A meteorological disaster risk assessment method and system based on the Internet of Things

Abstract

The present invention discloses a meteorological disaster risk assessment method and system based on the Internet of Things, which relates to the technical field of meteorological disaster risk assessment. The method comprises the following steps: obtaining geographic location data of the area to be assessed, determining elevation and historical meteorological environment data, and generating a continuous elevation surface and a meteorological environment surface; constructing a risk assessment coordinate system, and combining elevation and meteorological data to determine a characteristic coefficient matrix for each coordinate point; dividing the area into n sub-areas to be assessed based on this; obtaining data on the natural disaster resistance characteristics and disaster resistance infrastructure of each sub-area, and determining an initial disaster resistance coefficient; selecting the first sub-area in sequence, and determining a homogeneity coefficient by combining adjacent sub-areas; obtaining meteorological forecast data, and combining the homogeneity coefficient and the initial disaster resistance coefficient to determine the risk sub-coefficient and the regional total risk coefficient. The system includes acquisition, data management, risk assessment, early warning, and display modules, and achieves accurate assessment and early warning through multi-dimensional data processing and analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological disaster risk assessment, and in particular to a meteorological disaster risk assessment method and system based on the Internet of Things. Background Art

[0002] Meteorological disasters (such as heavy rain, high winds, and high temperatures) are characterized by suddenness, widespread impact, and extensive damage, posing a serious threat to the ecological environment, infrastructure, and property safety. As global climate change intensifies, the frequency of extreme weather events is increasing year by year. Accurate meteorological disaster risk assessment has become a core component of disaster prevention and mitigation systems. Scientifically assessing regional meteorological disaster risks can help formulate response strategies, optimize resource allocation, and effectively reduce disaster losses. Therefore, developing efficient and intelligent meteorological disaster risk assessment methods and systems is of great practical significance.

[0003] The existing meteorological disaster risk assessment methods and systems based on the Internet of Things have insufficient data fusion capabilities and mostly rely on single-type sensor data. The integration of multi-dimensional information such as elevation, terrain, and historical disasters is low, resulting in one-sided assessment results. At the same time, their spatial resolution is limited, and they mostly use large areas as assessment units. There is a lack of consideration of the refined characteristics of sub-regions, making it difficult to reflect local risk differences. Therefore, it is necessary to provide a meteorological disaster risk assessment method and system based on the Internet of Things to solve the above-mentioned problems. Summary of the Invention

[0004] In order to solve the above technical problems, a meteorological disaster risk assessment method and system based on the Internet of Things is provided. This technical solution solves the problems of the existing meteorological disaster risk assessment method and system based on the Internet of Things proposed in the above background technology, which have insufficient data fusion capabilities, mostly rely on a single type of sensor data, and have low integration of multi-dimensional information such as elevation, terrain, and historical disasters, resulting in one-sided assessment results. At the same time, its spatial resolution is limited, and it mostly uses large areas as assessment units. It lacks consideration of the refined characteristics of sub-regions and is difficult to reflect local risk differences.

[0005] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0006] A meteorological disaster risk assessment method based on the Internet of Things, comprising:

[0007] Obtaining geographic location data of the area to be assessed, thereby determining the corresponding elevation data and historical meteorological environment data, and determining a continuous elevation surface and meteorological environment surface of the area to be assessed;

[0008] Construct a risk assessment coordinate system, and determine the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system by combining the corresponding elevation data and historical meteorological environment data;

[0009] According to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, the area to be assessed is divided into n sub-areas to be assessed;

[0010] Obtaining the natural disaster resistance characteristic data of the sub-region to be assessed from the elevation data corresponding to the area to be assessed, and obtaining the disaster resistance infrastructure data of the sub-region to be assessed, thereby determining the initial disaster resistance coefficient of the sub-region to be assessed;

[0011] Determine the first sub-region to be evaluated from the n sub-regions to be evaluated, and simultaneously obtain historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be evaluated. Then, determine the homogeneity coefficient of the first sub-region to be evaluated by combining the sub-regions to be evaluated adjacent to the first sub-region to be evaluated.

[0012] Obtain meteorological forecast data, and determine the meteorological disaster risk sub-coefficient of the sub-region to be evaluated based on the homogeneity coefficient of the first sub-region to be evaluated and the initial coefficient of disaster resistance of the sub-region to be evaluated, thereby determining the meteorological disaster risk coefficient of the sub-region to be evaluated.

[0013] In an optional embodiment, obtaining geographic location data of the area to be evaluated, thereby determining corresponding elevation data and historical meteorological environment data, and determining a continuous elevation surface and meteorological environment surface of the area to be evaluated, specifically includes:

[0014] Using the acquisition module, obtain the latitude and longitude coordinates of the area to be evaluated and obtain the geographical location data of the area to be evaluated;

[0015] Through the GIS interface, query the elevation data corresponding to the coordinate set and generate the elevation matrix;

[0016] Call the meteorological data platform API to obtain the historical meteorological environment data of the area to be evaluated for the past T years and construct the historical meteorological tensor M;

[0017] Perform time series normalization on the historical meteorological tensor M, obtain the annual average value of each meteorological element in the historical meteorological tensor, and generate the historical meteorological environment feature vector;

[0018] Establish an association mapping table between geographic location data, elevation data, and historical meteorological environment data, and store the association relationship between the elevation matrix corresponding to the latitude and longitude coordinate set and the historical meteorological environment feature vector;

[0019] The Kriging interpolation algorithm is used to perform spatial interpolation on the discrete data in the elevation data and historical meteorological environment data to generate continuous elevation surfaces and meteorological environment surfaces for the area to be evaluated;

[0020] Based on the continuous elevation surface and meteorological environment surface of the area to be evaluated, the elevation surface slope matrix, elevation surface aspect matrix, meteorological environment surface slope matrix and meteorological environment surface aspect matrix are obtained;

[0021] The frequency of disastrous weather in historical meteorological and environmental data is statistically analyzed to generate a disaster frequency vector.

[0022] In an optional embodiment, the risk assessment coordinate system is constructed, and the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system is determined in combination with the corresponding elevation data and historical meteorological environment data, specifically including:

[0023] Project the continuous elevation surface of the area to be assessed to obtain a reference figure of the origin of the risk assessment coordinate system;

[0024] Obtain the geometric center point corresponding to the reference figure of the risk assessment coordinate system origin as the origin of the risk assessment coordinate system, and construct the risk assessment coordinate system with the minimum circumscribed rectangle of the area to be assessed as the boundary;

[0025] Grid the risk assessment coordinate system to obtain a set of coordinate points;

[0026] Obtaining a correlation mapping table of geographic location data, elevation data, and historical meteorological environment data, and querying the elevation value and historical meteorological characteristic vector corresponding to each coordinate point in the coordinate point set from the correlation mapping table;

[0027] Define the characteristic coefficient calculation function, and simultaneously normalize the elevation value corresponding to each coordinate point to obtain the standard elevation value;

[0028] Perform principal component analysis on the historical meteorological feature vector to obtain the principal component matrix, and select the first principal component as the comprehensive value of the meteorological feature;

[0029] Substitute the standard elevation value and the comprehensive value of meteorological characteristics into the characteristic coefficient calculation function to obtain the characteristic coefficient of each coordinate point in the risk assessment coordinate system, thereby constructing the characteristic coefficient matrix.

