions into a single, complete geometric shape. When dealing with multiple flood-inundated areas or diverse distributions of hazard-prone objects, the engine automatically identifies connected or intersecting parts by analyzing the boundaries and topological relationships of each geometric object. It then performs a fusion operation, eliminating internal boundaries and generating a unified polygon. This ensures the integrity of the analysis area and facilitates comprehensive flood impact assessment and loss statistics.

[0050] Implementation Principle: Analyze multiple input geometric shapes and identify their adjacency relationships and topological structures. By merging the boundaries of adjacent geometric shapes and eliminating internal boundaries, a new geometric shape is generated. For polygons, this is done by determining whether their edges are adjacent or overlapping, merging adjacent edges into a single edge, and sequentially integrating the vertices of all geometric shapes to form a combined polygon vertex sequence.

[0051] Geometric Difference Algorithm: This algorithm can accurately remove the portion of a geometric region covered by another geometric region from one geometric region. For example, when determining the area of ​​farmland unaffected by flooding within a certain area, the original geometric region of the farmland is subtracted from the geometric region of the flooded area to obtain the remaining uninundated farmland geometry. This allows for precise calculation of the area of ​​unaffected farmland, providing accurate data for agricultural loss assessment.

[0052] Implementation Principle: Based on the subtracted geometry, the subtracted geometry is determined point by point to determine whether it covers its internal area. For polygons, a line-surface intersection algorithm is used to determine the boundary edge between the subtracted and subtracted polygons. The subtracted polygon is then divided into two parts: one for the unsubtracted area and the other for the subtracted area. The unsubtracted area is retained as the difference result.

[0053] Intersection Detection Algorithm: Utilizing rapid scanning and spatial indexing technology, it performs batch intersection analysis on large numbers of geometric shapes. When analyzing the relationship between flood inundation areas and numerous geographic features (such as roads, residential areas, and critical infrastructure), it can quickly and accurately identify features that intersect with the flooded area, effectively targeting them for subsequent impact analysis and loss assessment. This improves analysis efficiency and avoids ineffective processing of irrelevant features.

[0054] How the algorithm works: Spatial indexing technology is used to quickly screen geometric figures and identify potentially intersecting pairs of geometric bodies. Each pair of geometric bodies is then carefully checked to determine whether their boundaries intersect. For line segment intersection detection, the vector cross product is used to calculate the direction vectors of the two line segments to determine whether they are collinear or parallel. If they are neither collinear nor parallel, the intersection coordinates are further calculated to determine whether the line segments intersect.

[0055] Geometric Area Calculation: By comprehensively accounting for factors such as Earth curvature and projection distortion, the engine can perform high-precision area calculations for areas of various complex geometries. Whether it's an irregular flood inundation area or a distribution of different types of hazard-bearing structures, the engine can quickly and accurately calculate their actual area. This provides reliable data support for flood area statistics in flood impact analysis and asset footprint calculations in loss assessment, ensuring the accuracy of analytical results.

[0056] The algorithm works by decomposing a geometric region into simple shapes (such as triangles and trapezoids). The area of ​​each simple shape is calculated separately, and then the sum is used to obtain the area of ​​the entire geometric region. For polygons, a triangulation algorithm is used to divide them into triangles. The area of ​​each triangle is calculated and added together. The area result is corrected for the effects of Earth curvature and projection distortion.

[0057] Geometric Length Calculation: This feature provides precise length measurement for various linear geometric shapes, such as roads, rivers, and pipelines. In flood risk analysis, by measuring the length of linear objects like flooded roads and rivers, combined with corresponding loss calculation models, it is possible to accurately assess the extent of flood damage to transportation, water conservancy, and other infrastructure, providing a crucial reference for post-disaster repair and reconstruction efforts.

