Multi-source data and random distribution coupled building disaster damage assessment uncertainty optimization method

Abstract

The invention discloses a multi-source data and random distribution coupled building disaster damage assessment uncertainty optimization method. The method comprises the following steps: collecting historical flood disaster data of different cities and different types of buildings; meshing division is carried out on the evaluation area; simulating and generating high-resolution water depth data of each grid unit by using a HiPIMS model, and verifying and generating disaster damage data of each grid in combination with remote sensing data; generalizing the cost of different types of buildings by using the building cost data in the urban statistical yearbook; evaluating the flood vulnerability of the building based on a gridding method; and quantifying the evaluation uncertainty of the building flood vulnerability by adopting a confidence level analysis method. According to the method, the problems of multi-source data fusion and building accurate classification are solved, the scale difference between flood characteristics and social economic data is unified, the confidence level analysis method is introduced to quantify uncertainty, the evaluation precision and reliability are remarkably improved, and scientific support is provided for urban disaster prevention and reduction, risk management and planning.

Description

Technical Field

[0001] The invention relates to the technical field of urban building disaster damage assessment, and in particular to an uncertainty optimization method for building disaster damage assessment coupled with multi-source data and random distribution. Background Art

[0002] The existing urban building damage assessment technology has the following problems: first, the data source is single and lacks multi-source data fusion, resulting in inaccurate building classification; second, the loss rate estimation method is too simple and does not take into account the uncertainty of the water depth-loss rate curve, making it difficult to accurately reflect the actual loss; third, the spatial resolution is insufficient and the spatial distribution of building damage cannot be accurately quantified; in addition, the application of high-performance hydrodynamic models is insufficient, the calculation efficiency is low, and it is difficult to obtain accurate real-time flooding depth information; at the same time, the assessment is highly dependent on historical disaster data, which limits the applicability of the method in cities with scarce data; finally, the building classification method is simple and does not fully consider the difference in losses of buildings with different functions. In addition, the model lacks a reliable verification mechanism, and the reliability and applicability of the assessment results are difficult to guarantee. These problems restrict the accuracy of damage assessment and decision-making support capabilities, and are in urgent need of optimization and improvement. Summary of the invention

[0003] In view of the above technical problems in the related art, the present invention provides an uncertainty optimization method for building damage assessment coupling multi-source data with random distribution, which can solve the above problems.

[0004] To achieve the above technical objectives, the technical solution of the present invention is implemented as follows:

[0005] An uncertainty optimization method for building damage assessment coupling multi-source data and random distribution includes the following steps:

[0006] S100. Collect historical flood data of different types of buildings in different cities, including water depth, flood loss rate and building attribute characteristics, and clean, deduplicate and standardize the data. In the process of collecting building damage data, combine electronic map data, POI data, remote sensing data and socio-economic survey data, and use multi-source data fusion technology to classify building functions;

[0007] S200, dividing the region into multi-level spatial grids;

[0008] S300, using the HiPIMS model to simulate and obtain the water depth data of each grid unit, and verify and generate the disaster loss data of each grid in combination with remote sensing data;

[0009] S400, use the construction cost data of different types of buildings in the city statistical yearbook to estimate the building cost;

[0010] S500. Building flood vulnerability assessment based on the grid method;

[0011] S510. Determine the loss rate function of each type of building at different water depths;

[0012] S520. Estimate the cost value of each type of building and estimate the proportion of the area of each type of building in the grid;

[0013] S530. Statistically calculate the total loss of buildings in the grid and estimate the flood loss of the entire area;

[0014] S600. Use the confidence level analysis method to quantify the uncertainty of building flood vulnerability and improve the reliability of disaster loss assessment.

[0015] Furthermore, the attribute characteristic information of the buildings in S100 includes the location, type, area, and storey height of the buildings.

[0016] Furthermore, the specific steps for building function classification in S100 are as follows:

[0017] S110. Use high-resolution remote sensing images and GIS tools, combine image recognition algorithms and visual interpretation methods to extract the buildings in the selected area, and assign a unique identification number to each building for spatial distribution analysis;

[0018] S120. Classify based on the building attribute information of the electronic map;

[0019] S130. Screen and remove duplicates from the collected POI data. According to the land use type and function classification standard, use the POI data frequency density analysis method to assign preliminary function attributes to the buildings whose functions cannot be determined through electronic map labels;

[0020] S140. For the buildings that cannot be classified above, use the kernel density estimation method for function classification. The kernel density estimation method uses a Gaussian distribution kernel function to smooth the POI data around the building and estimate the function distribution probability of the building.