[0030] In an optional embodiment, the region to be assessed is divided according to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system to obtain n sub-regions to be assessed, specifically including:

[0031] The K-means clustering algorithm is used to divide the coordinate points corresponding to the characteristic coefficient matrix into n categories, and the range of the number of clusters is set simultaneously, and the elbow method is used to determine the optimal number of clusters;

[0032] Initialize the cluster center and obtain the Euclidean distance from each coordinate point to the cluster center;

[0033] Set the center change rate threshold, iteratively update the cluster center until the center change rate is less than the center change rate threshold, and obtain the standard cluster center;

[0034] Based on the standard cluster centers, the coordinate points are mapped back to the risk assessment coordinate system to generate polygonal boundaries of n sub-areas to be assessed;

[0035] Determine the area of ​​the polygonal boundary corresponding to each sub-region to be evaluated and generate a sub-region morphological feature vector;

[0036] Count the mean and variance of the characteristic coefficients of the coordinate points in each sub-region to be evaluated to generate a sub-region characteristic statistical vector;

[0037] Establish an association table between the sub-region to be evaluated and the geographical location data, and store the sub-region morphological feature vector and the sub-region feature statistical vector;

[0038] Obtaining a silhouette coefficient based on the sub-region morphological feature vector and the sub-region feature statistical vector;

[0039] Set the silhouette coefficient threshold. If the silhouette coefficient is greater than the silhouette coefficient threshold, the division result corresponding to the silhouette coefficient is used as the final division result, thereby determining n sub-areas to be evaluated. If the silhouette coefficient is less than or equal to the silhouette coefficient threshold, redetermine the optimal number of clusters and re-divide the area to be evaluated until the silhouette coefficient is greater than the silhouette coefficient threshold.

[0040] In an optional embodiment, obtaining the natural disaster resistance characteristic data of the sub-region to be assessed from the elevation data corresponding to the area to be assessed, and obtaining the disaster resistance infrastructure data of the sub-region to be assessed, thereby determining the initial coefficient of the disaster resistance capacity of the sub-region to be assessed, specifically includes:

[0041] For each sub-area to be evaluated, extract its elevation matrix to obtain the terrain relief and average elevation;

[0042] Define a set of natural disaster resistance characteristic indicators and obtain the vegetation coverage of the sub-area to be evaluated through remote sensing image analysis;

[0043] The soil permeability of the sub-area to be assessed is queried through the soil database, and combined with the vegetation coverage of the sub-area to be assessed, a natural disaster resistance feature vector is constructed;

[0044] IoT devices are used to collect disaster resilience infrastructure data within the sub-region, including flood levee length, drainage system capacity, and number of shelters, and simultaneously construct a disaster resilience infrastructure vector.

[0045] Perform Z-score normalization on the natural disaster resistance feature vector and the disaster resistance infrastructure vector to obtain the natural disaster resistance feature standard vector and the disaster resistance infrastructure standard vector;

[0046] Define the calculation function of the initial coefficient of disaster resistance, and determine the initial coefficient of disaster resistance of the sub-area to be assessed based on the standard vector of natural disaster resistance characteristics and the standard vector of disaster resistance infrastructure.

[0047] In an optional embodiment, the method of sequentially determining a first sub-region to be evaluated from the n sub-regions to be evaluated, synchronously acquiring historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be evaluated, and then determining the homogeneity coefficient of the first sub-region to be evaluated in combination with the sub-regions to be evaluated adjacent to the first sub-region to be evaluated, specifically includes:

[0048] Traverse n sub-regions to be evaluated and take the i-th sub-region as the first sub-region to be evaluated ;

[0049] Get the first sub-region to be evaluated The corresponding historical meteorological tensor is extracted from the first sub-area to be evaluated The corresponding historical meteorological disaster information, including the number of disasters and the scope of disaster impact, is used to simultaneously construct the disaster feature vector;

[0050] Use the spatial adjacency matrix to determine the first sub-region to be evaluated Adjacent sub-areas to be evaluated , and then obtain the first sub-region to be evaluated Adjacent sub-areas to be evaluated The corresponding standard vectors of natural disaster resistance characteristics and disaster resistance infrastructure;

[0051] Based on the natural disaster resistance characteristic standard vector and the disaster resistance infrastructure standard vector, the first sub-area to be assessed is determined and the adjacent sub-areas to be evaluated The cosine similarity of , thereby determining the first sub-region to be evaluated The homogeneity coefficient.

[0052] In an optional embodiment, the meteorological forecast data is obtained, and based on the homogeneity coefficient of the first sub-region to be evaluated and the initial coefficient of disaster resistance of the sub-region to be evaluated, the meteorological disaster risk sub-coefficient of the sub-region to be evaluated is determined, thereby determining the meteorological disaster risk coefficient of the sub-region to be evaluated, specifically including:

[0053] Call the weather forecast platform API to obtain the future The weather forecast data of the day is used to construct the forecast weather tensor;

[0054] Disaster weather identification is performed on the forecast meteorological tensor, the threshold method is used to determine the disaster type, and the predicted disaster feature vector is generated;

[0055] Determine the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the predicted disaster characteristic vector, the homogeneity coefficient of the first sub-region to be assessed, and the initial coefficient of the disaster resistance capacity of the sub-region to be assessed;

[0056] Normalize the meteorological disaster risk sub-coefficients of the sub-region to be assessed to obtain the meteorological disaster risk standard sub-coefficients;

[0057] Obtain the sub-region area values ​​of n sub-regions to be assessed in the risk assessment coordinate system, and obtain the meteorological disaster risk coefficient of the sub-region to be assessed by combining the meteorological disaster risk standard sub-coefficient of the sub-region to be assessed.