[0058] Implementation Principle: Different length calculation methods are used depending on the type of geometric figure. For straight line segments, the Euclidean distance between two points is directly calculated. For polyline segments, the polyline is decomposed into multiple straight line segments, and the distance of each segment is calculated separately and summed. For curves, parametric equations or numerical integration methods are used for approximate calculations. Distance conversion in geographic coordinate systems is also considered, and latitude and longitude are converted to actual length units.

[0059] Geometric Centroid Calculation: Automatically calculates and determines the geometric center of a geometric area based on its shape and distribution. For flood-inundated areas and hazard-prone areas, determining the center point helps quickly locate the core of the area, facilitating spatial analysis and visualization. It also serves as a starting point for other analytical functions, such as buffer analysis and distance attenuation analysis.

[0060] Implementation Principle: For regular geometric shapes (such as rectangles and circles), the center coordinates are directly calculated based on their geometric properties. For irregular geometric shapes, the geometric area is decomposed into multiple simple geometric shapes, the center point of each simple shape is calculated, and then the center coordinates of the entire shape are calculated based on the area or mass distribution of the geometric body. For polygons, the center point is determined by using a centroid calculation algorithm, either through integration or weighted averaging of discrete points.

[0061] Cluster analysis algorithm: Density-based spatial clustering application is used to identify the clustering structure of flood risk points and disaster-stricken areas, help discover concentrated areas and diffusion trends of flood impacts, and provide macro-guidance for flood control planning and emergency response.

[0062] How it works: Clustering is performed based on density by setting a radius ε and a minimum number of points, MinPts. First, the number of points in each point's neighborhood is calculated. If the number of points in the neighborhood is greater than or equal to MinPts, the point is used as the core point and the cluster is expanded. This process is repeated until all reachable points are included in the cluster.

[0063] Convex hull calculation: It can generate the minimum convex polygonal bounding box for any complex geometric shape. In flood risk analysis, generating a convex hull for multiple discrete flood risk points or flooded areas can visually demonstrate the overall distribution and morphological characteristics of the risk area, helping to identify the concentrated areas and diffusion directions of flood risk, and providing macro-level guidance for flood control planning and layout.

[0064] Implementation Principle: Implementing a convex hull algorithm, using the Graham scan method as an example, first finds the bottom-leftmost point in the set of geometric points. Then, points are selected in a counterclockwise direction, ensuring that the cross product of the vectors formed by each newly selected point and the first two points remains consistent (i.e., always positive or negative). This generates a convex polygon containing all points. The algorithm gradually constructs the convex hull of the set of geometric points by continuously adjusting and filtering points.

[0065] In practice, flood data from different water depths are integrated into a single result set based on the water depth levels defined in the loss rate curve. The geometric union algorithm, Union, is then used to merge the geometric shapes that meet the conditions, generating flood inundation areas at different water depth levels.

[0066] Based on the spatial intersection judgment algorithm Intersects, flood data are associated with the administrative boundary table to ensure that only flood data and administrative boundary data that have intersections in spatial locations will be included in the subsequent analysis to determine the scope of administrative divisions inundated by floods.

[0067] The loss rate information of flood data is filtered, and the flood inundation data corresponding to different flood depth levels are queried separately to achieve a detailed analysis of the relationship between the inundation range and administrative boundaries under different flood depth scenarios, providing a data basis for subsequent flood risk assessments at different depths.

[0068] The geometric intersection algorithm, Intersection, calculates the intersection of the flood data geometry and the administrative boundary geometry to obtain the specific geometry of the flood inundation area within the administrative division. The flood-affected area is determined, and the actual inundation area boundaries of the flood are calculated within each administrative division.

[0069] Using the geometric area algorithm, Area, we extract the area of ​​the administrative divisions themselves and the area of ​​their intersections, and calculate the ratio of the intersection area to the administrative division area. This area data provides the basis for subsequent calculations of the ratio of the flood-inundated area to the administrative division area, quantifying the impact of flooding on each administrative division. By analyzing the spatial intersection of the flood-affected area and surface-type hazard-prone features, we calculate the area, number, length, and ratio of the intersections, providing data for flood risk maps reflecting the impact of floods on different hazard-prone features.