[0021] Furthermore, the kernel function for building function classification in S140 is a two-dimensional Gaussian distribution, and the expression is as follows:

[0022]

[0023] In the formula, the coefficient A represents the Gaussian function amplitude, x 0 , y 0 represent the position coordinates of the center point, and σ x , σ y represent the variances in the x and y directions respectively, which affect the search / diffusion radius;

[0024] After determining the kernel function, four kernel density layers are constructed based on the four known classifications of buildings. For unknown classification buildings, the sum of the kernel density layers in the area of ​​the unknown classification building is calculated. The unknown classification building is classified into the classification category with the highest density value. The mathematical expression is as follows:

[0025]

[0026]

[0027]

[0028]

[0029] In the formula, K 1 , K 2 , K 3 , K 4 Represents the kernel density value of each type of building, N b represents the number of pixels in the building area, K 1i , K 2i , K 3i , K 4i Represents the kernel density value of the corresponding pixel on the kernel density layer of each type of building.

[0030] Furthermore, in S400, the cost per unit area of ​​each type of building is calculated by the following formula:

[0031]

[0032] Where θ i，j N represents the unit area value cost of the i-th type of building in the j-th administrative area. i，j A represents the number of pixels of the i-th type of building in the j-th administrative area, k Represents the area of ​​each pixel, G j Represents the cost value of the j-th administrative area.

[0033] Furthermore, in S510, the discrete water depth-loss rate curve points are fitted by using the curve fitting toolbox of Matlab, and the most suitable fitting function is selected according to the curve characteristics of different types of buildings.

[0034] Furthermore, assuming that the building loss rate follows a Gaussian distribution, it is estimated based on the maximum gap of the water depth-loss rate curve of similar buildings at the same water depth.

[0035] Furthermore, the Monte Carlo simulation method was used to simulate the possible loss rate distribution under each water depth condition through a large number of experiments, and then the loss rate value with the highest probability was selected as the final estimate through the maximum probability value method.

[0036] Furthermore, the total flood loss of buildings in each grid in S530 is calculated by the following formula:

[0037]

[0038] Where C b represents the estimated total flood loss of buildings in the grid, i.e., the grid flood vulnerability, θ i represents the value cost per unit area of ​​the i-th type of building, f i (d) represents the loss rate of the i-th type of building at water depth h, N i represents the number of pixels occupied by the i-th type of building, A k Represents the area of ​​each pixel.

[0039] Furthermore, S600 specifically includes the following steps:

[0040] S610, calculating the building vulnerability index according to the loss value and water depth data of each building under different flood disaster scenarios;

[0041] S620. Conduct confidence level analysis on the vulnerability index of each building through Monte Carlo simulation and confidence interval calculation;

[0042] S630. Combine the loss data and vulnerability indicators of different types of buildings to quantify the losses of urban buildings.

[0043] Beneficial effects of the present invention: This application comprehensively utilizes multi-source data to accurately classify building functions, greatly improving the accuracy of disaster damage assessment; unifies flood characteristics and socioeconomic data through a gridding method, solves the problem of inconsistent data at different scales, and achieves accurate assessment of building damage. In the analysis of urban flood building vulnerability, the Matlab curve fitting tool is used to generate a water depth-loss rate mean curve, and the Monte Carlo simulation is used to simulate the loss rate distribution under different water depths, and the maximum probability value method is used to determine the final estimation result. At the same time, combined with urban grid division, histogram statistical analysis and confidence level analysis of building flood vulnerability, the uncertainty in the disaster damage assessment process is effectively reduced, further improving the scientificity and applicability of flood damage assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] The present invention is further described in detail below with reference to the accompanying drawings.