[0058] Furthermore, a meteorological disaster risk assessment system based on the Internet of Things is proposed, which is used to implement any of the above assessment methods, including:

[0059] An acquisition module is used to obtain geographic location data of the area to be evaluated, thereby determining corresponding elevation data and historical meteorological environment data, and also to obtain meteorological forecast data;

[0060] A data management module, the data management module is used to determine the continuous elevation surface and meteorological environment surface of the area to be assessed, perform data preprocessing on the data acquired by the acquisition module, perform data processing and normalization on the data output by the data management module, and construct a risk assessment coordinate system. In combination with the corresponding elevation data and historical meteorological environment data, the module determines the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, divides the area to be assessed into n sub-areas to be assessed based on the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, and manages the area to be assessed and the sub-areas to be assessed;

[0061] a risk assessment module, the risk assessment module being used to obtain natural disaster resistance characteristic data of a sub-region to be assessed from elevation data corresponding to the region to be assessed, and to obtain disaster resistance infrastructure data of the sub-region to be assessed, thereby determining an initial disaster resistance coefficient of the sub-region to be assessed; and sequentially determining a first sub-region to be assessed from the n sub-regions to be assessed, and synchronously obtaining historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be assessed, and then determining a homogeneity coefficient of the first sub-region to be assessed in combination with sub-regions to be assessed adjacent to the first sub-region to be assessed; and determining a meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the homogeneity coefficient of the first sub-region to be assessed and the initial disaster resistance coefficient of the sub-region to be assessed, thereby determining the meteorological disaster risk coefficient of the region to be assessed;

[0062] A risk warning module is used to issue a risk warning to the area to be assessed based on the meteorological disaster risk coefficient of the area to be assessed;

[0063] The display module is used to present the process and results of meteorological disaster risk assessment to users.

[0064] In an optional embodiment, the data management module includes:

[0065] A data processing unit, which is used to determine the continuous elevation surface and meteorological environment surface of the area to be evaluated, perform data preprocessing on the data acquired by the acquisition module, and perform data processing and normalization on the data output by the data management module;

[0066] A coordinate system management unit, the coordinate system management unit is used to construct a risk assessment coordinate system and determine a characteristic coefficient matrix for each coordinate point in the risk assessment coordinate system in combination with corresponding elevation data and historical meteorological environment data;

[0067] The area management unit is used to divide the area to be evaluated into n sub-areas to be evaluated according to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, and manage the area to be evaluated and the sub-areas to be evaluated.

[0068] In an optional embodiment, the risk assessment module includes:

[0069] A regional disaster resistance assessment unit, configured to obtain natural disaster resistance characteristic data of a sub-region to be assessed from elevation data corresponding to the region to be assessed, and to obtain disaster resistance infrastructure data of the sub-region to be assessed, thereby determining an initial coefficient of disaster resistance capacity of the sub-region to be assessed;

[0070] a regional mutual impact assessment unit, the regional mutual impact assessment unit being configured to sequentially determine a first sub-region to be assessed from the n sub-regions to be assessed, synchronously obtain historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be assessed, and then determine a homogeneity coefficient for the first sub-region to be assessed in combination with adjacent sub-regions to be assessed;

[0071] The regional risk assessment unit is used to determine the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the homogeneity coefficient of the first sub-region to be assessed and the initial coefficient of the disaster resistance capacity of the sub-region to be assessed, thereby determining the meteorological disaster risk coefficient of the region to be assessed.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] This proposal proposes an IoT-based meteorological disaster risk assessment method. By constructing a risk assessment coordinate system, the method combines elevation data and historical meteorological environmental data to determine the characteristic coefficient matrix of coordinate points. Based on this matrix, the method divides the assessment area into n sub-regions to be assessed. This method achieves a refined division of the assessment area and accurately captures the differences in meteorological disaster risks in different sub-regions, laying the foundation for subsequent targeted assessments.

[0074] This proposal proposes a meteorological disaster risk assessment method based on the Internet of Things. It calculates the initial disaster resistance coefficient by integrating natural disaster resistance characteristic data and disaster resistance infrastructure data, determines the homogeneity coefficient based on the characteristics of adjacent sub-regions, and then associates it with meteorological forecast data to obtain risk sub-coefficients and total risk coefficients. It achieves in-depth integration of multi-dimensional data, comprehensively considers the region's own disaster resistance capacity, the impact of adjacent regions and future meteorological trends, and improves the comprehensiveness and accuracy of risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a flow chart of a meteorological disaster risk assessment method based on the Internet of Things proposed by the present invention;

[0076] Figure 2 This is a flow chart for obtaining the elevation surface and the meteorological environment surface in the present invention;

[0077] Figure 3 This is a flow chart for obtaining the characteristic coefficient matrix in the present invention;

[0078] Figure 4 This is a system framework diagram of a meteorological disaster risk assessment system based on the Internet of Things proposed in the present invention. DETAILED DESCRIPTION

[0079] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0080] Reference Figure 1 - Figure 4 As shown, a meteorological disaster risk assessment method based on the Internet of Things includes:

[0081] Obtaining geographic location data of the area to be assessed, thereby determining the corresponding elevation data and historical meteorological environment data, and determining a continuous elevation surface and meteorological environment surface of the area to be assessed;

[0082] Construct a risk assessment coordinate system, and determine the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system by combining the corresponding elevation data and historical meteorological environment data;

[0083] According to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, the area to be assessed is divided into n sub-areas to be assessed;

[0084] Obtaining the natural disaster resistance characteristic data of the sub-region to be assessed from the elevation data corresponding to the area to be assessed, and obtaining the disaster resistance infrastructure data of the sub-region to be assessed, thereby determining the initial disaster resistance coefficient of the sub-region to be assessed;

[0085] Determine the first sub-region to be evaluated from the n sub-regions to be evaluated, and simultaneously obtain historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be evaluated. Then, determine the homogeneity coefficient of the first sub-region to be evaluated by combining the sub-regions to be evaluated adjacent to the first sub-region to be evaluated.

[0086] Obtain meteorological forecast data, and determine the meteorological disaster risk sub-coefficient of the sub-region to be evaluated based on the homogeneity coefficient of the first sub-region to be evaluated and the initial coefficient of disaster resistance of the sub-region to be evaluated, thereby determining the meteorological disaster risk coefficient of the sub-region to be evaluated.

[0087] Furthermore, the geographical location data of the area to be evaluated is obtained to determine the corresponding elevation data and historical meteorological environment data, and to determine the continuous elevation surface and meteorological environment surface of the area to be evaluated, specifically including:

[0088] Using the acquisition module, obtain the latitude and longitude coordinates of the area to be evaluated and obtain the geographical location data of the area to be evaluated;

[0089] Through the GIS interface, query the elevation data corresponding to the coordinate set and generate the elevation matrix;

[0090] Call the meteorological data platform API to obtain the historical meteorological environment data of the area to be evaluated for the past T years and construct the historical meteorological tensor M;

[0091] Perform time series normalization on the historical meteorological tensor M, obtain the annual average value of each meteorological element in the historical meteorological tensor, and generate the historical meteorological environment feature vector;

[0092] Establish an association mapping table between geographic location data, elevation data, and historical meteorological environment data, and store the association relationship between the elevation matrix corresponding to the latitude and longitude coordinate set and the historical meteorological environment feature vector;

[0093] The Kriging interpolation algorithm is used to perform spatial interpolation on the discrete data in the elevation data and historical meteorological environment data to generate continuous elevation surfaces and meteorological environment surfaces for the area to be evaluated;

[0094] Based on the continuous elevation surface and meteorological environment surface of the area to be evaluated, the elevation surface slope matrix, elevation surface aspect matrix, meteorological environment surface slope matrix and meteorological environment surface aspect matrix are obtained;

[0095] The frequency of disastrous weather in historical meteorological and environmental data is statistically analyzed to generate a disaster frequency vector.

