A weather disaster risk assessment method and system

Abstract

This invention relates to the field of meteorological disaster risk assessment technology, specifically disclosing a method and system for meteorological disaster risk assessment. The method involves collecting five types of multi-source heterogeneous data, including meteorological monitoring and geographic underlying surface data, and performing standardized preprocessing. Disaster types are identified using a threshold library, and risk factor weights are dynamically allocated using an entropy-weighted hierarchical analysis coupling algorithm. A single-hazard assessment model is constructed from the dimensions of disaster hazard, vulnerability of the disaster-bearing body, and adaptability to prevention and control, calculating the basic risk value. A modified model is constructed by introducing coupling correlation coefficients, considering the superposition effect of multiple hazards to obtain an initial comprehensive risk value. Five risk levels are determined by matching dynamic thresholds based on season, geomorphology, and disaster stage, ultimately outputting a refined assessment result and updating it dynamically in real time. This invention solves the problems of traditional assessments being limited to a single dimension, having fixed parameters, and not considering disaster coupling, resulting in more accurate and real-time assessments that can provide reliable support for disaster prevention and emergency decision-making.

Description

Technical Field

[0001] This invention relates to the field of meteorological disaster risk assessment technology, specifically to a meteorological disaster risk assessment method and system. Background Technology

[0002] Frequent occurrences of various meteorological disasters, such as torrential rains, typhoons, cold waves, strong winds, frost, and droughts, pose serious threats to regional agricultural production, infrastructure, living environment, and the safety of people's lives and property. Accurate, real-time, and comprehensive meteorological disaster risk assessment is the core foundation for disaster early warning, emergency response, and disaster prevention and mitigation planning.

[0003] Current meteorological disaster risk assessment technologies generally suffer from numerous inherent flaws, severely limiting the accuracy and practicality of assessment results. Firstly, existing assessment models often rely on single meteorological monitoring data as the core basis for judgment, depending solely on basic meteorological parameters such as real-time rainfall, wind speed, and temperature for risk classification. This neglects crucial related factors such as the underlying surface environment, the attributes of regional disaster-bearing bodies, historical disaster patterns, and human prevention and control capabilities. This results in a singular assessment dimension, making it highly susceptible to omissions and misjudgments. For example, under the same rainfall conditions, the disaster risk varies greatly in mountainous areas, riverbanks, and old urban areas; a single data assessment cannot reflect the differentiated risk characteristics of different regions.

[0004] Secondly, existing assessment methods are mostly static, fixed-threshold assessment models, using uniform evaluation standards for different seasons, landforms, and disaster types, which cannot adapt to the dynamic evolution characteristics of meteorological disasters. Meteorological disasters are characterized by suddenness, stages, and cascading effects. Meteorological parameters and environmental conditions continuously change during the disaster process, making it difficult for static threshold assessments to capture dynamic fluctuations in risk in real time, resulting in problems such as delayed risk assessment and untimely early warning.

[0005] Third, existing technologies mostly assess single meteorological disasters independently, without considering the risks of multiple disasters coupled together. In real disaster scenarios, rainstorms can easily trigger landslides and urban flooding, while typhoons bring multiple disasters such as strong winds, rainstorms, and storm surges. The superposition of multiple disasters can significantly amplify disaster risks, and existing single-disaster assessment models cannot identify coupled risks, resulting in assessment results that are too lenient and difficult to support emergency decision-making in complex disaster scenarios.

[0006] Fourth, the existing assessment system has weak data fusion capabilities. Multi-source monitoring data suffers from problems such as time misalignment, inconsistent dimensions, and significant noise interference. Data is directly substituted into the model without standardized preprocessing and correlation matching, resulting in insufficient accuracy of the assessment data. At the same time, it lacks integrated capabilities for risk tracing, dynamic updates, and result visualization and push. The overall level of intelligence and automation is low, which cannot meet the application needs of refined disaster prevention and mitigation. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing meteorological disaster risk assessment technologies, such as single-dimensionality, static and rigid approaches, inability to adapt to multi-hazard coupling scenarios, low data accuracy, and insufficient intelligence. This invention provides a meteorological disaster risk assessment method and system that enables dynamic fusion of multi-source heterogeneous data, hierarchical assessment of multi-hazard coupling risks, and real-time dynamic updates of risk levels. This improves the accuracy, real-time nature, and comprehensiveness of meteorological disaster risk assessment, providing reliable data support and decision-making basis for disaster prevention and mitigation, emergency dispatch, and disaster early warning.

