Three-proof emergency command method and system based on meteorological big data

Abstract

The application belongs to the technical field of emergency management, and discloses a three-proofing emergency command method and system based on meteorological big data; including: collecting disaster meteorological correlation data, identifying meteorological evolution modes corresponding to various disasters; collecting current meteorological data and environmental state data in real time, generating a three-proofing risk progression curve; obtaining meteorological forecast data, calculating the meteorological similarity between each meteorological evolution mode, combining the three-proofing risk progression curve, predicting the disaster risk elements of various disasters, and determining the intervention disaster; according to the disaster risk elements of the intervention disaster, analyzing the feasible time period and expected effect of different weather modification operations; intelligently identifying the best operation type and the best time window, and intelligently scheduling three-proofing emergency resources to generate a preventive resource deployment scheme; the application constructs a full-chain intelligent three-proofing emergency command system, and breaks through the technical bottlenecks of traditional emergency systems in terms of early warning timeliness, intervention capacity, resource allocation and the like.

Description

Technical Field

[0001] This invention relates to the field of emergency management technology, and more specifically, to a method and system for emergency command of three-defense (flood, drought, and landslide) based on meteorological big data. Background Technology

[0002] Traditional disaster prevention and mitigation emergency management mainly relies on emergency response after a disaster occurs, using limited data from meteorological monitoring stations for simple threshold warnings. When monitoring indicators exceed preset values, emergency plans are activated. This passive emergency mode suffers from problems such as poor warning timeliness, delayed resource allocation, and limited disaster prevention measures. Although meteorological monitoring technology has been developing in recent years, accumulating massive amounts of historical and real-time meteorological data, existing systems only use this big meteorological data for weather forecasting and simple disaster warnings, failing to delve into the intrinsic correlation between meteorological evolution patterns and disaster occurrence, resulting in a lack of scientific basis for disaster prevention decisions.

[0003] Existing disaster prevention, mitigation, and hazard mitigation emergency command systems suffer from several technical deficiencies: First, they lack the ability to dynamically monitor the disaster formation process, making it impossible to quantitatively assess the state evolution and risk accumulation process of disaster-bearing bodies (such as forest fuels, watershed soil, and buildings) under meteorological conditions. Second, they lack proactive intervention mechanisms, only initiating emergency responses when a disaster is about to occur or has already occurred, missing the optimal opportunity to eliminate potential disaster risks in advance through means such as weather modification. Third, they fail to fully utilize the value of meteorological big data, unable to predict disaster risks by analyzing the similarity between historical disaster cases and current meteorological conditions. Fourth, emergency resource allocation lacks foresight, often only being urgently deployed after a disaster occurs, resulting in delays in rescue efforts and waste of resources. Therefore, there is an urgent need for an intelligent disaster prevention, mitigation, and hazard mitigation emergency command system that can fully leverage the value of meteorological big data, achieve dynamic monitoring of disaster risks, and support proactive intervention decisions.

[0004] In view of this, the present invention proposes a method and system for emergency command of three-defense based on meteorological big data to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the existing technology and achieve the above objectives, the present invention provides the following technical solution: a method for emergency command of disaster prevention, mitigation, and hazard mitigation based on meteorological big data, comprising:

[0006] Collect meteorological data related to disasters, identify meteorological evolution patterns corresponding to various disasters in the meteorological data related to disasters, and build a database of meteorological precursor patterns for disaster prevention, mitigation, and disaster relief.

[0007] Real-time collection of current meteorological and environmental data; assessment of the progressive evolution of vulnerability of disaster-bearing bodies corresponding to various disasters; generation of three-defense risk progression curves.

[0008] Acquire meteorological forecast data, calculate the meteorological similarity between the meteorological forecast data and various meteorological evolution models in the disaster prevention and mitigation meteorological precursor model library, and combine the disaster prevention and mitigation risk progression curve to predict the disaster risk factors of various disasters, and determine the disaster intervention based on the disaster risk factors;

[0009] Based on the disaster risk factors of disaster intervention, the feasible time periods and expected effects of different artificial weather modification operations corresponding to disaster intervention are analyzed, and the best operation type and the best time window are intelligently identified based on the feasible time periods and expected effects.

[0010] Based on the optimal operation type and time window for disaster intervention, the system intelligently schedules emergency resources for disaster prevention, mitigation, and hazard mitigation, generates preventive resource deployment plans, and directs various disaster prevention forces to execute these plans.

[0011] Furthermore, methods for collecting disaster meteorological correlation data include:

[0012] Acquire historical meteorological data covering the target area, including meteorological observation records collected at various historical moments in the target area; acquire disaster archive data corresponding to the same time period as the historical meteorological data, including records of various disaster events that have occurred in the target area in the past;

[0013] Based on the occurrence time in each disaster event record, the corresponding associated meteorological sequence is extracted from historical meteorological data; each associated meteorological sequence is decomposed into meteorological elements to obtain a meteorological element sequence corresponding to each meteorological element; each meteorological element sequence is associated with the corresponding disaster event record to form disaster meteorological association data.

[0014] Furthermore, methods for constructing a database of meteorological precursor models for three-defense systems include:

[0015] Based on the disaster type of each disaster event record in the disaster meteorological association data, the disaster meteorological association data is divided into fire association dataset, flood association dataset and wind disaster association dataset;

[0016] For fire-related datasets, meteorological evolution patterns of fires are extracted and used as fire precursor patterns.

[0017] For flood-related datasets, meteorological evolution patterns of floods are extracted and used as precursor patterns for floods.

[0018] For wind disaster-related datasets, meteorological evolution patterns of wind disasters are extracted and used as precursor patterns for wind disasters.

[0019] By summarizing the precursor patterns of fires, floods, and winds, a database of meteorological precursor patterns for disaster prevention, mitigation, and hazard mitigation is constructed.

[0020] Furthermore, methods for generating the three-defense risk progression curve include:

[0021] Current meteorological data includes meteorological observation records collected in the target area at the current moment; environmental status data includes fire status data, flood status data, and wind disaster status data.

[0022] Among them, fire status data includes combustible material moisture content and surface dryness; flood status data includes soil moisture content, river water level and drainage network load rate; wind disaster status data includes building wind resistance rating, number of temporary facilities and tree fall risk index;

[0023] Based on current meteorological and fire status data, assess the progressive evolution of the vulnerability of the fire-affected body, form a progressive sequence of fire vulnerability, and plot a progressive curve of fire risk.

[0024] Based on current meteorological and flood status data, assess the progressive evolution of the vulnerability of flood-affected bodies, form a progressive sequence of flood vulnerability, and plot a progressive curve of flood risk.

[0025] Based on current meteorological data and wind disaster status data, assess the progressive evolution of the vulnerability of the corresponding disaster-bearing bodies, form a progressive sequence of wind disaster vulnerability, and plot a progressive curve of wind disaster risk.

[0026] By integrating the fire risk progression curve, the flood risk progression curve, and the wind risk progression curve, a three-defense risk progression curve is generated.