[0096] Specifically, the latitude and longitude set of the area to be evaluated is ,in, is the longitude value, is the latitude value, i∈m. Through the GIS (Geographic Information System) interface, query the elevation data corresponding to the coordinate set G and generate the elevation matrix ,in Represents the elevation value of the coordinate point in the i-th row and j-th column, j∈n, and n is the total number of columns. The historical meteorological environment data of the area to be evaluated in the past T years include wind speed, rainfall, temperature, etc., so the historical meteorological tensor constructed is , where t is the year dimension, k is the meteorological element dimension (wind speed, rainfall, etc.), and p is the number of meteorological element types. The historical meteorological environment feature vector is ,in , Indicates the The time average characteristic value of the meteorological element, such as For “wind speed”, it is the average wind speed of the area to be evaluated in the past T years. The corresponding “rainfall” is the average rainfall in the past T years, which is used to extract the long-term statistical characteristics of meteorological elements as the basic environmental parameters for risk assessment. Indicates the tth year (or tth time period), Spatial distribution data of meteorological elements, It is a spatial wildcard, which means "covering all spatial coordinate points in the area to be evaluated" (corresponding to the two-dimensional space of the latitude and longitude grid). Establish a mapping table of geographic location information, elevation, and meteorological information. , store coordinate points Corresponding and The Kriging interpolation algorithm is used to perform spatial interpolation on the elevation and meteorological information of discrete coordinate points to generate a continuous elevation surface H(x,y) and a meteorological environment surface M(x,y) of the area to be evaluated. The slope and aspect matrix S=[s i,j ] m×n 、A=[a i,j ] m×n , where the slope s i,j Reflecting the impact of terrain undulation on disasters, the slope matrix A=[ai,j ] m×n It is a two-dimensional matrix used to describe the terrain orientation characteristics of the area to be assessed. Its core function is to quantify the inclination angle of the terrain slope relative to the north direction, and provide basic terrain parameters for risk simulation of meteorological disasters (such as slope runoff caused by heavy rain and terrain blocking effect of strong wind). Each element a in the matrix is i,j The slope angle of the coordinate point in the i-th row and j-th column of the grid in the assessment area ranges from 0° to 360° (with due north as 0∘ and increasing clockwise). The frequency of disastrous weather (such as heavy rain and strong wind) in historical meteorological environmental data is counted to generate a disaster frequency vector , where q is the number of disaster types.

[0097] Furthermore, a risk assessment coordinate system is constructed, and the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system is determined by combining the corresponding elevation data and historical meteorological environment data, including:

[0098] Project the continuous elevation surface of the area to be assessed to obtain a reference figure of the origin of the risk assessment coordinate system;

[0099] Obtain the geometric center point corresponding to the reference figure of the risk assessment coordinate system origin as the origin of the risk assessment coordinate system, and construct the risk assessment coordinate system with the minimum circumscribed rectangle of the area to be assessed as the boundary;

[0100] Grid the risk assessment coordinate system to obtain a set of coordinate points;

[0101] Obtaining a correlation mapping table of geographic location data, elevation data, and historical meteorological environment data, and querying the elevation value and historical meteorological characteristic vector corresponding to each coordinate point in the coordinate point set from the correlation mapping table;

[0102] Define the characteristic coefficient calculation function, and simultaneously normalize the elevation value corresponding to each coordinate point to obtain the standard elevation value;

[0103] Perform principal component analysis on the historical meteorological feature vector to obtain the principal component matrix, and select the first principal component as the comprehensive value of the meteorological feature;

[0104] Substitute the standard elevation value and the comprehensive value of meteorological characteristics into the characteristic coefficient calculation function to obtain the characteristic coefficient of each coordinate point in the risk assessment coordinate system, thereby constructing the characteristic coefficient matrix.

[0105] Specifically, the risk assessment coordinate system is gridded to obtain a set of coordinate points: , N is the total number of grids, and the coordinate range is For each coordinate point, query the corresponding elevation value from the associated mapping table and historical meteorological environment characteristic vector , where p is the index value, is the elevation value corresponding to the p-th coordinate point.

[0106] The expression formula of the characteristic coefficient calculation function is:

[0107]

[0108] Where, Both are weight coefficients. To perform principal component analysis on meteorological feature vectors, the first principal component is extracted as the comprehensive value of meteorological features. The mean of the principal component analysis results for the historical meteorological feature vectors, is its standard deviation, is the maximum elevation value, is the minimum elevation value.

[0109] It can be understood that the elevation value corresponding to each coordinate point is normalized to obtain the standard elevation value. The specific formula is: , is the standard elevation value corresponding to the pth coordinate point. his,p Perform principal component analysis (PCA) to obtain the principal component matrix P PCA =[p i,k ] p×k , select the first principal component p 1,k As a comprehensive value of meteorological characteristics, The Z-score standardization is performed on the PCA results. The standard value range after standardization is approximately [−3, 3]. Substitute the standard elevation value and the comprehensive value of meteorological characteristics into the characteristic coefficient function to calculate the characteristic coefficient c of each coordinate point. p =C(h p ,F his,p ), construct the characteristic coefficient matrix C=[c p ] 1×N ; Use spatial autocorrelation analysis to calculate the spatial correlation index IC of the characteristic coefficient matrix and verify the spatial distribution law of the characteristic coefficient; Based on the characteristic coefficient matrix C, use Gaussian filtering to smooth the characteristic coefficients of the coordinate points and generate a smoothed characteristic coefficient matrix C smooth =[c smooth,p ] 1×N , reduce noise interference.