[0008] To achieve the above objectives, the technical solution provided by this invention is: a meteorological disaster risk assessment method, comprising the following steps: S1. Real-time acquisition and standardized preprocessing of multi-source heterogeneous data: Real-time acquisition of five major categories of core data in the target assessment area, including meteorological monitoring data, geographic underlying surface data, disaster-bearing body data, historical disaster data, and prevention and control capacity data. Cleaning, noise reduction, time-series alignment, and dimension normalization of various heterogeneous data are performed to construct a standardized meteorological disaster assessment dataset. S2. Disaster type identification and dynamic allocation of risk factor weights: Based on real-time meteorological monitoring data and combined with a disaster characteristic threshold library, the system automatically identifies the current and future predicted single or multiple meteorological disaster types in the target area. It adopts an entropy weight-analysis coupling algorithm and dynamically adjusts the weight coefficients of each risk factor in combination with the real-time environmental characteristics of the area to avoid the evaluation bias caused by fixed weights. S3. Layered construction of single-hazard risk assessment models: For the various meteorological disasters identified, a layered single-hazard risk assessment model is constructed from three dimensions: disaster hazard, vulnerability of disaster-bearing body, and regional prevention and control adaptability. The hazard index, vulnerability index, and prevention and control index corresponding to a single disaster are calculated respectively, and finally the basic risk value of a single disaster is synthesized. S4. Multi-hazard Coupling Risk Correction Calculation: Based on the basic risk value of a single hazard, a multi-hazard coupling correlation coefficient is introduced. Combining the superimposed linkage characteristics of different hazards, a multi-hazard coupling risk correction model is constructed to superimpose and correct the basic risk value of a single hazard, eliminating the limitations of single assessment and obtaining the initial comprehensive risk value. S5. Dynamic Threshold Matching and Risk Level Determination: Construct a dynamic risk threshold system that is divided by season, topography, and time period. Based on the seasonal attributes, topography, and disaster evolution stage of the target area, match the corresponding dynamic threshold interval, classify and correct the initial comprehensive risk value, and determine the final meteorological disaster risk level. S6. Risk Result Output and Dynamic Update: Outputs refined gridded risk assessment results for the target area in real time, generates risk distribution maps, risk source tracing reports, and emergency response plans. At the same time, iterates and updates monitoring data and risk assessment results in real time according to the preset update frequency to achieve dynamic tracking and assessment of disaster risks.

[0009] Furthermore, in step S1, the multi-source heterogeneous data specifically includes: Meteorological monitoring data includes real-time rainfall, hourly wind speed, temperature, humidity, air pressure, visibility, snowfall, and weather forecasts for the next 1-72 hours; geographic underlying surface data includes topographic elevation, slope, aspect, water system distribution, vegetation coverage, soil moisture, and geological type data; disaster-bearing body data includes regional population density, building distribution, transportation network, water conservancy facilities, agricultural planting areas, and industrial and mining enterprise distribution data; historical disaster data includes the frequency, impact range, degree of loss, and evolution pattern of similar meteorological disasters in the target area over the past 10 years; and prevention and control capacity data includes regional flood control material reserves, emergency team configuration, drainage facility capacity, and disaster early warning coverage data.

[0010] The data preprocessing specifically includes: removing abnormal noise from the monitoring data through mean filtering algorithm, aligning the time sequence of data with different sampling frequencies through timestamp calibration, and mapping the raw data of different dimensions and units to the 0-1 standard range through range standardization algorithm to complete the data standardization process.

[0011] Furthermore, in step S2, the disaster feature threshold library pre-stores the basic judgment thresholds and related feature parameters of various meteorological disasters, covering the identification standards of seven core meteorological disasters: rainstorm, typhoon, cold wave, strong wind, frost, drought, and blizzard. The entropy weight-analysis coupling algorithm mines the weight features of the data itself through the objective entropy weight method, and integrates disaster prevention experience weights through the analysis method, dynamically balancing objective data and subjective experience to achieve adaptive adjustment of risk factor weights.