[0046] Figure 1 It is a flow chart of an uncertainty optimization method for building damage assessment coupled with multi-source data and random distribution according to an embodiment of the present invention;

[0047] Figure 2 It is a principle diagram of using a kernel density estimation method to perform functional classification on buildings that cannot be classified according to an embodiment of the present invention;

[0048] Figure 3 It is a schematic diagram of conducting building flood vulnerability assessment based on a gridding method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0050] like Figure 1-3As shown, according to the present invention, an uncertainty optimization method for building disaster damage assessment coupling multi-source data and random distribution is disclosed, comprising the following steps: S100, collecting historical flood data of different types of buildings in different cities, including water depth, flood loss rate and attribute characteristic information of buildings, and cleaning, de-duplication and standardization of the data. In the process of collecting building disaster damage data, multi-source data fusion technology is used to classify building functions by combining electronic map data, POI data, remote sensing data and socio-economic survey data; S200, multi-level grid division of the region; S300, using HiPIMS model to simulate and obtain the water depth of each grid unit; S400, use the construction cost data of different types of buildings in the city statistical yearbook to estimate the building cost; S500, conduct flood vulnerability assessment of buildings based on the grid method; S510, determine the loss rate function of each type of building under different water depths; S520, estimate the cost value of each type of building and the area proportion of each type of building in the grid; S530, after counting the total loss of buildings in the grid, estimate the flood loss of the entire area; S600, use the confidence level analysis method to quantify the uncertainty of building flood vulnerability and improve the reliability of disaster loss assessment.

[0051] Embodiment 1:

[0052] When using multi-source data fusion technology to classify building functions, the specific technical means include the following:

[0053] (1) Spatial distribution information extraction

[0054] Using 0.5-meter resolution remote sensing images and GIS tools, combined with convolutional neural networks (CNN) and visual interpretation, we automatically extracted buildings in the disaster-damaged area, and corrected misjudgments and omissions through manual verification to ensure the accuracy and completeness of the results. Each building was assigned a unique number to facilitate spatial distribution analysis.

[0055] (2) Classification by electronic map labeling

[0056] Using web crawlers and map data parsing tools, building attribute labels are extracted from electronic maps and classified according to preset classification rules and keyword libraries.

[0057] (3) POI data classification

[0058] The collected POI data are screened and deduplicated, and corresponding functions are assigned to buildings whose functions cannot be determined through electronic map labels based on land use type and functional classification standards and the number of surrounding POI data.

[0059] (4) Kernel density estimation classification

[0060] For buildings that cannot be classified, the kernel density estimation method (KDE) can be used to perform functional classification based on the surrounding POI data. The Gaussian kernel function is used to smooth local features and expand to the global. The POI density is calculated by setting an appropriate diffusion radius to infer the function of the building, such as densely distributed residential POIs in residential areas and concentrated industrial POIs in industrial areas, which improves classification accuracy and reduces data uncertainty.

[0061] The principle formula of the kernel density estimation method is as follows:

[0062]

[0063] In the formula, K(x) represents the kernel function, which is a non-negative function, R represents the neighborhood radius or search radius, and x i N is the element whose attributes are known. k Represents the number of known features in the kernel density estimate.

[0064] According to the above conditions, the kernel function of building function classification is set to a two-dimensional Gaussian distribution, and the expression is as follows:

[0065]

[0066] In the formula, the coefficient A represents the amplitude of the Gaussian function, x 0 ,y 0 represents the position coordinates of the center point, σ x , σ y Represents the variance in the x and y directions, respectively, which affects the search / diffusion radius.

[0067] Four kernel density layers for residential, commercial, industrial and public buildings are constructed based on the kernel function, and machine learning algorithms (decision trees) are used to classify the functions of buildings. K-means clustering is used to optimize data selection and sample expansion to ensure that more building types and functional attributes are covered. For unknown classifications, the sum of each kernel density layer within the building area is classified into the category with the largest sum. The mathematical expression is as follows:

[0068]

[0069]

[0070]

[0071]

[0072] In the formula, K 1 , K 2 , K 3 , K 4Represents the kernel density value of each type of building, N b represents the number of pixels in the building area, K 1i , K 2i , K 3i , K 4i Represents the kernel density value of the corresponding pixel on the kernel density layer of each type of building.

[0073] In order to solve the problem of inconsistent scales between flood characteristic grids and socio-economic data, a gridding method was used to divide the region into multi-level regular grids to unify different types of spatial data (flood characteristics, socio-economic attributes) for comprehensive analysis. For flood characteristic data, the water depth data of each grid cell was obtained by using the HiPIMS model (high-resolution flood simulation model) and multi-source remote sensing data.