[0110] Furthermore, the area to be assessed is divided into n sub-areas to be assessed based on the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, specifically including:

[0111] The K-means clustering algorithm is used to divide the coordinate points corresponding to the characteristic coefficient matrix into n categories, and the range of the number of clusters is set simultaneously, and the elbow method is used to determine the optimal number of clusters;

[0112] Initialize the cluster center and obtain the Euclidean distance from each coordinate point to the cluster center;

[0113] Set the center change rate threshold, iteratively update the cluster center until the center change rate is less than the center change rate threshold, and obtain the standard cluster center;

[0114] Based on the standard cluster centers, the coordinate points are mapped back to the risk assessment coordinate system to generate polygonal boundaries of n sub-areas to be assessed;

[0115] Determine the area of ​​the polygonal boundary corresponding to each sub-region to be evaluated and generate a sub-region morphological feature vector;

[0116] Count the mean and variance of the characteristic coefficients of the coordinate points in each sub-region to be evaluated to generate a sub-region characteristic statistical vector;

[0117] Establish an association table between the sub-region to be evaluated and the geographical location data, and store the sub-region morphological feature vector and the sub-region feature statistical vector;

[0118] Obtaining a silhouette coefficient based on the sub-region morphological feature vector and the sub-region feature statistical vector;

[0119] Set the silhouette coefficient threshold. If the silhouette coefficient is greater than the silhouette coefficient threshold, the division result corresponding to the silhouette coefficient is used as the final division result, thereby determining n sub-areas to be evaluated. If the silhouette coefficient is less than or equal to the silhouette coefficient threshold, redetermine the optimal number of clusters and re-divide the area to be evaluated until the silhouette coefficient is greater than the silhouette coefficient threshold.

[0120] Specifically, the K-means clustering algorithm is used to transform the characteristic coefficient matrix C smooth The corresponding coordinate points are divided into n categories, and the value range of cluster number n is set to [2,10]. The optimal number of clusters is determined by the elbow method; the cluster center C is initialized. center =[c 1,0 ,c 2,0 ,…,c n,0 ], calculate the Euclidean distance from each coordinate point to the cluster center: ; Iteratively update the cluster center until the convergence condition is met (such as the center change rate is less than ϵ=10 −5 ), get the final cluster center According to the clustering results, the coordinate points are mapped back to the geographic space to generate the polygonal boundaries B={B1,B2,…,B n}, where Bi is the boundary coordinate set of the i-th sub-region; calculate the area of ​​each sub-region ,perimeter , generate sub-region morphological feature vector F shape,i =[A i ,L i ]; Count the mean and variance of the characteristic coefficients of the coordinate points in each sub-region to generate the sub-region characteristic statistical vector ,in, , N i The number of coordinate points in the sub-area; establish an association table between the sub-area and geographic information , store the basic features of the sub-region; use the silhouette coefficient to evaluate the clustering effect and calculate the silhouette coefficient S C , verify the rationality of the sub-region division, if S C <0.5, then readjust the number of clusters n.

[0121] Furthermore, the natural disaster resistance characteristic data of the sub-region to be assessed is obtained from the elevation data corresponding to the area to be assessed, and the disaster resistance infrastructure data of the sub-region to be assessed is obtained, so as to determine the initial disaster resistance coefficient of the sub-region to be assessed, which specifically includes:

[0122] For each sub-area to be evaluated, extract its elevation matrix to obtain the terrain relief and average elevation;

[0123] Define a set of natural disaster resistance characteristic indicators and obtain the vegetation coverage of the sub-area to be evaluated through remote sensing image analysis;

[0124] The soil permeability of the sub-area to be assessed is queried through the soil database, and combined with the vegetation coverage of the sub-area to be assessed, a natural disaster resistance feature vector is constructed;

[0125] IoT devices are used to collect disaster resilience infrastructure data within the sub-region, including flood levee length, drainage system capacity, and number of shelters, and simultaneously construct a disaster resilience infrastructure vector.

[0126] Perform Z-score normalization on the natural disaster resistance feature vector and the disaster resistance infrastructure vector to obtain the natural disaster resistance feature standard vector and the disaster resistance infrastructure standard vector;

[0127] Define the calculation function of the initial coefficient of disaster resistance, and determine the initial coefficient of disaster resistance of the sub-area to be assessed based on the standard vector of natural disaster resistance characteristics and the standard vector of disaster resistance infrastructure.

[0128] Specifically, for each sub-region B to be evaluated i , extract its elevation matrix , calculate the terrain relief , average elevation . Define the natural disaster resistance characteristic index set I natural ={terrain relief, vegetation coverage, soil permeability}, and obtain the sub-region vegetation coverage V through remote sensing image analysis i , obtain the soil permeability K by querying the soil database i , construct the natural disaster resistance feature vector F natural,i =[R i ,V i ,K i ] Use IoT devices to collect disaster prevention infrastructure data in the sub-area, including the length of the flood levee L dyke,i , drainage system capacity C drain,i 、Number of shelters N shelter,i , building disaster-resistant infrastructure vector F infra,i =[L dyke,i ,C drain,i ,N shelter,i ]. The natural disaster resistance feature vector and infrastructure vector are normalized using z-score normalization: ,in 、 are the mean and standard deviation of the natural features, 、 are the means and standard deviations of infrastructure characteristics.

[0129] The calculation formula of the initial coefficient of disaster resistance is:

[0130]

[0131] Where, The weights of natural and human factors were determined by the analytic hierarchy process (AHP);

[0132] Furthermore, it is necessary to calculate the initial coefficient of disaster resistance of the sub-region , and normalized, we get ;

[0133] Establish a correlation table between disaster resilience and sub-regions , Grey correlation analysis was used to calculate the correlation between natural disaster resistance characteristics, infrastructure and disaster resistance capacity to verify the validity of the selected indicators. If the correlation was less than 0.6, the indicators were supplemented or adjusted.

[0134] Furthermore, a first sub-region to be evaluated is determined from the n sub-regions to be evaluated in sequence, and historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be evaluated are simultaneously obtained. Then, the homogeneity coefficient of the first sub-region to be evaluated is determined by combining the sub-regions to be evaluated adjacent to the first sub-region to be evaluated, specifically including:

[0135] Traverse n sub-regions to be evaluated and take the i-th sub-region as the first sub-region to be evaluated ;

[0136] Get the first sub-region to be evaluated The corresponding historical meteorological tensor is extracted from the first sub-area to be evaluated Corresponding historical meteorological disaster information, including the number of disasters that occurred and the scope of disaster impact , and simultaneously construct disaster feature vectors ;

[0137] Use the spatial adjacency matrix to determine the first sub-region to be evaluated Adjacent sub-areas to be evaluated , and then obtain the first sub-region to be evaluated Adjacent sub-areas to be evaluated The corresponding standard vectors of natural disaster resistance characteristics and disaster resistance infrastructure;

[0138] Based on the natural disaster resistance characteristic standard vector and the disaster resistance infrastructure standard vector, the first sub-area to be assessed is determined and the adjacent sub-areas to be evaluated The cosine similarity of , thereby determining the first sub-region to be evaluated The homogeneity coefficient.

[0139] Specifically, determine the first sub-area to be evaluated and the adjacent sub-areas to be evaluated The cosine similarity of , the cosine similarity calculation formula used is: . Determine the first sub-area to be evaluated The calculation of the homogeneity coefficient needs to consider the weighted sum of the similarities of adjacent sub-regions:

[0140]

[0141] Where, is the self-similarity of the first sub-region to be evaluated, The first sub-region to be evaluated and the adjacent sub-areas to be evaluated The cosine similarity of is the number of adjacent sub-regions.