[0012] Furthermore, in step S3, the disaster risk index is calculated based on meteorological intensity, disaster duration, and disaster impact range, representing the destructive capacity of the meteorological disaster itself; the disaster-bearing body vulnerability index is calculated based on the exposure and sensitivity of disaster-bearing bodies such as population, buildings, and industries, representing the susceptibility of regional disaster losses; and the regional prevention and control adaptability index is calculated based on infrastructure protection capacity, emergency response capacity, and basic disaster resistance conditions, representing the region's ability to resist disasters and reduce losses.

[0013] Furthermore, in step S4, the multi-hazard coupling correlation coefficient is preset with graded values ​​based on the linkage correlation degree of different hazards. The higher the correlation degree, the larger the coupling coefficient. High coupling coefficients are used for strongly correlated hazards such as rainstorms and landslides, strong winds and typhoons, and cold waves and frosts. The baseline coupling coefficient is used for independent hazards without obvious correlation. The multi-hazard coupling risk correction model obtains the initial comprehensive risk value by superimposing the risk values ​​of each single hazard and the coupling linkage increment, which accurately reflects the multi-hazard superposition amplification effect.

[0014] Furthermore, in step S5, the dynamic risk threshold system divides the threshold range into multiple dimensions based on the four seasons, four types of landforms (mountainous / plain / hilly / waterside), and four evolutionary stages (disaster incubation / occurrence / continuation / regression). It abandons the traditional fixed threshold and achieves differentiated and precise judgment of risk levels. The risk levels are divided into five levels: low risk, medium risk, relatively high risk, high risk, and extremely high risk.

[0015] The present invention also provides a meteorological disaster risk assessment system for implementing the above-mentioned meteorological disaster risk assessment method, including a data acquisition module, a data preprocessing module, a disaster identification and weight configuration module, a single disaster risk calculation module, a multi-disaster coupling correction module, a risk level determination module, a result output and update module, and a storage database; The data acquisition module is used to collect meteorological monitoring data, geographic underlying surface data, disaster-bearing body data, historical disaster data, and prevention and control capability data of the target assessment area in real time. The data preprocessing module is used to clean, denoise, align time series, and normalize dimensions of the collected multi-source heterogeneous data to generate a standardized evaluation dataset. The disaster identification and weight configuration module is used to identify the types of meteorological disasters in the target area and dynamically allocate the weights of each risk factor through the entropy weight-analysis coupling algorithm. The single-hazard risk calculation module is used to calculate the basic risk value of various single meteorological disasters from three dimensions: disaster hazard, vulnerability of the disaster-bearing body, and regional prevention and control adaptability. The multi-hazard coupling correction module is used to superimpose and correct the basic risk value of a single hazard based on the hazard coupling correlation coefficient to obtain an initial comprehensive risk value. The risk level determination module is used to match the dynamic risk threshold system, correct and determine the final meteorological disaster risk level; The result output and update module is used to output refined risk assessment results, generate visual charts and emergency suggestions, and dynamically update assessment data and results in real time. The storage database is used to store raw collected data, standardized datasets, disaster threshold databases, weight parameters, and assessment result data, providing data support for model iteration and risk tracing.

[0016] The advantages of this invention compared to the prior art are: This invention adopts a multi-source heterogeneous data fusion assessment system, which breaks through the limitations of existing single meteorological data assessment. It integrates data from five dimensions: meteorology, geography, disaster-bearing bodies, historical disasters, and prevention and control capabilities. It conducts risk assessment from all aspects, including the causes of disasters, the affected objects, and the resilience capabilities, and completely solves the problems of one-sided assessment dimensions and large biases in traditional assessments, thus greatly improving the comprehensiveness of the assessment.

[0017] This invention adopts a dynamic weight allocation and dynamic threshold determination mechanism, abandoning the traditional static evaluation mode with fixed weights and fixed thresholds. It can adaptively adjust the evaluation parameters according to the regional topography, season, and disaster evolution status, adapting to the dynamic evolution characteristics of meteorological disasters, effectively avoiding the defects of static evaluation such as lag and misjudgment, and improving the real-time performance and accuracy of evaluation.

[0018] This invention introduces a multi-hazard coupling risk correction model, which fully considers the amplification effect of multiple hazards superimposed and linked, and breaks through the limitations of traditional single-hazard independent assessment. It can accurately identify coupled disaster risks in complex scenarios, fit the actual disaster occurrence scenarios, and the assessment results are more in line with actual disaster prevention needs.