[0027] Furthermore, methods for predicting disaster risk factors for various disasters include:

[0028] The meteorological forecast data includes meteorological forecast records for the target area at various future times. From the meteorological forecast data, fire forecast vectors, flood forecast vectors, and wind forecast vectors are extracted in sequence, and the meteorological similarity of fire forecast vectors, flood forecast vectors, and wind forecast vectors is calculated.

[0029] The fire risk progression curve is extrapolated to predict the progression value at each future moment, forming a fire prediction progression curve. The moment when the progression value in the fire prediction progression curve first reaches the preset fire progression threshold is obtained and marked as the fire critical moment. The remaining time of the fire is calculated based on the fire critical moment and the current moment. The risk evolution rate is calculated based on the fire risk progression curve. The fire meteorological similarity and risk evolution rate are weighted and summed based on the preset fire risk weights to obtain the fire occurrence probability. The fire impact range is obtained based on the fire precursor pattern corresponding to the fire meteorological similarity. The remaining time of the fire, the fire occurrence probability, and the fire impact range are integrated to form fire risk elements.

[0030] Based on the prediction methods for fire risk factors, the disaster risk factors for floods and wind disasters are predicted separately and used as the risk factors for floods and wind disasters, respectively.

[0031] Furthermore, methods for determining disaster interventions based on disaster risk factors include:

[0032] For each group of disaster risk factors, the necessity of intervention is determined sequentially. A fire intervention threshold is preset, and the fire risk factors are compared with the fire intervention threshold. The fire intervention threshold includes a fire time threshold and a fire probability threshold. If the remaining time of the fire is less than or equal to the fire time threshold, and the probability of the fire occurring is greater than or equal to the fire probability threshold, then the fire is determined to require intervention. If the remaining time of the fire is greater than the fire time threshold, or the probability of the fire occurring is less than the fire probability threshold, then the fire is determined not to require intervention. Using the same determination method, the necessity of intervention for floods and winds is determined separately based on the flood risk factors and wind risk factors.

[0033] All disaster types identified as requiring intervention are compiled into an intervention disaster list. If multiple disaster types exist in several disaster prevention lists, the intervention urgency corresponding to each disaster type in the intervention disaster list is calculated. The disaster types in the intervention disaster list are sorted in descending order according to the intervention urgency to generate an intervention disaster sequence. According to the ascending order of the intervention disaster sequence, each disaster type is marked as an intervention disaster and assigned a decreasing numerical label as the disaster priority for each disaster type.

[0034] Furthermore, methods for analyzing the feasible timing and expected effects of different weather modification operations in response to disaster intervention include:

[0035] Based on the disaster type of the intervention, a set of candidate operation types is obtained from a pre-built operation type library. The set of candidate operation types includes the candidate operation type and the operation implementation conditions. The candidate operation type is the artificial weather modification operation type. Meteorological forecast records for each future time are extracted from the meteorological forecast data. Each future time meteorological forecast record is compared with each operation implementation condition. If the parameters in the meteorological forecast record all meet the same operation implementation conditions, the corresponding future time is taken as the feasible time for the corresponding candidate operation type. The feasible times corresponding to each candidate operation type are summarized to form the feasible time period corresponding to each candidate operation type.

[0036] Different numerical labels are assigned to different candidate operation types and marked as operation labels. The disaster risk factors of the disaster to be intervened for each candidate operation type, the meteorological forecast record at a feasible time, and the operation label are combined to obtain multiple sets of change prediction sets. Each set of change prediction sets is input into the trained change prediction model to predict the meteorological change set after each intervention disaster is applied by each candidate operation type at different feasible times, and this is used as the expected effect of each candidate operation type. The meteorological change set includes temperature change value, relative humidity change value, wind speed change value, and precipitation change value.

[0037] Furthermore, methods for intelligently identifying the optimal job type and optimal time window include:

[0038] The feasible time periods for each candidate operation type are compared with the remaining time of the corresponding disaster; the remaining time of the disaster includes the remaining time of fire, flood and wind disasters; if at least one feasible moment in the feasible time period is within the remaining time of the corresponding disaster, the corresponding candidate operation type is taken as the valid operation type.

[0039] Based on the expected effects of each effective operation type, calculate the risk reduction rate of each effective operation type for the corresponding disaster at different feasible times; calculate the time period score of each effective operation type based on the feasible time period of each effective operation type; calculate the benefit score of each effective operation type based on the risk reduction rate, and combine it with the corresponding time period score to calculate the comprehensive operation score of each effective operation type; calculate the weighted sum of the benefit score and the time period score based on the preset scoring weights to obtain the comprehensive operation score; compare the comprehensive operation scores of the effective operation types corresponding to the same disaster, and select the effective operation type with the highest comprehensive operation score as the best operation type for the corresponding disaster.

[0040] Based on the feasible time periods for each optimal job type, identify the time intervals consisting of consecutive feasible moments and mark them as candidate time windows; for each candidate time window, calculate the corresponding window optimization score in sequence; compare the window optimization scores of the candidate time windows corresponding to the same optimal job type, and select the candidate time window with the highest window optimization score as the optimal time window for the corresponding optimal job type.

[0041] Furthermore, methods for generating preventative resource deployment plans include:

[0042] Based on the disaster type and optimal operation type of each intervention disaster, obtain the emergency resource demand list corresponding to each intervention disaster from the pre-built resource allocation library; obtain the current schedulable emergency resource status information; and match resource supply and demand according to the disaster priority of the intervention disaster based on the emergency resource demand list and emergency resource status information; determine the schedulable resource list for each intervention disaster based on the matching results. The schedulable resource list includes the operational equipment, personnel, and materials to be scheduled, and is collectively referred to as schedulable resources.

[0043] Determine the center location of the target area and obtain the current location and dispatch speed of each dispatch resource from the emergency resource status information; calculate the resource dispatch path corresponding to each dispatch resource based on the current location of each dispatch resource and the center location of the target area; calculate the dispatch departure time of each dispatch resource based on the length of the corresponding resource dispatch path and the dispatch speed; summarize the list of dispatch resources for the same intervention disaster and the dispatch departure time of each dispatch resource to generate an emergency resource dispatch plan for each intervention disaster; summarize the resource dispatch plans for each intervention disaster to generate a preventive resource deployment plan.

[0044] The three-defense emergency command system based on meteorological big data, implementing the aforementioned three-defense emergency command method based on meteorological big data, includes:

[0045] The meteorological mining module is used to collect disaster meteorological correlation data, identify meteorological evolution patterns corresponding to various disasters in the disaster meteorological correlation data, and build a database of meteorological precursor patterns for disaster prevention, mitigation, and disaster relief.

[0046] The risk monitoring module is used to collect current meteorological and environmental data in real time, assess the progressive evolution of the vulnerability of disaster-bearing bodies corresponding to various disasters, and generate a progressive curve of three-defense risk.