[0070] Spatial data association: associate the administrative boundary analysis table with the surface type disaster-prone land features, and use the spatial intersection judgment algorithm Intersects to screen out data pairs that have intersections in spatial positions.

[0071] Conditional filtering: Use the maximum water depth and minimum water depth contained in the loss rate curve to filter the disaster-prone ground feature data, and perform query filtering for different depth levels.

[0072] Spatial intersection operation: The geometric intersection algorithm Intersection is used to calculate the intersection of the geometric shape of the administrative boundary and the geometric shape of the disaster-affected landform, and the geometric shape of the flood inundation range within the disaster-affected landform unit is obtained.

[0073] At the same time, this embodiment is based on the spatial analysis module to extract the number of point-type disaster-prone areas, the area of ​​surface-type disaster-prone features, and the length of line-type disaster-prone features. The number of point-type disaster-prone areas is used to obtain the number of disaster-prone features that meet the spatial intersection condition, and the ratio of the number of disaster-prone feature units to the number of intersection parts is calculated to quantify the impact of floods on disaster-prone features. The area of ​​surface-type disaster-prone features is extracted using the geometric area algorithm "Area", and the ratio of the area of ​​the disaster-prone feature unit to the area of ​​the intersection part is calculated to quantify the impact of floods on disaster-prone features. The length of line-type disaster-prone features is obtained by extracting the length of the disaster-prone feature unit using the geometric length algorithm "Length", and the ratio of the length of the disaster-prone feature unit to the length of the intersection part is calculated to quantify the impact of floods on disaster-prone features.

[0074] Based on the spatial analysis module, the inundation layer is superimposed with the administrative district layer, the residential map layer, and the cultivated land surface layer to obtain the flooded administrative district area, flooded residential area, and flooded cultivated land area corresponding to different flooding depth levels in different scenarios, and the areas are displayed in the visualization module.

[0075] A disaster damage assessment method comprises the following steps: Step 1: Collect inundation data, geographic data, and asset data of the hazard-prone body; and clean, convert, and standardize the collected raw data; Step 2: derive flood characteristics, geographic characteristics, and asset characteristics respectively from the inundation data, geographic data, and asset data in the historical data; The geographical feature is the regional number, and the specific processing method is as follows: for the disaster-prone body, the disaster-prone body is gridded based on the distribution of administrative regions, and the gridded areas are numbered; Inundation characteristics include inundation depth, inundation duration, arrival time and maximum flow velocity corresponding to the area number; The asset characteristics include asset data distributed over corresponding area numbers; Step three: Cluster analysis is performed on the historical asset data and flooding data for each area number, generating a flooding dataset and an asset loss rate dataset for that area number. Based on the loss rate function assigned to the corresponding asset, the loss rate function is trained using the flooding dataset and the asset loss rate dataset to determine the asset loss rate caused by the flooding data for that area number. This results in an asset loss rate assessment model. The loss rate function uses linear, exponential, logarithmic, step, and modified exponential functions. Using the selected function form, combined with the organized flooding dataset and asset loss rate dataset, a statistical analysis method is used to fit the loss rate curve. By adjusting the function parameters, the error between the fitted curve and the actual data points is minimized, ultimately resulting in a loss rate curve equation for each asset type. The statistical analysis method used is either the least squares method or maximum likelihood estimation.

[0076] Step 4: Extract and process the real-time inundation data, geographic data, and asset data in the same manner as in Steps 1 and 2 to obtain inundation characteristics and asset characteristics under the corresponding area number. Input the inundation characteristics and asset characteristics into the asset loss rate assessment model to obtain the losses of various assets. Step 5: Add up the losses of various assets to obtain the total asset losses.