[0142] Furthermore, the homogeneity coefficient Perform normalization and obtain , then establish the association table between homogeneity coefficient and sub-region , and the rationality of the homogeneity coefficient was verified by spatial autocorrelation.

[0143] Furthermore, meteorological forecast data is obtained, and based on the homogeneity coefficient of the first sub-region to be assessed and the initial coefficient of disaster resistance of the sub-region to be assessed, the meteorological disaster risk sub-coefficient of the sub-region to be assessed is determined, thereby determining the meteorological disaster risk coefficient of the sub-region to be assessed, specifically including:

[0144] Call the weather forecast platform API to obtain the future The weather forecast data of the day is used to construct the forecast weather tensor;

[0145] Disaster weather identification is performed on the forecast meteorological tensor, the threshold method is used to determine the disaster type, and the predicted disaster feature vector is generated;

[0146] Determine the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the predicted disaster characteristic vector, the homogeneity coefficient of the first sub-region to be assessed, and the initial coefficient of the disaster resistance capacity of the sub-region to be assessed;

[0147] Normalize the meteorological disaster risk sub-coefficients of the sub-region to be assessed to obtain the meteorological disaster risk standard sub-coefficients;

[0148] Obtain the sub-region area values ​​of n sub-regions to be assessed in the risk assessment coordinate system, and obtain the meteorological disaster risk coefficient of the sub-region to be assessed by combining the meteorological disaster risk standard sub-coefficient of the sub-region to be assessed.

[0149] Specifically, the weather forecast platform API is called to obtain the future The weather forecast data for the day is used to construct the forecast weather tensor ; Disaster weather identification is performed on the forecast meteorological tensor, and the threshold method is used to judge heavy rain (such as rainfall > R th ), strong wind (such as wind speed>V th ) and other disasters, generate the predicted disaster feature vector: ;

[0150] Define the calculation function of meteorological disaster risk sub-coefficient, combining the homogeneity coefficient and the initial coefficient of disaster resistance:

[0151]

[0152] where ϵ=10 −3 To avoid the minimum value with zero denominator;

[0153] Risk sub-coefficient Perform normalization and obtain ;

[0154] Calculate the meteorological disaster risk coefficient of the area to be assessed using the weighted summation method:

[0155]

[0156] in is the sub-region area weight;

[0157] Establish a correlation table between risk factors, sub-regions, and regions ;

[0158] Monte Carlo simulation is used to quantify the uncertainty of meteorological forecast data and generate confidence intervals for risk factors. .

[0159] Furthermore, it is necessary to issue risk warnings for the areas to be assessed based on the meteorological disaster risk coefficients of the areas to be assessed:

[0160] Determine the risk warning level: Multiple risk warning levels are set, such as low, medium, high, and extremely high. Each level is assigned a risk factor range. For example: Low risk: A risk factor in the range [0, 0.2] indicates a low likelihood of a meteorological disaster in the region, and the impact of a disaster would be minimal. Medium risk: A risk factor in the range [0.2, 0.5] indicates a certain risk of meteorological disasters in the region, and a disaster could have a certain degree of impact on local facilities, production, and daily life. High risk: A risk factor in the range [0.5, 0.8] indicates a high risk of meteorological disasters in the region, and a disaster could cause widespread damage and losses. Extremely high risk: A risk factor in the range [0.8, 1] indicates an extremely high risk of meteorological disasters in the region, and a disaster could cause severe property damage and infrastructure destruction. Experienced personnel will then develop a warning release mechanism. The system will then use the risk factor and threshold to determine which warnings to issue. This process also requires dissemination of information to professionals for manual review.

[0161] Furthermore, a meteorological disaster risk assessment system based on the Internet of Things is proposed, which is used to implement any of the above assessment methods, including:

[0162] The acquisition module is used to obtain the geographical location data of the area to be evaluated, thereby determining the corresponding elevation data and historical meteorological environment data, and is also used to obtain meteorological forecast data;

[0163] The data management module is used to determine the continuous elevation surface and meteorological environment surface of the area to be assessed, perform data preprocessing on the data acquired by the acquisition module, and perform data processing and normalization on the data output by the data management module to construct a risk assessment coordinate system. In combination with the corresponding elevation data and historical meteorological environment data, the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system is determined. The characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system is used to divide the area to be assessed into n sub-areas to be assessed, and the area to be assessed and the sub-areas to be assessed are managed;

[0164] A risk assessment module, the risk assessment module is used to obtain natural disaster resistance characteristic data of the sub-region to be assessed from the elevation data corresponding to the area to be assessed, and obtain disaster resistance infrastructure data of the sub-region to be assessed, so as to determine the initial disaster resistance coefficient of the sub-region to be assessed, and to determine the first sub-region to be assessed from the n sub-regions to be assessed in sequence, and to simultaneously obtain historical meteorological disaster data, natural disaster resistance characteristic data and disaster resistance infrastructure data corresponding to the first sub-region to be assessed, and then determine the homogeneity coefficient of the first sub-region to be assessed in combination with the sub-regions to be assessed adjacent to the first sub-region to be assessed, and to determine the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the homogeneity coefficient of the first sub-region to be assessed and the initial disaster resistance coefficient of the sub-region to be assessed, so as to determine the meteorological disaster risk coefficient of the area to be assessed;

[0165] Risk warning module: The risk warning module is used to issue risk warnings to the area to be assessed based on the meteorological disaster risk coefficient of the area to be assessed;

[0166] The display module is used to present the process and results of meteorological disaster risk assessment to users.

[0167] Furthermore, the data management module includes:

[0168] The data processing unit is used to determine the continuous elevation surface and meteorological environment surface of the area to be evaluated, perform data preprocessing on the data acquired by the acquisition module, and perform data processing and normalization on the data output by the data management module;

[0169] The coordinate system management unit is used to construct a risk assessment coordinate system and determine the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system in combination with the corresponding elevation data and historical meteorological environment data;

[0170] The regional management unit is used to divide the area to be assessed into n sub-areas to be assessed according to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, and manage the area to be assessed and the sub-areas to be assessed.

[0171] Furthermore, the risk assessment module includes:

[0172] A regional disaster resistance assessment unit is used to obtain natural disaster resistance characteristic data of the sub-region to be assessed from the elevation data corresponding to the sub-region to be assessed, and obtain disaster resistance infrastructure data of the sub-region to be assessed, so as to determine the initial disaster resistance coefficient of the sub-region to be assessed;

[0173] A regional mutual impact assessment unit is used to sequentially determine a first sub-region to be assessed from the n sub-regions to be assessed, synchronously obtain historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be assessed, and then determine the homogeneity coefficient of the first sub-region to be assessed by combining the sub-regions to be assessed adjacent to the first sub-region to be assessed;

[0174] The regional risk assessment unit is used to determine the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the homogeneity coefficient of the first sub-region to be assessed and the initial coefficient of the disaster resistance capacity of the sub-region to be assessed, thereby determining the meteorological disaster risk coefficient of the region to be assessed.