[0047] The disaster assessment module is used to acquire meteorological forecast data, calculate the meteorological similarity between the meteorological forecast data and various meteorological evolution models in the disaster prevention and mitigation meteorological precursor model library, and combine the disaster prevention and mitigation risk progression curve to predict the disaster risk factors of various types of disasters, and determine the disaster intervention based on the disaster risk factors.

[0048] The intervention identification module is used to analyze the feasible time periods and expected effects of different artificial weather modification operations corresponding to the disaster intervention based on the disaster risk factors of the disaster intervention, and intelligently identify the best operation type and the best time window based on the feasible time periods and expected effects.

[0049] The emergency command module is used to intelligently dispatch NBC (nuclear, biological, and chemical) emergency resources based on the optimal operation type and time window for disaster intervention, generate preventive resource deployment plans, and command various disaster prevention forces to execute the preventive resource deployment plans.

[0050] The technical effects and advantages of the emergency command method and system for disaster prevention, mitigation, and calamity control based on meteorological big data in this invention are as follows:

[0051] By deeply exploring the inherent correlation between historical disasters and meteorological evolution, a database of meteorological precursor models for disaster prevention, mitigation, and hazard mitigation is constructed. Combined with dynamic quantitative assessment of the vulnerability of disaster-bearing bodies, a progressive risk curve for disaster prevention, mitigation, and hazard mitigation is generated, realizing a paradigm shift from traditional "post-disaster emergency response" to "pre-disaster prevention." Innovatively, by calculating the similarity between meteorological forecast data and precursor models, the probability of disaster occurrence, remaining time, and scope of impact are accurately predicted. Based on deep learning models, the feasible time periods and expected effects of different artificial weather modification operations are analyzed, and the best operation type and optimal time window are intelligently identified. In the early stages of disaster formation, artificial rain enhancement, rain suppression, and wind reduction technologies are used to proactively eliminate or weaken disaster risks, filling the technological gap in traditional emergency systems that lack proactive intervention capabilities. Emergency resources are intelligently matched based on disaster priority and optimal time window. The shortest path algorithm optimizes the scheduling path and accurately calculates the departure time, enabling the forward-looking deployment of emergency resources and avoiding the delays and waste of resources caused by the lag in resource scheduling in the traditional model. It fully releases the application value of meteorological big data, upgrading meteorological big data from a simple weather forecasting tool to a core driving force supporting intelligent decision-making. It constructs a full-chain intelligent three-defense emergency command system integrating "risk identification, dynamic monitoring, proactive intervention, and intelligent scheduling", breaking through the technical bottlenecks of traditional emergency systems in terms of early warning timeliness, intervention capability, and resource allocation. It has important practical value and promotion significance for improving natural disaster prevention capabilities and protecting people's lives and property. Attached Figure Description

[0052] Figure 1 This is a flowchart of the emergency command method for three-defense based on meteorological big data according to Embodiment 1 of the present invention;

[0053] Figure 2 This is a schematic diagram of the three-defense emergency command system based on meteorological big data according to Embodiment 2 of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Example 1

[0056] Please see Figure 1 As shown in this embodiment, the emergency command method for disaster prevention, mitigation, and hazard mitigation based on meteorological big data includes the following:

[0057] Collect meteorological data related to disasters, identify meteorological evolution patterns corresponding to various disasters in the meteorological data related to disasters, and construct a database of meteorological precursor patterns for disaster prevention, mitigation, and disaster relief.

[0058] The methods for collecting disaster meteorological correlation data include:

[0059] Historical meteorological data covering the target area will be obtained from the meteorological data center. This historical meteorological data includes meteorological observation records collected at various historical moments in the target area. The target area refers to the specific geographical area targeted by the flood control and disaster relief emergency command center. Meteorological observation records include, but are not limited to, station number, observation time, temperature value, relative humidity value, wind speed value, wind direction value, air pressure value, and precipitation. Disaster archive data corresponding to the same time period as the historical meteorological data will be obtained from the emergency management department. The disaster archive data includes records of various disaster events that have occurred in the target area in the past. Disaster event records include, but are not limited to, disaster type, occurrence time, and scope of impact. Disaster types include fire, flood, and wind disaster.

[0060] Based on the occurrence time in each disaster event record, corresponding related meteorological sequences are extracted from historical meteorological data. Specifically, using the occurrence time in the disaster event record as a benchmark, a preset time window is traced back to obtain all meteorological observation records within the time window, and these records are arranged in ascending order according to the observation time to form a related meteorological sequence. The length of the time window is preset by those skilled in the art based on the actual situation. For example, if the length of the time window is set to 30 days, then meteorological observation records within 30 days before the disaster are extracted.

[0061] Each associated meteorological sequence is broken down into meteorological elements to obtain a meteorological element sequence corresponding to each meteorological element. Meteorological elements include, for example, temperature, relative humidity, wind speed, wind direction, air pressure, and precipitation. Meteorological element sequences include, for example, temperature sequence, humidity sequence, wind speed sequence, wind direction sequence, air pressure sequence, and precipitation sequence. Each meteorological element sequence is then associated with the corresponding disaster event record to form disaster meteorological association data.

[0062] The methods for constructing a database of meteorological precursor models for three-defense systems include:

[0063] Based on the disaster type of each disaster event record in the disaster meteorological association data, the disaster meteorological association data is divided into fire association dataset, flood association dataset and wind disaster association dataset;

[0064] For fire-related datasets, meteorological evolution patterns corresponding to fires are extracted and used as fire precursor patterns. Specifically, cumulative features are calculated for temperature sequences of all associated meteorological sequences in the fire-related dataset to obtain the duration of continuous high temperatures and cumulative temperature deviation for each temperature sequence; persistent features are calculated for humidity sequences of all associated meteorological sequences in the fire-related dataset to obtain the duration of continuous low humidity and average humidity level for each humidity sequence; and scarcity features are calculated for precipitation sequences of all associated meteorological sequences in the fire-related dataset to obtain the duration of continuous no precipitation and cumulative precipitation deficit for each precipitation sequence. The cumulative, persistent, and scarcity features of the same associated meteorological sequence are combined into a fire feature vector, and a temporal clustering algorithm is used to perform cluster analysis on all fire feature vectors to identify fire precursor patterns before the occurrence of fires.

[0065] For flood-related datasets, meteorological evolution patterns corresponding to floods are extracted and used as flood precursor patterns. Specifically, intensity features are calculated for precipitation sequences of all associated meteorological sequences in the flood-related dataset to obtain the number of short-duration heavy rainfalls and the total cumulative rainfall for each precipitation sequence. Change features are calculated for air pressure sequences of all associated meteorological sequences in the flood-related dataset to obtain the number of sudden air pressure drops and the amplitude of air pressure fluctuations for each air pressure sequence. The intensity and change features of the same associated meteorological sequence are combined into a flood feature vector, and a temporal clustering algorithm is used to perform cluster analysis on all flood feature vectors to identify flood precursor patterns before the occurrence of floods.