[0077] This embodiment integrates meteorological, hydrological, geographic, and asset data through a multi-source data fusion module, which, after cleaning and conversion, forms a standardized data foundation. The data analysis module uses gridded spatial correlation technology to extract dynamic features such as flooding depth and duration, and establishes a multidimensional feature library based on the attributes of the disaster-prone body. The loss rate assessment module innovatively uses linear, exponential, logarithmic, step, and modified exponential function families to generate refined loss rate curves through cluster analysis and parameter fitting, significantly improving the accuracy of loss estimates. The disaster loss assessment module incorporates spatial analysis algorithms to quantify the economic losses of various types of assets at different water depth levels, covering all elements such as housing, agriculture, industry, commerce, and infrastructure. The visualization platform uses WebGL and GeoServer to implement dynamic simulation and heat map display. The spatial analysis module supports precise spatial topology analysis through algorithms such as geometric merging and intersection calculation. This method ensures real-time computing efficiency through asynchronous queue processing and gRPC communication technology, ultimately outputting multidimensional loss assessment results by type, feature, and region, providing precise spatial and temporal quantitative support for flood control decision-making.

[0078] The above-described embodiments of the present invention do not constitute a limitation on the scope of protection of the present invention. The basic concept of the present invention is to build a standardized data foundation by integrating meteorological, hydrological, geographical and asset data; to use gridded spatial correlation technology to extract dynamic characteristics such as flooding depth and duration, and to establish a multi-dimensional feature library in combination with the attributes of the disaster-prone body; to innovatively use linear, exponential, logarithmic, step and modified exponential function families to generate refined loss rate curves through cluster analysis, significantly improving the accuracy of loss estimation. Combined with spatial analysis algorithms, quantify the losses of various types of assets under different water depth levels, covering all elements of housing, agriculture, industry, commerce and infrastructure, and intuitively display the evolution of floods and the distribution of losses through a dynamic visualization platform. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. A disaster damage assessment device, characterized by: include Data collection and pre-processing module: used to collect flooding data, geographic data and asset data of the disaster-prone body; and to clean, convert and standardize the collected raw data; Data analysis module: Utilizes the flood data, geographic data and asset data in historical data to obtain flood characteristics, geographic characteristics and asset characteristics respectively. The geographical features include regional numbers. For disaster-prone bodies, the disaster-prone bodies are gridded based on the distribution of administrative regions, and the gridded areas are numbered. Inundation characteristics include inundation depth, inundation duration, arrival time and maximum flow velocity corresponding to the area number; The asset characteristics include asset data distributed over corresponding area numbers; Loss rate assessment module: Based on the area number, cluster analysis is performed on the historical asset data and inundation data for the area number. According to the distribution characteristics of the sorted data, the corresponding loss rate function is matched to fit the loss rate curve. Based on the loss rate curve, the asset loss rate of different types of flooded and hazard-bearing objects is assessed; Disaster damage assessment module: used to store the asset loss rate generated by the inundation data acting on the corresponding area number, establish the relationship between the asset loss rate and the inundation characteristics, and calculate the direct economic losses caused by the flood to various assets based on the asset value; Visualization module: Combining the PostGIS spatial database and GeoServer map service, the Vue+OpenLayers front-end framework is used to build a visualization platform to provide intuitive and dynamic flood simulation and statistical analysis results.

2. The damage assessment device according to claim 1, wherein: The loss rate function includes linear function, exponential function, logarithmic function, step function and modified exponential function; the linear function is used for linear loss relationship in low water depth range; the exponential function is used for rapidly rising loss rate in high water depth range; the piecewise function is used for nonlinear changes in different water depth ranges; the logarithmic function is used for scenarios where the loss rate slowly increases with water depth; The step function is used for scenarios with sudden changes in loss rate within the water depth range, and the modified exponential function is used to evaluate the impact of the flooding duration t.

3. The damage assessment device according to claim 1, wherein: According to the water depth levels defined in the loss rate curve, flood data in different water depth ranges are integrated into the same result set.