[0019] This invention realizes intelligent and automated operation of the entire process, including data preprocessing, disaster identification, risk calculation, level determination, result output, and dynamic updating. It is also equipped with risk tracing and emergency suggestion generation functions, which can complete refined gridded risk assessment without manual intervention, greatly improving assessment efficiency. It can provide real-time, accurate, and comprehensive decision support for grassroots disaster prevention, emergency dispatch, and disaster early warning, and has strong practicality and adaptability. Attached Figure Description

[0020] Figure 1 This is a flowchart of a meteorological disaster risk assessment method according to the present invention.

[0021] Figure 2 This is a system block diagram of a meteorological disaster risk assessment system according to the present invention. Detailed Implementation

[0022] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present invention.

[0023] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0024] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0025] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0026] The following detailed description of a meteorological disaster risk assessment method and system according to the present invention, with reference to the accompanying drawings, provides further insight.

[0027] Combined with appendix Figure 1-2 The specific implementation process of the meteorological disaster risk assessment method and system of the present invention is as follows: A meteorological disaster risk assessment method is applied to a scenario of refined disaster prevention and mitigation assessment for regional meteorological disasters. The specific implementation steps are as follows: Step S1: Real-time acquisition and standardized preprocessing of multi-source heterogeneous data.

[0028] The system collects five core categories of data in real time for the target assessment area, including: meteorological monitoring data: real-time rainfall, hourly wind speed, temperature, humidity, air pressure, visibility, snowfall, and 72-hour weather forecast data; geographic underlying surface data: topographic elevation, slope, aspect, water system distribution, vegetation coverage, soil moisture, and geological type data; disaster-bearing body data: regional population density, building distribution, transportation network, water conservancy facilities, agricultural planting areas, and industrial and mining enterprise distribution data; historical disaster data: frequency, impact range, degree of loss, and evolution pattern data of similar meteorological disasters in the past 10 years; and prevention and control capacity data: flood control material reserves, emergency team configuration, drainage facility capacity, and disaster early warning coverage data.

[0029] The collected raw data undergoes standardized preprocessing: mean filtering is used to filter out abnormal sensor values ​​and sudden noise data, eliminating invalid data; a unified timestamp calibration method is used to align the data at different sampling frequencies (seconds, minutes, hours, etc.) to ensure the consistency of the data's time dimension; and a range standardization algorithm is used to map all raw data of different dimensions and orders of magnitude to the 0-1 standard range, eliminating the influence of dimensional differences, and finally constructing a standardized and structured meteorological disaster assessment dataset.

[0030] Since the dimensions and orders of magnitude of each indicator differ, this invention uses the range standardization method to normalize the positive and negative indicators and uniformly map them to the interval [0, 1].

[0031] Standardized formula for positive indicators (higher values ​​indicate higher risk, such as rainfall, wind speed, and population density): ; Standardized formula for negative indicators (higher values ​​indicate lower risk, such as prevention and control capabilities, drainage efficiency, and material reserves): ; In the formula: For the first The first sample The original values ​​of the indicators; These are standardized values; , The first The maximum and minimum values ​​of each indicator.

[0032] Step S2: Disaster type identification and dynamic allocation of risk factor weights.

[0033] The system retrieves a pre-set disaster characteristic threshold database, which contains basic judgment thresholds and associated characteristic parameters for seven core meteorological disasters: rainstorms, typhoons, cold waves, strong winds, frost, drought, and blizzards. Real-time meteorological monitoring data is compared with the parameters in the threshold database to automatically identify single or multiple coupled disaster types currently occurring in the target area and predicted for the next 72 hours.

[0034] The entropy weight-analysis coupled algorithm is used for weight allocation: the objective weight of each risk factor is calculated based on real collected data using the entropy weight method, and the subjective weight is set by combining industry disaster prevention and mitigation experience and regional disaster characteristics using the analysis method. The two types of weights are coupled and integrated, and the weight coefficients of all risk factors are adaptively and dynamically adjusted by combining the real-time topography, meteorology and disaster-bearing body distribution characteristics of the region, so as to avoid the assessment bias caused by fixed weights.

[0035] The objective weights of the entropy weight method are calculated as follows: Calculate the first Weighting of each indicator: ; Calculate the information entropy of the indicator: ,Regulation hour, ; Calculate objective entropy weight: ; AHP Subjective Weights: Subjective weights are obtained by constructing a judgment matrix through expert scoring. ; Coupled combination weight formula: ; in: This is the weighted fusion coefficient, with a preferred value of 0.5; For the final dynamic coupling weight, satisfying .