[0066] For wind disaster-related datasets, meteorological evolution patterns corresponding to wind disasters are extracted and used as precursor patterns. Specifically, cumulative features are calculated for wind speed sequences of all associated meteorological sequences in the wind disaster-related dataset to obtain the duration of sustained strong winds and maximum gust speed for each wind speed sequence; trend features are calculated for air pressure sequences of all associated meteorological sequences in the wind disaster-related dataset to obtain the duration of sustained air pressure decline and minimum air pressure value for each air pressure sequence; the cumulative features and trend features of the same associated meteorological sequence are combined into a wind disaster feature vector, and a temporal clustering algorithm is used to perform cluster analysis on all wind disaster feature vectors to identify precursor patterns of wind disasters before they occur.

[0067] By summarizing the precursor patterns of fires, floods, and winds, a database of meteorological precursor patterns for disaster prevention, mitigation, and hazard mitigation is constructed.

[0068] The cumulative characteristics are defined as follows: continuous high temperature duration refers to the longest duration during which the temperature value is continuously greater than the preset high temperature threshold in the temperature sequence, which is used to reflect the cumulative impact of continuous high temperature on the environment; cumulative temperature deviation refers to the index obtained by summing the differences between all temperature values ​​greater than the high temperature threshold and the high temperature threshold in the temperature sequence, which is used to reflect the cumulative impact of high temperature intensity on the environment.

[0069] The definition of the continuous feature is as follows: the continuous low humidity duration refers to the longest duration during which the relative humidity value is continuously less than the preset low humidity threshold in the humidity sequence, which is used to reflect the degree of dryness; the average humidity level refers to the arithmetic mean of all relative humidity values ​​in the humidity sequence, which is used to reflect the overall humidity status.

[0070] The definition of the scarcity feature is as follows: the duration of continuous no precipitation refers to the longest duration in a precipitation sequence where the precipitation is continuously equal to 0 or less than a preset trace threshold, which is used to reflect the persistence of drought; the cumulative precipitation deficit refers to the summation of the differences between all precipitation amounts below the preset normal threshold and the normal threshold in a precipitation sequence, and the absolute value of the summation is used to reflect the degree of accumulation of insufficient precipitation.

[0071] The intensity characteristics are defined as follows: the number of short-duration heavy precipitation events refers to the number of consecutive heavy precipitation events in a precipitation sequence, which reflects the frequency and intensity of heavy precipitation events; consecutive heavy precipitation events are defined as: a continuous period of time in which the precipitation amount is greater than a preset heavy precipitation threshold and the duration is greater than or equal to a preset duration threshold; the cumulative total precipitation refers to the index obtained by summing up all precipitation amounts in a precipitation sequence, which reflects the intensity of the total precipitation.

[0072] The characteristics of change are defined as follows: the number of times the pressure drops sharply refers to the number of times in the pressure sequence the difference between the corresponding pressure values ​​at two adjacent moments (i.e., the pressure value at the previous moment minus the pressure value at the next moment) is greater than a preset pressure threshold, which is used to reflect the frequency of rapid pressure drops; the pressure fluctuation amplitude refers to the index obtained by calculating the difference between the maximum and minimum pressure values ​​in the pressure sequence, which is used to reflect the overall amplitude of pressure changes.

[0073] The cumulative feature is defined as follows: the duration of continuous strong winds refers to the longest duration during which the wind speed value is continuously greater than the preset wind speed threshold in the wind speed sequence, which is used to reflect the duration of strong winds and their cumulative impact; the maximum gust wind speed refers to the wind speed value with the largest value in the wind speed sequence, which is used to reflect the instantaneous wind intensity and destructive force.

[0074] The trend characteristics are defined as follows: the duration of continuous pressure decline refers to the longest duration of continuous pressure decline in the pressure sequence, which is used to reflect the duration of the pressure decline trend; the minimum pressure value refers to the smallest pressure value in the pressure sequence, which is used to reflect the extreme level reached by the pressure change.

[0075] It should be understood that the identification methods for fire precursor patterns, flood precursor patterns, and wind precursor patterns are all the same. This embodiment uses the fire precursor pattern identification method as an example for specific description. The method for identifying fire precursor patterns before a fire occurs is as follows: all fire feature vectors are clustered to obtain multiple feature clusters; the fire feature vector corresponding to the center of each feature cluster is obtained and marked as the center feature vector; each center feature vector is compared with a preset fire feature range, which includes the normal value range of each feature within the fire feature vector; if the center feature vector is within the fire feature range, the corresponding center feature vector is not considered a fire precursor pattern; if the center feature vector exceeds the fire feature range, the corresponding center feature vector is considered a fire precursor pattern.

[0076] It should be noted that the temporal clustering algorithm is a well-known technology in this field, and the specific process will not be described in detail here; the high temperature threshold, low humidity threshold, trace threshold, normal threshold, heavy precipitation threshold, duration threshold, air pressure threshold, wind speed threshold and fire characteristic range are all preset by those skilled in the art according to the actual situation.

[0077] Real-time collection of current meteorological and environmental data is used to assess the progressive evolution of vulnerability of disaster-bearing bodies corresponding to various disasters and generate progressive risk curves for disaster prevention, mitigation, and disaster relief.

[0078] Among them, the current meteorological data covering the target area is obtained from the official API of the meteorological department. The current meteorological data includes meteorological observation records collected in the target area at the current time; the environmental status data refers to the status information of the disaster-bearing body related to the three types of disasters: fire, flood and wind disaster, including fire status data, flood status data and wind disaster status data, which are obtained through the environmental monitoring system.

[0079] Among them, fire status data includes combustible material moisture content and surface dryness; flood status data includes soil moisture content, river water level and drainage network load rate; wind disaster status data includes building wind resistance rating, number of temporary facilities and tree fall risk index;

[0080] Combustible material moisture content refers to the proportion of water in the total weight of combustible materials such as dead branches and fallen leaves under the forest canopy; surface dryness refers to the degree of dryness of the surface soil; soil moisture content refers to the proportion of water in the soil to the total volume of the soil; river water level refers to the current water level of the river; drainage network load rate refers to the ratio of the actual drainage volume of the network to the designed drainage capacity; building wind resistance rating refers to the wind force intensity that the building can actually withstand; number of temporary facilities refers to the number of temporary buildings or facilities in the target area that are susceptible to wind; tree fall risk index refers to the risk that trees may fall under the action of wind.

[0081] The environmental monitoring system consists of remote sensing satellites, IoT sensors, and ground inspection systems. IoT sensors include, but are not limited to, soil moisture sensors, near-infrared moisture sensors, ultrasonic water level gauges, flow sensors, building structure health monitoring sensors (such as tilt sensors, strain gauges, etc.), tree tilt sensors, etc. The ground inspection system includes, but is not limited to, drone inspections (such as visible light, infrared, lidar, etc.), fixed video surveillance (such as roadside cameras, construction site cameras, etc.), and manual-assisted inspections (such as handheld temperature and humidity meters, handheld soil moisture meters, etc.).