4. The damage assessment device according to claim 1, wherein: It also includes a spatial analysis module, which includes a geometric merging algorithm, a geometric intersection algorithm, a geometric intersection calculation, a geometric area calculation, a geometric length calculation, and a geometric centroid calculation; it performs spatial topology checks on gridded layers, land use data, and disaster-prone body distribution data to ensure that there are no topological errors such as unclosed, gaps, overlaps, and self-intersections.

5. The damage assessment device according to claim 4, characterized in that: Based on the spatial analysis module, the number of point-type disaster-prone areas, the area of ​​surface-type disaster-prone features, and the length of line-type disaster-prone features are extracted. The number of point-type disaster-prone areas is used to obtain the number of disaster-prone features that meet the spatial intersection condition, and the ratio of the number of disaster-prone feature units to the number of intersection parts is calculated to quantify the impact of floods on disaster-prone features. The area of ​​surface-type disaster-prone features is extracted using the geometric area algorithm "Area", and the ratio of the area of ​​the disaster-prone feature unit to the area of ​​the intersection is calculated to quantify the impact of floods on the disaster-prone features. The length of line-type disaster-prone features is obtained by extracting the length of the disaster-prone feature unit using the geometric length algorithm "Length", and the ratio of the length of the disaster-prone feature unit to the length of the intersection is calculated to quantify the impact of floods on the disaster-prone features.

6. The damage assessment device according to claim 4, wherein: Based on the spatial analysis module, the inundation layer is overlaid with the administrative district layer, the residential map layer, and the cultivated land surface layer to obtain the flooded administrative district area, flooded residential area, and flooded cultivated land area corresponding to different flooding depth levels in different scenarios, and the areas are displayed in the visualization module.

7. The disaster damage assessment device according to claim 1, characterized in that: Asset data include household property, household housing, agricultural losses, industrial losses, industrial output value, commercial assets, commercial business income, railways, provincial roads and above, and roads below provincial roads; for each type of asset, the loss rate under different flooding characteristics is compiled separately.

8. A disaster damage assessment method, characterized by: The steps include: Step 1: Collect inundation data, geographic data, and asset data of the hazard-prone body; and clean, convert, and standardize the collected raw data; Step 2: derive flood characteristics, geographic characteristics, and asset characteristics respectively from the inundation data, geographic data, and asset data in the historical data; The geographical feature is the regional number, and the specific processing method is as follows: for the disaster-prone body, the disaster-prone body is gridded based on the distribution of administrative regions, and the gridded areas are numbered; Inundation characteristics include inundation depth, inundation duration, arrival time and maximum flow velocity corresponding to the area number; The asset characteristics include asset data distributed over corresponding area numbers; Step 3: Perform cluster analysis on the historical asset data and flooding data for each area number to generate a flooding dataset and an asset loss rate dataset based on the area number. Based on the loss rate function set for the corresponding asset, the loss rate function is trained using the flooding dataset and the asset loss rate dataset to obtain the asset loss rate caused by the flooding data for each area number. This results in an asset loss rate assessment model. Step 4: Extract and process the real-time inundation data, geographic data, and asset data in the same manner as in Steps 1 and 2 to obtain inundation characteristics and asset characteristics under the corresponding area number. Input the inundation characteristics and asset characteristics into the asset loss rate assessment model to obtain the losses of various assets. Step 5: Add up the losses of various assets to obtain the total asset losses.

9. The disaster damage assessment method according to claim 6, wherein: The loss rate function includes linear function, exponential function, logarithmic function, step function and modified exponential function. The selected function form is combined with the sorted flooding data set and asset loss rate data set, and the loss rate curve is fitted by adopting statistical analysis method. By adjusting the parameters of the function, the error between the fitting curve and the actual data points is minimized, and finally the loss rate curve equation for each type of asset is obtained.

10. The disaster damage assessment method according to claim 9, characterized in that: The statistical analysis method is least squares method or maximum likelihood estimation.

Images