[0036] Step S3: Construct a single-hazard risk assessment model in layers and calculate the basic risk value.

[0037] For each identified meteorological hazard, an independent single-hazard risk assessment model is constructed layered from three core dimensions: Disaster risk dimension: Combining three major indicators—disaster meteorological intensity, duration, and spatial impact range—a disaster risk index is calculated by weighting the indicators to quantify the destructive capacity of the disaster itself. Disaster Risk Index: ; Vulnerability dimension of disaster-bearing bodies: The spatial exposure and disaster sensitivity of various disaster-bearing bodies such as population, buildings, transportation, agriculture, industry and mining in the region are statistically analyzed, and the vulnerability index of disaster-bearing bodies is calculated to quantify the degree of vulnerability to disasters in the region; Vulnerability index of disaster-bearing body: ; Regional prevention and control adaptability dimension: Based on the regional infrastructure protection level, emergency team response capabilities, material reserve guarantee capabilities, and disaster early warning coverage effect, the prevention and control index is calculated to quantify the region's ability to resist disasters and reduce losses; Regional epidemic prevention and control adaptability index: ; in: These represent the number of indicators for risk, vulnerability, and prevention / control; each weight is a normalized coupled weight.

[0038] By weighting and synthesizing three types of indices, the basic risk value for each type of single meteorological disaster is obtained.

[0039] Combining disaster damage characteristics, vulnerability of affected bodies, and regional resilience, a quantifiable single-hazard risk formula is constructed: ;in: This is the basic risk value for a single disaster type; The dynamic weights for hazard, vulnerability, and preventability are respectively satisfied with the following constraints: ; The prevention and control index adopts a negative correction logic; the stronger the prevention and control capability, the lower the basic risk value, which is in line with the actual disaster prevention law.

[0040] Step S4: Multi-hazard coupling risk correction calculation.

[0041] Based on the interconnected characteristics of disasters, a multi-hazard coupling correlation coefficient is preset at different levels: for strongly correlated disaster combinations such as rainstorm-landslide, strong wind-typhoon, and cold wave-frost, a high coupling correlation coefficient is configured to highlight the amplifying effect of disaster superposition; for independent disasters without obvious linkage characteristics, a baseline coupling correlation coefficient is configured to avoid over-correction. The coupling correlation coefficients are then substituted into the multi-hazard coupling risk correction model to superimpose and couple the basic risk values ​​of each single hazard, eliminating the limitations of independent assessment of single hazards and obtaining the initial comprehensive risk value of the target area.

[0042] Assuming regional coexistence For meteorological disasters, the basic risk of each individual disaster type is: ,disaster With disaster Coupling correlation coefficient is The initial overall risk value is: ; Among them: the first term is the linear superposition value of the risks of each single disaster; the second term is the multi-disaster coupling increment term, which is used to characterize the disaster linkage amplification effect; The preset coupling coefficient is set to 0.2~0.5 for strongly correlated disasters, 0~0.2 for weakly correlated disasters, and 0 for uncorrelated disasters.

[0043] Step S5: Dynamic threshold matching and risk level determination.

[0044] A pre-constructed multi-dimensional dynamic risk threshold system is invoked, covering four seasons (spring, summer, autumn, and winter), four landform types (mountains, plains, hills, and waterfront), and differentiated threshold ranges for four evolution stages of disaster: incubation, occurrence, duration, and dissipation. Based on the target area's current season, topographical features, and real-time disaster evolution stage, the corresponding dynamic threshold range is precisely matched, and the initial comprehensive risk value is graded and corrected. Finally, the risk is uniformly classified into five levels: low risk, medium risk, relatively high risk, high risk, and extremely high risk, determining the final meteorological disaster risk level for the region.

[0045] This invention abandons fixed thresholds and constructs a three-dimensional dynamic threshold function based on season, landform, and evolution stages. This function is then used to normalize and correct the comprehensive risk value, resulting in a final standardized risk value. : ; in: Based on season landforms Disaster Phase The dynamic upper and lower threshold values.