[0082] It should be noted that comprehensive environmental status indicators such as building wind resistance rating and tree fall risk index are calculated by combining data from multiple sensors and ground inspection information with statistical scoring, engineering models, or multi-source data fusion. The specific calculation methods are well-known technologies in this field and will not be elaborated on here. For example, building wind resistance rating and tree fall risk index are conventional assessment indicators in the fields of building engineering and landscape management, respectively. Their calculation methods have been clearly defined and maturely applied in relevant national standards and industry specifications, and therefore belong to well-known technologies in this field.

[0083] The methods for assessing the progressive evolution of vulnerability of various disaster-bearing bodies include:

[0084] The progressive evolution of the vulnerability of fire-affected bodies is assessed to form a progressive sequence of fire vulnerability. Specifically, based on the temperature and relative humidity values ​​in the current meteorological data, the atmospheric dryness index is calculated at the current moment; based on the moisture content of combustibles and surface dryness in the fire status data, the combustible flammability index is calculated at the current moment; the atmospheric dryness index and the combustible flammability index are normalized separately, and corresponding fire weights are assigned to each; based on the fire weights, the normalized atmospheric dryness index and the combustible flammability index are weighted and summed to obtain the fire vulnerability at the current moment. A fire vulnerability time-series queue is constructed to store the fire vulnerability at each time point. The fire vulnerability at the current time point is added to the fire vulnerability time-series queue, and the corresponding fire vulnerability increment is calculated sequentially based on the fire vulnerability at every two adjacent time points in the fire vulnerability time-series queue, forming a fire vulnerability increment sequence. The fire vulnerability increment is equal to the fire vulnerability at the next time point minus the fire vulnerability at the previous time point. The progressive evolution trend of the fire vulnerability increment sequence is analyzed using a moving average algorithm or an exponential smoothing algorithm to form a progressive fire vulnerability sequence.

[0085] The progressive evolution of vulnerability of flood-affected bodies is assessed to form a progressive vulnerability sequence. Specifically, based on precipitation and air pressure values ​​in current meteorological data, the precipitation intensity index is calculated for the current moment; based on soil moisture content and river level in flood status data, the surface saturation index is calculated for the current moment; and based on drainage network load rate in flood status data, the drainage pressure index is calculated for the current moment. The precipitation intensity index, surface saturation index, and drainage pressure index are normalized and assigned corresponding flood weights. Based on the flood weights, the normalized precipitation intensity index is... The flood vulnerability index, surface saturation index, and drainage pressure index are weighted and summed to obtain the flood vulnerability level at the current moment. A flood vulnerability time series queue is constructed to store the flood vulnerability level at each moment. The flood vulnerability level at the current moment is added to the flood vulnerability time series queue, and the corresponding flood vulnerability level increment is calculated sequentially based on the flood vulnerability level at every two adjacent moments in the flood vulnerability time series queue, forming a flood vulnerability level increment sequence. The progressive evolution trend of the flood vulnerability level increment sequence is analyzed using a moving average algorithm or an exponential smoothing algorithm to form a progressive flood vulnerability level sequence.

[0086] The progressive evolution of vulnerability of affected structures in response to wind disasters is assessed, forming a progressive sequence of wind disaster vulnerability. Specifically, based on wind speed and air pressure values ​​from current meteorological data, the wind intensity index for the current moment is calculated; based on the wind resistance level of buildings and the number of temporary facilities from wind disaster status data, the structural vulnerability index for the current moment is calculated; based on the tree fall risk index from wind disaster status data, the vegetation risk index for the current moment is obtained; the wind intensity index, structural vulnerability index, and vegetation risk index are normalized respectively, and corresponding wind disaster weights are assigned to each; based on the wind disaster weights, the normalized wind intensity index is... The wind damage vulnerability index, structural vulnerability index, and vegetation risk index are weighted and summed to obtain the wind damage vulnerability level at the current moment. A wind damage vulnerability time series queue is constructed to store the wind damage vulnerability level at each moment. The wind damage vulnerability level at the current moment is added to the wind damage vulnerability time series queue, and the corresponding wind damage vulnerability increment is calculated sequentially based on the wind damage vulnerability level of every two adjacent moments in the wind damage vulnerability time series queue, forming a wind damage vulnerability increment sequence. The progressive evolution trend of the wind damage vulnerability increment sequence is analyzed using a moving average algorithm or an exponential smoothing algorithm to form a progressive wind damage vulnerability sequence.

[0087] The atmospheric dryness index is calculated as follows: the temperature value is normalized to obtain the standard temperature value; the relative humidity value is normalized to obtain the standard humidity value; the difference between the temperature value and the standard humidity value is calculated, and the result is multiplied by the standard temperature value to obtain the atmospheric dryness index; the atmospheric dryness index is used to reflect the dryness of the atmospheric environment, and the larger the value, the drier the atmosphere.

[0088] The method for calculating the flammability index of a combustible is as follows: calculate the difference between the moisture content of the combustible and the dryness of the combustible to obtain the dryness of the combustible; calculate the average of the dryness of the combustible and the dryness of the surface to obtain the flammability index of the combustible; the flammability index of the combustible is used to reflect the flammability of the combustible, and the larger the value, the easier the combustible is to burn.

[0089] The precipitation intensity index is calculated as follows: Precipitation is normalized to obtain standard precipitation; the rate of change of air pressure (current air pressure minus previous air pressure) is calculated based on the air pressure value, and this rate of change is normalized to obtain the standard rate of change of air pressure; corresponding precipitation weights are assigned to both standard precipitation and the standard rate of change of air pressure, and a weighted sum is calculated based on these weights to obtain the precipitation intensity index; the precipitation intensity index reflects the intensity and persistence of precipitation, with a higher value indicating stronger precipitation.

[0090] The method for calculating the surface saturation index is as follows: the soil moisture content is normalized to obtain the standard soil moisture content; the river water level is normalized to obtain the standard river water level; the average of the standard soil moisture content and the standard river water level is calculated to obtain the surface saturation index; the surface saturation index is used to reflect the degree of water saturation on the surface, and the larger the value, the more saturated the surface is.

[0091] The drainage pressure index is calculated by normalizing the load rate of the drainage network. The drainage pressure index reflects the pressure bearing capacity of the drainage system; the larger the value, the greater the drainage pressure.

[0092] The wind intensity index is calculated as follows: Wind speed values ​​are normalized to obtain standard wind speed values; the air pressure drop is calculated based on the air pressure value (i.e., the air pressure value at the previous moment minus the air pressure value at the current moment), and the air pressure drop is normalized to obtain standard air pressure drop amplitude; corresponding wind force weights are assigned to the standard wind speed value and the standard air pressure drop amplitude, and a weighted sum is calculated based on these weights to obtain the wind intensity index; the wind intensity index reflects the intensity and development trend of wind, with a larger value indicating stronger winds.

[0093] The structural vulnerability index is calculated as follows: the wind resistance level of the building is reverse normalized to obtain the building vulnerability level; the number of temporary facilities is normalized to obtain the temporary facility risk level; the average of the building vulnerability level and the temporary facility risk level is calculated to obtain the structural vulnerability index; the structural vulnerability index is used to reflect the vulnerability of the building structure, and the larger the value, the more vulnerable the structure.