[0074] Embodiment 2:

[0075] When screening vulnerability samples, in order to reduce the error in estimating building flood losses, according to the central limit theorem, cities with similar scale and economic development level should be selected. Using the construction cost data of different types of buildings in the city statistical yearbook, the building cost can be generalized. The accuracy of building cost estimation can be improved by combining big data technology with other relevant information. The cost value per unit area of ​​various types of buildings is calculated according to the following formula:

[0076]

[0077] Where θ i，j N represents the unit area value cost of the i-th type of building in the j-th administrative area, i，j G represents the number of pixels of the i-th type of building in the j-th administrative area. j Represents the cost value of the j-th administrative area.

[0078] When conducting a flood vulnerability assessment of buildings based on a grid-based approach, the area is divided into regular grids (squares or rectangles), and each grid integrates information such as building type, area, cost value, and water depth. The loss rate of the building is calculated using the water depth-loss rate curve, and the loss value is calculated by combining the cost value and area weights. Finally, the loss values ​​of each grid are accumulated to estimate the flood loss of the entire area.

[0079] The Matlab curve fitting tool is used to fit the discrete water depth-loss rate data points of different building types, and the appropriate function is selected to reduce the influence of the discreteness of the curve, providing an accurate basis for the loss rate estimation.

[0080] Furthermore, due to the uncertainty of the water depth-loss rate curve, it is assumed that the building loss rate follows a Gaussian distribution at a specific water depth, centered on the vulnerability mean, and the variance is estimated based on the maximum gap of the curve. Monte Carlo simulation is used to simulate the loss rate, and the distribution law of all possible loss rate values ​​is analyzed through histogram statistics to generate a probability density distribution, and the value with the highest probability is selected as the final estimate.

[0081] Determine the building loss rate function under different water depths, and combine the cost value of the building and the area ratio within the grid to calculate the total loss of the building in each grid. Based on the minimum spatial scale grid unit, the total regional loss is accumulated through the grid tool. According to the loss rate and unit area cost value of each type of building derived and calculated, the total flood loss of the building in each grid is calculated according to the following formula:

[0082]

[0083] Where C b represents the estimated total flood loss of buildings in the grid, i.e., the grid flood vulnerability, Ω i represents the value cost per unit area of ​​the i-th type of building, f i (h) represents the loss rate of the i-th type of building at water depth h, ψ i Represents the area ratio of the i-th type of building in the grid.

[0084] In order to facilitate the calculation of grid loss, the formula is optimized as follows:

[0085]

[0086] Where θ i represents the value cost per unit area of ​​the i-th type of building, f i (d) represents the loss rate of the i-th type of building at water depth h, N i represents the number of pixels occupied by the i-th type of building, A k Represents the area of ​​each pixel.

[0087] In order to further improve the reliability of disaster loss assessment, the present invention adopts confidence level analysis to quantify the uncertainty of building flood vulnerability. The specific steps are as follows: Generate loss estimates through Monte Carlo simulation, and construct a probability density function in combination with histogram statistics to determine the confidence level within a specific confidence interval. The specific steps include determining the random event simulation method, selecting a confidence interval (95%), statistically distributing the results and calculating the confidence level, and finally obtaining the threshold range (maximum and minimum values) of the loss. This method can quantify the uncertainty of the model, intuitively describe the reliability of loss estimates, provide a scientific basis for disaster prevention and mitigation policies, and is applicable to different building types and water depth conditions to achieve flexible and accurate risk prediction.

[0088] In summary, the uncertainty of urban disaster loss assessment is reduced and its accuracy and applicability are improved through urban area grid division, building flood loss rate estimation based on histogram statistics and confidence level analysis of building flood vulnerability.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An uncertainty optimization method for building damage assessment coupling multi-source data and random distribution, characterized in that: The steps include: S100. Collect historical flood data of different types of buildings in different cities, including water depth, flood loss rate and building attribute characteristics, and clean, deduplicate and standardize the data. In the process of collecting building damage data, combine electronic map data, POI data, remote sensing data and socio-economic survey data, and use multi-source data fusion technology to classify building functions; S200, dividing the region into multi-level spatial grids; S300, using the HiPIMS model to simulate and obtain the water depth data of each grid unit, and verify and generate the disaster loss data of each grid in combination with remote sensing data; S400, use the construction cost data of different types of buildings in the city statistical yearbook to estimate the building cost; S500, Flood vulnerability assessment of buildings based on grid-based approach; S510, determining the loss rate function of each type of building at different water depths; S520, estimating the cost value of each type of building and estimating the area proportion of each type of building in the grid; S530, calculate the total loss of buildings in the grid and estimate the flood loss in the entire area; S600. Use confidence level analysis to quantify the uncertainty of building flood vulnerability and improve the reliability of damage assessment.