[0046] in accordance with The numerical range is divided into five risk levels: low risk [0, 0.2), medium risk [0.2, 0.4), relatively high risk [0.4, 0.6), high risk [0.6, 0.8], and extremely high risk [0.8, 1.0].

[0047] Step S6: Output and dynamic update of risk results.

[0048] Based on the assessment results, refined gridded risk assessment data for the target area is generated, automatically creating a spatial distribution map of the entire risk area. This identifies core risk triggers, high-risk areas, and weak prevention and control links, generating a risk tracing report. Combining regional prevention and control capabilities with disaster characteristics, targeted emergency response, early warning and control, and material allocation recommendations are provided. Monitoring data is collected and updated in real-time at a preset hourly update frequency, repeating the above assessment process to continuously update the risk assessment results, achieving dynamic tracking and assessment of meteorological disaster risks throughout their entire lifecycle.

[0049] The present invention also provides a meteorological disaster risk assessment system for implementing the above-mentioned meteorological disaster risk assessment method, including: The data acquisition module is used to collect meteorological monitoring data, geographic underlying surface data, disaster-bearing body data, historical disaster data, and prevention and control capability data of the target assessment area in real time. The data preprocessing module is used to clean, denoise, align time series, and normalize dimensions of the collected data to generate a standardized evaluation dataset. The disaster identification and weight configuration module is used to identify the types of meteorological disasters in the target area and dynamically allocate the weights of each risk factor through the entropy weight-analysis coupling algorithm. The single-hazard risk calculation module is used to calculate the basic risk value of various single meteorological disasters from three dimensions: disaster hazard, vulnerability of the disaster-bearing body, and regional prevention and control adaptability. The multi-hazard coupling correction module is used to superimpose and correct the basic risk value of a single hazard based on the hazard coupling correlation coefficient to obtain the initial comprehensive risk value; The risk level determination module is used to match the dynamic risk threshold system, correct and determine the final meteorological disaster risk level; The results output and update module is used to output refined risk assessment results, generate visual charts and emergency suggestions, and dynamically update assessment data and results in real time. The storage database is used to store raw collected data, standardized datasets, disaster threshold libraries, weight parameters, and assessment result data.

[0050] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for assessing the risk of a meteorological disaster, characterized in that, Includes the following steps: S1. Real-time acquisition and standardized preprocessing of multi-source heterogeneous data: Real-time acquisition of meteorological monitoring data, geographic underlying surface data, disaster-bearing body data, historical disaster data, and prevention and control capability data of the target assessment area, followed by preprocessing such as cleaning, noise reduction, time-series alignment, and dimension normalization to construct a standardized meteorological disaster assessment dataset; S2. Disaster type identification and dynamic allocation of risk factor weights: Based on real-time meteorological monitoring data and combined with a preset disaster characteristic threshold library, the system automatically identifies the current and future predicted single or multiple meteorological disaster types in the target area. It adopts an entropy weight-analysis coupling algorithm and dynamically adjusts the weight coefficients of each risk factor in combination with the real-time environmental characteristics of the area. S3. Construct a hierarchical single-hazard risk assessment model and calculate the basic risk value: For the various meteorological disasters identified, construct a hierarchical single-hazard risk assessment model from three dimensions: disaster hazard, vulnerability of the disaster-bearing body, and regional prevention and control adaptability. Calculate the hazard index, vulnerability index, and prevention and control index corresponding to a single disaster, and synthesize the basic risk value of the single disaster. S4. Multi-hazard coupling risk correction calculation: Based on the basic risk values ​​of each single hazard, a multi-hazard coupling correlation coefficient is introduced. Combining the superposition and linkage characteristics of different disasters, the basic risk values ​​of the single hazards are superimposed and corrected through the multi-hazard coupling risk correction model to obtain the initial comprehensive risk value. S5. Dynamic Threshold Matching and Risk Level Determination: Construct a dynamic risk threshold system that is divided by season, topography, and time period. Based on the seasonal attributes, topography, and disaster evolution stage of the target area, match the corresponding dynamic threshold interval, classify and correct the initial comprehensive risk value, and determine the final meteorological disaster risk level. S6. Risk Result Output and Dynamic Update: Outputs refined gridded risk assessment results for the target area in real time, generates risk distribution maps, risk source tracing reports and emergency response plans, and iteratively updates monitoring data and risk assessment results according to a preset update frequency to achieve dynamic tracking and assessment of disaster risks.