[0175] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A meteorological disaster risk assessment method based on the Internet of Things, characterized in that: include: Obtaining geographic location data of the area to be assessed, thereby determining the corresponding elevation data and historical meteorological environment data, and determining a continuous elevation surface and meteorological environment surface of the area to be assessed; Construct a risk assessment coordinate system, and determine the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system by combining the corresponding elevation data and historical meteorological environment data; According to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, the area to be assessed is divided into n sub-areas to be assessed; Obtaining the natural disaster resistance characteristic data of the sub-region to be assessed from the elevation data corresponding to the area to be assessed, and obtaining the disaster resistance infrastructure data of the sub-region to be assessed, thereby determining the initial disaster resistance coefficient of the sub-region to be assessed; Determine the first sub-region to be evaluated from the n sub-regions to be evaluated, and simultaneously obtain historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be evaluated. Then, determine the homogeneity coefficient of the first sub-region to be evaluated by combining the sub-regions to be evaluated adjacent to the first sub-region to be evaluated. Obtain meteorological forecast data, and determine the meteorological disaster risk sub-coefficient of the sub-region to be evaluated based on the homogeneity coefficient of the first sub-region to be evaluated and the initial coefficient of disaster resistance of the sub-region to be evaluated, thereby determining the meteorological disaster risk coefficient of the sub-region to be evaluated.

2. The method for meteorological disaster risk assessment based on the Internet of Things according to claim 1, characterized in that: The acquisition of geographic location data of the area to be evaluated, thereby determining corresponding elevation data and historical meteorological environment data, and determining a continuous elevation surface and meteorological environment surface of the area to be evaluated, specifically includes: Using the acquisition module, obtain the latitude and longitude coordinates of the area to be evaluated and obtain the geographical location data of the area to be evaluated; Through the GIS interface, query the elevation data corresponding to the coordinate set and generate the elevation matrix; Call the meteorological data platform API to obtain the historical meteorological environment data of the area to be evaluated for the past T years and construct the historical meteorological tensor M; Perform time series normalization on the historical meteorological tensor M, obtain the annual average value of each meteorological element in the historical meteorological tensor, and generate the historical meteorological environment feature vector; Establish an association mapping table between geographic location data, elevation data, and historical meteorological environment data, and store the association relationship between the elevation matrix corresponding to the latitude and longitude coordinate set and the historical meteorological environment feature vector; The Kriging interpolation algorithm is used to perform spatial interpolation on the discrete data in the elevation data and historical meteorological environment data to generate continuous elevation surfaces and meteorological environment surfaces for the area to be evaluated; Based on the continuous elevation surface and meteorological environment surface of the area to be evaluated, the elevation surface slope matrix, elevation surface aspect matrix, meteorological environment surface slope matrix and meteorological environment surface aspect matrix are obtained; The frequency of disastrous weather in historical meteorological and environmental data is statistically analyzed to generate a disaster frequency vector.

3. The method for meteorological disaster risk assessment based on the Internet of Things according to claim 1, characterized in that: The risk assessment coordinate system is constructed, and the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system is determined in combination with the corresponding elevation data and historical meteorological environment data, specifically including: Project the continuous elevation surface of the area to be assessed to obtain a reference figure of the origin of the risk assessment coordinate system; Obtain the geometric center point corresponding to the reference figure of the risk assessment coordinate system origin as the origin of the risk assessment coordinate system, and construct the risk assessment coordinate system with the minimum circumscribed rectangle of the area to be assessed as the boundary; Grid the risk assessment coordinate system to obtain a set of coordinate points; Obtaining a correlation mapping table of geographic location data, elevation data, and historical meteorological environment data, and querying the elevation value and historical meteorological characteristic vector corresponding to each coordinate point in the coordinate point set from the correlation mapping table; Define the characteristic coefficient calculation function, and simultaneously normalize the elevation value corresponding to each coordinate point to obtain the standard elevation value; Perform principal component analysis on the historical meteorological feature vector to obtain the principal component matrix, and select the first principal component as the comprehensive value of the meteorological feature; Substitute the standard elevation value and the comprehensive value of meteorological characteristics into the characteristic coefficient calculation function to obtain the characteristic coefficient of each coordinate point in the risk assessment coordinate system, thereby constructing the characteristic coefficient matrix.

4. The method for meteorological disaster risk assessment based on the Internet of Things according to claim 1, characterized in that: The area to be assessed is divided into n sub-areas to be assessed according to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, specifically including: The K-means clustering algorithm is used to divide the coordinate points corresponding to the characteristic coefficient matrix into n categories, and the range of the number of clusters is set simultaneously, and the elbow method is used to determine the optimal number of clusters; Initialize the cluster center and obtain the Euclidean distance from each coordinate point to the cluster center; Set the center change rate threshold, iteratively update the cluster center until the center change rate is less than the center change rate threshold, and obtain the standard cluster center; Based on the standard cluster centers, the coordinate points are mapped back to the risk assessment coordinate system to generate polygonal boundaries of n sub-areas to be assessed; Determine the area of ​​the polygonal boundary corresponding to each sub-region to be evaluated and generate a sub-region morphological feature vector; Count the mean and variance of the characteristic coefficients of the coordinate points in each sub-region to be evaluated to generate a sub-region characteristic statistical vector; Establish an association table between the sub-region to be evaluated and the geographical location data, and store the sub-region morphological feature vector and the sub-region feature statistical vector; Obtaining a silhouette coefficient based on the sub-region morphological feature vector and the sub-region feature statistical vector; Set the silhouette coefficient threshold. If the silhouette coefficient is greater than the silhouette coefficient threshold, the division result corresponding to the silhouette coefficient is used as the final division result, thereby determining n sub-areas to be evaluated. If the silhouette coefficient is less than or equal to the silhouette coefficient threshold, redetermine the optimal number of clusters and re-divide the area to be evaluated until the silhouette coefficient is greater than the silhouette coefficient threshold.

5. The method for meteorological disaster risk assessment based on the Internet of Things according to claim 1, characterized in that: The method of obtaining the natural disaster resistance characteristic data of the sub-region to be assessed from the elevation data corresponding to the sub-region to be assessed, and obtaining the disaster resistance infrastructure data of the sub-region to be assessed, thereby determining the initial disaster resistance coefficient of the sub-region to be assessed, specifically includes: For each sub-area to be evaluated, extract its elevation matrix to obtain the terrain relief and average elevation; Define a set of natural disaster resistance characteristic indicators and obtain the vegetation coverage of the sub-area to be evaluated through remote sensing image analysis; The soil permeability of the sub-area to be assessed is queried through the soil database, and combined with the vegetation coverage of the sub-area to be assessed, a natural disaster resistance feature vector is constructed; IoT devices are used to collect disaster resilience infrastructure data within the sub-region, including flood levee length, drainage system capacity, and number of shelters, and simultaneously construct a disaster resilience infrastructure vector. Perform Z-score normalization on the natural disaster resistance feature vector and the disaster resistance infrastructure vector to obtain the natural disaster resistance feature standard vector and the disaster resistance infrastructure standard vector; Define the calculation function of the initial coefficient of disaster resistance, and determine the initial coefficient of disaster resistance of the sub-area to be assessed based on the standard vector of natural disaster resistance characteristics and the standard vector of disaster resistance infrastructure.