[0094] The vegetation risk index, also known as the tree lodging risk index in wind disaster data, is used to reflect the risk of vegetation lodging under strong wind conditions. The higher the value, the higher the risk of lodging.

[0095] It should be noted that the weights for fire, flood, wind, precipitation, and wind are all preset by those skilled in the art based on actual conditions; the moving average algorithm and the exponential smoothing algorithm are well-known technologies in the field, and the specific processes will not be elaborated here; the calculation methods for the flood vulnerability increment and the wind vulnerability increment are consistent with the calculation method for the fire vulnerability increment; normalization refers to mapping the original values ​​to the range of 0 to 1 to eliminate dimensional differences.

[0096] The methods for generating the three-defense risk progression curve include:

[0097] With time as the horizontal axis and the progressive evolution value in the fire vulnerability progression sequence as the vertical axis, a fire risk progression curve is plotted to visually demonstrate the progressive evolution trend of fire vulnerability over time.

[0098] A flood risk progression curve is plotted with time on the horizontal axis and the progressive evolution value in the flood vulnerability progression sequence on the vertical axis. This curve is used to visually demonstrate the progressive evolution trend of flood vulnerability over time.

[0099] Plot a wind disaster risk progression curve with time as the horizontal axis and the progressive evolution value in the wind disaster vulnerability progression sequence as the vertical axis, to visually demonstrate the progressive evolution trend of wind disaster vulnerability over time.

[0100] By integrating the fire risk progression curve, the flood risk progression curve, and the wind risk progression curve, a three-defense risk progression curve is generated.

[0101] Acquire meteorological forecast data, calculate the meteorological similarity between the meteorological forecast data and various meteorological evolution models in the disaster prevention and mitigation meteorological precursor model library, and combine the disaster prevention and mitigation risk progression curve to predict the disaster risk factors of various types of disasters, and determine the disaster intervention based on the disaster risk factors.

[0102] Specifically, meteorological forecast data covering the target area is obtained from the official API of the meteorological department. The meteorological forecast data includes meteorological forecast records for the target area at various future times. The meteorological forecast records include, but are not limited to, station number, forecast time, forecast temperature value, forecast relative humidity value, forecast wind speed value, forecast precipitation, forecast cloud thickness, and forecast cloud base height. The forecast time range is preset by those skilled in the art based on the actual situation. For example, if the forecast time range is set to the next 72 hours, then meteorological forecast records for the next 72 hours are obtained.

[0103] Methods for calculating the meteorological similarity between weather forecast data and various meteorological evolution models include:

[0104] From the meteorological forecast data, fire feature vector, flood feature vector, and wind disaster feature vector are extracted in sequence and labeled as fire forecast vector, flood forecast vector, and wind disaster forecast vector, respectively.

[0105] The fire forecast vector and each fire precursor model are normalized respectively. The Euclidean distance between the normalized fire forecast vector and each fire precursor model is calculated, and the minimum Euclidean distance is taken as the fire feature distance. The fire feature distance is then converted to obtain the fire meteorological similarity.

[0106] The similarity conversion method is as follows: calculate the ratio of the fire feature distance to the preset distance normalization coefficient, and subtract the calculated result from the ratio to obtain the fire meteorological similarity.

[0107] Based on the method for calculating fire meteorological similarity, the meteorological similarity of floods corresponding to flood forecast vectors and the meteorological similarity of winds corresponding to wind forecast vectors are calculated sequentially.

[0108] The methods for predicting disaster risk factors for various types of disasters include:

[0109] The method predicts fire disaster risk factors and uses them as fire risk elements. Specifically, it uses linear regression or exponential smoothing algorithms to extrapolate the fire risk progression curve, predicting the progression value at each future moment to form a fire prediction progression curve. It identifies the moment when the progression value in the fire prediction progression curve first reaches a preset fire progression threshold and marks it as the fire critical moment. It calculates the difference between the fire critical moment and the current moment to obtain the remaining fire time. It calculates the slope of the fire risk progression curve (i.e., the slope of the tangent line at the current moment) and normalizes it to obtain the risk evolution rate. It sets corresponding fire risk weights for fire meteorological similarity and risk evolution rate, and calculates the fire occurrence probability by weighted summation based on these fire risk weights. It obtains disaster event records associated with the fire precursor patterns corresponding to the fire meteorological similarity and extracts the impact range from these disaster event records as the fire impact range. Finally, it integrates the remaining fire time, the fire occurrence probability, and the fire impact range to form fire risk elements.

[0110] Based on the prediction methods for fire risk factors, the disaster risk factors for floods and wind disasters are predicted separately and used as the risk factors for floods and wind disasters, respectively.

[0111] Among them, methods for determining disaster intervention based on disaster risk factors include:

[0112] For each group of disaster risk factors, the necessity of intervention is determined sequentially. Specifically, a fire intervention threshold is preset, and the fire risk factors are compared with the fire intervention threshold. The fire intervention threshold includes a fire time threshold and a fire probability threshold. If the remaining time of the fire is less than or equal to the fire time threshold, and the probability of the fire occurring is greater than or equal to the fire probability threshold, then the fire is determined to require intervention. If the remaining time of the fire is greater than the fire time threshold, or the probability of the fire occurring is less than the fire probability threshold, then the fire is determined not to require intervention. Using the same determination method, the necessity of intervention for floods and winds is determined separately based on the flood risk factors and wind risk factors.

[0113] All disaster types identified as requiring intervention are compiled into an intervention disaster list. If multiple disaster types exist in several disaster prevention lists, the disaster types in the intervention disaster lists are prioritized. Specifically, the intervention urgency corresponding to each disaster type in the intervention disaster list is calculated, and the disaster types in the intervention disaster list are sorted in descending order according to the intervention urgency to generate an intervention disaster sequence. According to the ascending order of the intervention disaster sequence, each disaster type is marked as an intervention disaster and assigned a decreasing numerical label as the disaster priority for each disaster type.

[0114] The calculation method for the intervention urgency is consistent for each type of disaster. This embodiment takes the intervention urgency corresponding to fire as an example for detailed explanation. The calculation method for intervention urgency is as follows: reverse normalize the remaining time of the fire to obtain the time urgency; normalize the probability of fire occurrence and the scope of fire impact to obtain the probability urgency and the scope urgency; set corresponding urgency weights for time urgency, probability urgency and scope urgency respectively, and calculate the intervention urgency by weighted summation based on the urgency weights.

[0115] It should be noted that the linear regression algorithm is a well-known technology in this field, and the specific process will not be described in detail here; the distance normalization coefficient, fire risk weight, fire progression threshold, fire intervention threshold and urgency weight are all preset by those skilled in the art according to the actual situation.

[0116] Based on the disaster risk factors of disaster intervention, the feasible time periods and expected effects of different artificial weather modification operations corresponding to disaster intervention are analyzed, and the best operation type and the best time window are intelligently identified based on the feasible time periods and expected effects.