2. The method for weather disaster risk assessment according to claim 1, characterized in that: In step S1, the meteorological monitoring data includes real-time rainfall, hourly wind speed, temperature, humidity, air pressure, visibility, snowfall, and weather forecast data for the next 1-72 hours; the geographic underlying surface data includes topographic elevation, slope, aspect, water system distribution, vegetation coverage, soil moisture, and geological type data; the disaster-bearing body data includes regional population density, building distribution, transportation network, water conservancy facilities, agricultural planting areas, and industrial and mining enterprise distribution data; the historical disaster data includes the frequency of occurrence, impact range, degree of loss, and evolution pattern of similar meteorological disasters in the target area over the past 10 years; and the prevention and control capability data includes regional flood control material reserves, emergency team configuration, drainage facility capacity, and disaster early warning coverage data.

3. The method of claim 2, wherein: In step S1, the preprocessing specifically includes: removing abnormal noise from the monitoring data through a mean filtering algorithm, aligning the time sequence of data with different sampling frequencies through timestamp calibration, and mapping the original data with different dimensions and scales to the 0-1 standard range through a range standardization algorithm to complete the data standardization process.

4. The method for weather disaster risk assessment according to claim 3, characterized in that: In step S2, the disaster feature threshold library pre-stores the basic judgment thresholds and associated feature parameters of seven core meteorological disasters: rainstorm, typhoon, cold wave, strong wind, frost, drought, and blizzard; the entropy weight-hierarchical analysis coupling algorithm combines objective data weights and disaster prevention experience weights to adaptively and dynamically adjust the risk factor weights.

5. The method for weather disaster risk assessment according to claim 4, characterized in that: In step S3, the disaster risk index is calculated based on meteorological intensity, disaster duration, and disaster impact range; the disaster-bearing body vulnerability index is calculated based on the exposure and sensitivity of population, buildings, and industries; and the regional prevention and control adaptability index is calculated based on infrastructure protection capabilities, emergency response capabilities, and basic disaster resistance conditions.

6. The meteorological disaster risk assessment method according to claim 5, characterized in that: In step S4, the multi-hazard coupling correlation coefficient is assigned a value according to the degree of linkage correlation of different hazards. A high coupling coefficient is configured for strongly correlated hazards such as rainstorm and landslide, strong wind and typhoon, and cold wave and frost. A baseline coupling coefficient is configured for independent hazards with no obvious correlation. The multi-hazard superposition and amplification effect is reflected through the coupling correlation coefficient.

7. The meteorological disaster risk assessment method according to claim 6, characterized in that: In step S5, the dynamic risk threshold system covers multi-dimensional threshold ranges for the four seasons, four types of landforms (mountainous / plain / hilly / waterside), and four evolution stages (disaster incubation / occurrence / continuation / regression), abandoning the fixed threshold judgment mode.

8. The meteorological disaster risk assessment method according to claim 7, characterized in that: In step S5, the risk level is divided into five levels: low risk, medium risk, relatively high risk, high risk, and extremely high risk.

9. A meteorological disaster risk assessment system, characterized in that, The method for implementing the meteorological disaster risk assessment method according to any one of claims 1-8 includes: The data acquisition module is used to collect meteorological monitoring data, geographic underlying surface data, disaster-bearing body data, historical disaster data, and prevention and control capability data of the target assessment area in real time. The data preprocessing module is used to clean, denoise, align time series, and normalize dimensions of the collected data to generate a standardized evaluation dataset. The disaster identification and weight configuration module is used to identify the types of meteorological disasters in the target area and dynamically allocate the weights of each risk factor through the entropy weight-analysis coupling algorithm. The single-hazard risk calculation module is used to calculate the basic risk value of various single meteorological disasters from three dimensions: disaster hazard, vulnerability of the disaster-bearing body, and regional prevention and control adaptability. The multi-hazard coupling correction module is used to superimpose and correct the basic risk value of a single hazard based on the hazard coupling correlation coefficient to obtain the initial comprehensive risk value; The risk level determination module is used to match the dynamic risk threshold system, correct and determine the final meteorological disaster risk level; The results output and update module is used to output refined risk assessment results, generate visual charts and emergency suggestions, and dynamically update assessment data and results in real time. The storage database is used to store raw collected data, standardized datasets, disaster threshold libraries, weight parameters, and assessment result data.

Images