6. The method for meteorological disaster risk assessment based on the Internet of Things according to claim 1, characterized in that: The method of sequentially determining a first sub-region to be evaluated from the n sub-regions to be evaluated, synchronously acquiring historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be evaluated, and then determining a homogeneity coefficient of the first sub-region to be evaluated in combination with adjacent sub-regions to be evaluated, specifically includes: Traverse n sub-regions to be evaluated and take the i-th sub-region as the first sub-region to be evaluated ; Get the first sub-region to be evaluated The corresponding historical meteorological tensor is extracted from the first sub-area to be evaluated The corresponding historical meteorological disaster information, including the number of disasters and the scope of disaster impact, is used to simultaneously construct the disaster feature vector; Use the spatial adjacency matrix to determine the first sub-region to be evaluated Adjacent sub-areas to be evaluated , and then obtain the first sub-region to be evaluated Adjacent sub-areas to be evaluated The corresponding standard vectors of natural disaster resistance characteristics and disaster resistance infrastructure; Based on the natural disaster resistance characteristic standard vector and the disaster resistance infrastructure standard vector, the first sub-area to be assessed is determined and the adjacent sub-areas to be evaluated The cosine similarity of , thereby determining the first sub-region to be evaluated The homogeneity coefficient.

7. The method for meteorological disaster risk assessment based on the Internet of Things according to claim 1, characterized in that: The obtaining of meteorological forecast data and determining the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the homogeneity coefficient of the first sub-region to be assessed and the initial coefficient of disaster resistance of the sub-region to be assessed, thereby determining the meteorological disaster risk coefficient of the sub-region to be assessed, specifically includes: Call the weather forecast platform API to obtain the future The weather forecast data of the day is used to construct the forecast weather tensor; Disaster weather identification is performed on the forecast meteorological tensor, the threshold method is used to determine the disaster type, and the predicted disaster feature vector is generated; Determine the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the predicted disaster characteristic vector, the homogeneity coefficient of the first sub-region to be assessed, and the initial coefficient of the disaster resistance capacity of the sub-region to be assessed; Normalize the meteorological disaster risk sub-coefficients of the sub-region to be assessed to obtain the meteorological disaster risk standard sub-coefficients; Obtain the sub-region area values ​​of n sub-regions to be assessed in the risk assessment coordinate system, and obtain the meteorological disaster risk coefficient of the sub-region to be assessed by combining the meteorological disaster risk standard sub-coefficient of the sub-region to be assessed.

8. A meteorological disaster risk assessment system based on the Internet of Things, used to implement the assessment method according to any one of claims 1 to 7, characterized in that: include: An acquisition module is used to obtain geographic location data of the area to be evaluated, thereby determining corresponding elevation data and historical meteorological environment data, and also to obtain meteorological forecast data; A data management module, the data management module is used to determine the continuous elevation surface and meteorological environment surface of the area to be assessed, perform data preprocessing on the data acquired by the acquisition module, perform data processing and normalization on the data output by the data management module, and construct a risk assessment coordinate system. In combination with the corresponding elevation data and historical meteorological environment data, the module determines the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, divides the area to be assessed into n sub-areas to be assessed based on the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, and manages the area to be assessed and the sub-areas to be assessed; a risk assessment module, the risk assessment module being used to obtain natural disaster resistance characteristic data of a sub-region to be assessed from elevation data corresponding to the region to be assessed, and to obtain disaster resistance infrastructure data of the sub-region to be assessed, thereby determining an initial disaster resistance coefficient of the sub-region to be assessed, and being used to sequentially determine a first sub-region to be assessed from the n sub-regions to be assessed, and simultaneously obtain historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be assessed, and then determine a homogeneity coefficient of the first sub-region to be assessed in combination with sub-regions to be assessed adjacent to the first sub-region to be assessed, and being used to determine a meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the homogeneity coefficient of the first sub-region to be assessed and the initial disaster resistance coefficient of the sub-region to be assessed, thereby determining the meteorological disaster risk coefficient of the region to be assessed; A risk warning module is used to issue a risk warning to the area to be assessed based on the meteorological disaster risk coefficient of the area to be assessed; The display module is used to present the process and results of meteorological disaster risk assessment to users.

9. The meteorological disaster risk assessment system based on the Internet of Things according to claim 8, characterized in that: The data management module includes: A data processing unit, which is used to determine the continuous elevation surface and meteorological environment surface of the area to be evaluated, perform data preprocessing on the data acquired by the acquisition module, and perform data processing and normalization on the data output by the data management module; A coordinate system management unit, the coordinate system management unit is used to construct a risk assessment coordinate system and determine a characteristic coefficient matrix for each coordinate point in the risk assessment coordinate system in combination with corresponding elevation data and historical meteorological environment data; The area management unit is used to divide the area to be evaluated into n sub-areas to be evaluated according to the characteristic coefficient matrix of each coordinate point in the risk assessment coordinate system, and manage the area to be evaluated and the sub-areas to be evaluated.

10. The meteorological disaster risk assessment system based on the Internet of Things according to claim 8, characterized in that: The risk assessment module includes: A regional disaster resistance assessment unit, configured to obtain natural disaster resistance characteristic data of a sub-region to be assessed from elevation data corresponding to the region to be assessed, and to obtain disaster resistance infrastructure data of the sub-region to be assessed, thereby determining an initial coefficient of disaster resistance capacity of the sub-region to be assessed; a regional mutual impact assessment unit, the regional mutual impact assessment unit being configured to sequentially determine a first sub-region to be assessed from the n sub-regions to be assessed, synchronously obtain historical meteorological disaster data, natural disaster resistance characteristic data, and disaster resistance infrastructure data corresponding to the first sub-region to be assessed, and then determine a homogeneity coefficient for the first sub-region to be assessed in combination with adjacent sub-regions to be assessed; The regional risk assessment unit is used to determine the meteorological disaster risk sub-coefficient of the sub-region to be assessed based on the homogeneity coefficient of the first sub-region to be assessed and the initial coefficient of the disaster resistance capacity of the sub-region to be assessed, thereby determining the meteorological disaster risk coefficient of the region to be assessed.

Images