[0117] The methods for analyzing the feasible timing and expected effects of different weather modification operations in response to disaster intervention include:

[0118] Based on the type of disaster to be intervened, a set of candidate operation types is obtained from a pre-built operation type library. The set of candidate operation types includes the candidate operation type (i.e., weather modification operation type) and the operation implementation conditions. For example, candidate operation types corresponding to fire include, but are not limited to, artificial rain enhancement and artificial humidification; candidate operation types corresponding to floods include, but are not limited to, artificial rain suppression and artificial precipitation reduction; candidate operation types corresponding to wind disasters include, but are not limited to, artificial fog dissipation and artificial wind reduction. The operation implementation conditions include temperature conditions, humidity conditions, cloud conditions, and wind speed conditions. Temperature conditions refer to the temperature range required for weather modification operations; humidity conditions refer to the relative humidity range required for weather modification operations; cloud conditions refer to the cloud thickness and cloud base height range required for weather modification operations; and wind speed conditions refer to the wind speed range required for weather modification operations. The operation type library is pre-built by those skilled in the art according to weather modification technical specifications.

[0119] For each candidate operation type, the corresponding feasible time period is analyzed. Specifically, meteorological forecast records for each future time are extracted from meteorological forecast data, and each future time meteorological forecast record is compared item by item with the operation implementation conditions. If the forecast temperature, forecast relative humidity, forecast wind speed, forecast cloud thickness, and forecast cloud base height in the meteorological forecast record all meet the same operation implementation conditions, then the corresponding future time is considered a feasible time for the corresponding candidate operation type. If any parameter in the meteorological forecast record does not meet the same operation implementation conditions, then the corresponding future time is considered an infeasible time for the corresponding candidate operation type. The feasible times corresponding to each candidate operation type are summarized to form the feasible time period for each candidate operation type.

[0120] Different numerical labels are assigned to different candidate operation types and marked as operation labels. For each candidate operation type, the disaster risk factors corresponding to the disaster intervention, a feasible time-based meteorological forecast record, and the operation label are combined to obtain multiple sets of change predictions. Each set of change predictions includes a set of disaster risk factors, a feasible time-based meteorological forecast record, and an operation label. Each set of change predictions is input into a trained change prediction model to predict the meteorological change set after each candidate operation type at different feasible times for each type of disaster intervention, which is then used as the expected effect of each candidate operation type. The meteorological change set includes temperature change, relative humidity change, wind speed change, and precipitation change.

[0121] Specifically, the change prediction model is a deep neural network model, comprising an input layer, hidden layers, and an output layer. Each hidden layer contains multiple neurons, and each neuron is connected to neurons in the next layer. These connections contain weights that determine the importance and impact of data transmitted within the neural network. An activation function is applied to each neuron between the hidden and output layers. This activation function introduces non-linearity, allowing the network to learn more complex patterns and features. The training process of the change prediction model includes:

[0122] Pre-collection Different sets of change predictions are set, and corresponding meteorological change sets are sequentially set for each set of change predictions. The variable is an integer greater than 1; the predicted change set and the corresponding meteorological change set are converted into a set of corresponding feature vectors; the meteorological change set corresponding to the predicted change set is collected by those skilled in the art during the historical analysis of the expected effects of different weather modification operations on different disaster interventions. Different sets of change forecasts are created, and candidate job types corresponding to the job labels in each set are implemented sequentially. After the candidate job types are implemented, the meteorological change sets corresponding to each set of change forecasts are collected. Different sets of change prediction sets are sequentially configured with corresponding meteorological change sets;

[0123] Each set of feature vectors is used as input to the change prediction model. The model outputs a set of predicted meteorological changes corresponding to each set of change predictions, and uses the set of actual meteorological changes corresponding to each set of change predictions as the prediction target. The actual meteorological change set is the pre-set set of meteorological changes corresponding to the change prediction set. The training objective is to minimize the sum of prediction errors of all change prediction sets. The formula for calculating the prediction error is: In the formula, For prediction error, The group number of the feature vector corresponding to the change prediction set. For the first The set of predicted meteorological changes corresponding to the set of predicted group changes. For the first The set of predicted changes corresponds to the set of actual meteorological changes; the change prediction model is trained until the sum of prediction errors converges, at which point training stops.

[0124] The methods for intelligently identifying the optimal job type and optimal time window include:

[0125] The feasible time period for each candidate operation type is compared with the remaining time of the corresponding disaster. The remaining time of the disaster includes the remaining time of fire, flood and wind disasters. If at least one feasible moment in the feasible time period is within the remaining time of the corresponding disaster, the corresponding candidate operation type is considered a valid operation type. If no feasible moment in the feasible time period is within the remaining time of the corresponding disaster, the corresponding candidate operation type is considered an invalid operation type.

[0126] Based on the expected effects of each effective operation type, the risk reduction rate of each effective operation type for the corresponding disaster at different feasible times is calculated. Specifically, for several potential disasters, such as fire, the humidity change value is normalized to obtain the humidity reduction contribution; the precipitation change value is normalized to obtain the precipitation reduction contribution; corresponding reduction weights are set for the humidity reduction contribution and the precipitation reduction contribution, and a weighted sum is calculated based on the reduction weights to obtain the fire risk reduction rate; for several potential disasters, such as flood, the ratio between the absolute value of the precipitation change value and the predicted precipitation in the corresponding change prediction set is calculated to obtain the precipitation reduction ratio; the precipitation reduction ratio is normalized to obtain the flood risk reduction rate; for several potential disasters, such as wind, the ratio between the absolute value of the wind speed change value and the predicted wind speed in the corresponding change prediction set is calculated to obtain the wind speed reduction ratio; the wind speed reduction ratio is normalized to obtain the wind risk reduction rate.

[0127] Based on the feasible time periods for each effective operation type, a time period score is calculated for each effective operation type. Specifically, the number of feasible moments within the feasible time period and the number of future moments within the forecast time range are counted to obtain the feasible quantity and the future quantity. The ratio of the feasible quantity to the future quantity is calculated to obtain the time period score. A benefit score for each effective operation type is calculated based on the risk reduction rate, and combined with the corresponding time period score, a comprehensive operation score for each effective operation type is calculated. Specifically, the average risk reduction rate of the effective operation type under all feasible moments is calculated to obtain the average risk reduction rate. The average risk reduction rate is normalized to obtain the benefit score. Corresponding scoring weights are set for the benefit score and the time period score, and the benefit score and the time period score are weighted and summed based on the scoring weights to obtain the comprehensive operation score. The comprehensive operation scores of the effective operation types corresponding to the same intervention disaster are compared, and the effective operation type with the highest comprehensive operation score is selected as the optimal operation type for the corresponding intervention disaster.