2. The uncertainty optimization method for building damage assessment based on multi-source data coupled with random distribution according to claim 1 is characterized in that: The property feature information of the building in S100 includes the location, type, area and floor height of the building.

3. The uncertainty optimization method for building damage assessment based on multi-source data coupled with random distribution according to claim 1 is characterized in that: The building function classification in S100 specifically includes the following steps: S110. Use high-resolution remote sensing images and GIS tools, combined with image recognition algorithms and visual interpretation methods, to extract buildings in the selected area and assign a unique identification number to each building for spatial distribution analysis; S120, classifying the building attribute information based on the electronic map; S130, screening and deduplication processing of the collected POI data, and using the POI data frequency density analysis method according to the land use type and functional classification standards to assign preliminary functional attributes to buildings whose functions cannot be determined through electronic map labels; S140. For the buildings that cannot be classified as above, a kernel density estimation method is used for functional classification. The kernel density estimation method uses a Gaussian distribution kernel function to smooth the POI data around the building to estimate the functional distribution probability of the building.

4. The uncertainty optimization method for building damage assessment based on coupling of multi-source data and random distribution according to claim 2 is characterized in that: The kernel function of building function classification in S140 is a two-dimensional Gaussian distribution, expressed as follows: In the formula, the coefficient A represents the amplitude of the Gaussian function, x0 and y0 represent the position coordinates of the center point, and σ x , σ y Represents the variance in the x and y directions, respectively, which affects the search / diffusion radius; After determining the kernel function, four kernel density layers are constructed based on the four known classifications of buildings. For unknown classification buildings, the sum of the kernel density layers in the area of ​​the unknown classification building is calculated. The unknown classification building is classified into the classification category with the highest density value. The mathematical expression is as follows: In the formula, K1, K2, K3, K4 represent the kernel density values ​​of various types of buildings, N b represents the number of pixels in the building area, K 1i , K 2i , K 3i , K 4i Represents the kernel density value of the corresponding pixel on the kernel density layer of each type of building.

5. The uncertainty optimization method for building damage assessment by coupling multi-source data with random distribution according to claim 1 is characterized in that: In S400, the cost per unit area of ​​each type of building is calculated using the following formula: Where θ i，j N represents the unit area value cost of the i-th type of building in the j-th administrative area, i，j A represents the number of pixels of the i-th type of building in the j-th administrative area, k Represents the area of ​​each pixel, G j Represents the cost value of the j-th administrative area.

6. The uncertainty optimization method for building damage assessment by coupling multi-source data with random distribution according to claim 1 is characterized in that: In S510, the curve fitting toolbox of Matlab is used to fit the discrete water depth-loss rate curve points, and the most suitable fitting function is selected according to the curve characteristics of different types of buildings.

7. The uncertainty optimization method for building damage assessment by coupling multi-source data with random distribution according to claim 6 is characterized in that: Assuming that the building loss rate follows a Gaussian distribution, it is estimated based on the maximum gap of the water depth-loss rate curve of similar buildings at the same water depth.

8. The uncertainty optimization method for building damage assessment by coupling multi-source data with random distribution according to claim 7 is characterized in that: The Monte Carlo simulation method is used to simulate the possible loss rate distribution under each water depth condition through a large number of experiments, and then the loss rate value with the highest probability is selected as the final estimate through the maximum probability value method.

9. The uncertainty optimization method for building damage assessment by coupling multi-source data with random distribution according to claim 1 is characterized in that: The total flood damage to buildings in each grid in S530 is calculated using the following formula: Where C b represents the estimated total flood loss of buildings in the grid, i.e., the grid flood vulnerability, θ i represents the value cost per unit area of ​​the i-th type of building, f i (d) represents the loss rate of the i-th type of building at water depth h, N i represents the number of pixels occupied by the i-th type of building, A k Represents the area of ​​each pixel.

10. The uncertainty optimization method for building damage assessment by coupling multi-source data with random distribution according to claim 1 is characterized in that: S600 specifically includes the following steps: S610, calculating the building vulnerability index according to the loss value and water depth data of each building under different flood disaster scenarios; S620, conduct confidence level analysis on the vulnerability index of each building through Monte Carlo simulation and confidence interval calculation; S630. Combine the loss data and vulnerability indicators of different types of buildings to quantify the losses of urban buildings.

Images