[0128] Based on the feasible time periods for each optimal operation type, time intervals consisting of consecutive feasible moments are identified and marked as candidate time windows. For each candidate time window, the corresponding window optimization score is calculated sequentially. Specifically, the difference between the start time and the current time of the candidate time window is calculated to obtain the window lead time. The window lead time is normalized to obtain the lead time score. The duration of the candidate time window is calculated (i.e., the end time of the candidate time window minus the start time), and the duration is normalized to obtain the duration score. The mean of the risk reduction rate corresponding to each feasible moment within the candidate time window is calculated to obtain the window reduction rate. The window reduction rate is normalized to obtain the reduction degree score. Corresponding window weights are set for the lead time score, duration score, and reduction degree score, and the lead time score, duration score, and reduction degree score are weighted and summed based on the window weights to obtain the window optimization score.

[0129] The window optimization scores of candidate time windows corresponding to the same best job type are compared, and the candidate time window with the highest window optimization score is selected as the best time window for the corresponding best job type.

[0130] It should be noted that the reduction weights, scoring weights, and window weights are all preset by those skilled in the art based on the actual situation.

[0131] Based on the optimal operation type and time window for disaster intervention, the system intelligently schedules emergency resources for disaster prevention, mitigation, and hazard mitigation, generates preventive resource deployment plans, and directs various disaster prevention forces to execute these plans.

[0132] The methods for generating preventative resource deployment plans include:

[0133] Based on the disaster type and optimal operation type for each type of intervention disaster, an emergency resource requirement list corresponding to each intervention disaster is obtained from a pre-built resource allocation database. This database stores emergency resource requirement lists for different disaster types and different weather modification operations. These lists include requirements for operational equipment, personnel, and materials. Equipment requirements include equipment type and minimum operational capability; personnel requirements include personnel positions and minimum qualification levels; and material requirements include material type and quantity. Operational equipment includes, but is not limited to, artificial rainmaking rocket launchers, anti-aircraft artillery systems, aircraft seeding equipment, and ground-based combustion furnaces. Personnel requirements include, but are not limited to, meteorological operation commanders, equipment operators, safety supervisors, and logistics support personnel. Operational materials include, but are not limited to, silver iodide catalysts, liquid nitrogen, dry ice, and flares. The resource allocation database is pre-built by those skilled in the art based on the specifications and historical records of weather modification operations.

[0134] The system retrieves the status information of currently available emergency resources from the emergency resource management system. This information includes information on operational equipment, personnel, and supplies. Specifically, the operational equipment information includes the equipment type, current location, operational capabilities (quantifying the performance level or efficiency of the equipment during weather modification operations), and dispatch speed for each type of equipment. The personnel information includes the personnel's job position, current location, qualification level (measuring the professional competence and operational qualifications of personnel performing weather modification operations), and dispatch speed for each type of personnel. The supplies information includes the type of supplies, inventory quantity, current location, expiration date, and dispatch speed for each type of supplies.

[0135] Based on the emergency resource demand list and emergency resource status information, resource supply and demand are matched according to the disaster priority of the intervention. Specifically, each demand in the emergency resource demand list is matched with the emergency resource status information item by item. For each demand, emergency resources that meet the demand conditions and are closest to the target area (i.e., the Euclidean distance between the current location and the corresponding center location of the target area is the smallest) are selected. The demand conditions are as follows: for equipment demand, equipment of the same type and with an operational capacity greater than or equal to the minimum operational capacity are selected; for personnel demand, personnel whose positions match the personnel and whose qualification level is greater than or equal to the minimum qualification level are selected; for materials demand, materials of the same type, with an inventory quantity greater than or equal to the demand quantity, and an expiration date later than the termination time corresponding to the optimal time window are selected.

[0136] For each type of disaster requiring intervention, a corresponding emergency resource dispatch plan is generated. Specifically, a dispatch resource list is determined based on the matching results, including the operational equipment, personnel, and materials to be dispatched. The center location of the target area is determined using remote sensing satellites. Based on the current location of each dispatch resource (i.e., the operational equipment, personnel, and materials to be dispatched in the dispatch resource list) and the center location of the target area, the shortest path algorithm is used to calculate the resource dispatch path corresponding to each dispatch resource to minimize the time it takes for the resources to reach the target area. The length of the resource dispatch path is calculated and used as the path length. The ratio of the path length to the dispatch speed corresponding to each dispatch resource is calculated to obtain the dispatch time of each dispatch resource. The difference between the start time corresponding to the optimal time window and the dispatch time corresponding to each dispatch resource is calculated to obtain the dispatch departure time of each dispatch resource. The dispatch resource list and the dispatch departure time of each dispatch resource are summarized to generate an emergency resource dispatch plan. It should be noted that the shortest path algorithm is a well-known technology in this field, and the specific process will not be elaborated on here.

[0137] The resource allocation plans for each type of disaster intervention are summarized to generate a preventive resource deployment plan. Based on the preventive resource deployment plan, dispatch instructions are issued to the management units corresponding to each dispatched resource, directing various disaster prevention forces to execute resource dispatch tasks according to the dispatch departure time.

[0138] This embodiment constructs a database of meteorological precursor models for disaster prevention, mitigation, and hazard mitigation by deeply exploring the inherent correlation between historical disasters and meteorological evolution. Combined with a dynamic quantitative assessment of the vulnerability of disaster-bearing bodies, it generates a progressive risk curve for disaster prevention, mitigation, and hazard mitigation, achieving a paradigm shift from traditional "post-disaster emergency response" to "pre-disaster prevention." Innovatively, it accurately predicts risk factors such as the probability of disaster occurrence, remaining time, and scope of impact by calculating the similarity between meteorological forecast data and precursor models. Furthermore, based on a deep learning model, it analyzes the feasible time periods and expected effects of different artificial weather modification operations, intelligently identifying the optimal operation type and best time window. In the early stages of disaster formation, it proactively eliminates or weakens potential disaster risks through artificial rain enhancement, rain suppression, and wind reduction techniques, filling the technological gap in traditional emergency systems that lack proactive intervention capabilities. Based on disaster priority and optimal time window, emergency resources are intelligently matched. The shortest path algorithm optimizes the scheduling path and accurately calculates the departure time, enabling the forward-looking deployment of emergency resources and avoiding the delays and waste of resources caused by the lag in resource scheduling in the traditional model. It fully releases the application value of meteorological big data, upgrading meteorological big data from a simple weather forecasting tool to a core driving force supporting intelligent decision-making. It constructs a full-chain intelligent three-defense emergency command system integrating "risk identification, dynamic monitoring, proactive intervention, and intelligent scheduling", breaking through the technical bottlenecks of traditional emergency systems in terms of early warning timeliness, intervention capability, and resource allocation. It has important practical value and promotion significance for improving natural disaster prevention capabilities and protecting people's lives and property.

[0139] Example 2

[0140] Please see Figure 2 As shown in the figure, the parts not described in detail in this embodiment are described in Embodiment 1. A three-defense emergency command system based on meteorological big data is provided, including a meteorological mining module, a risk monitoring module, a disaster assessment module, an intervention identification module, and an emergency command module. The modules are connected by wired and / or wireless means to realize data transmission between the modules.

[0141] The meteorological mining module is used to collect disaster meteorological correlation data, identify meteorological evolution patterns corresponding to various disasters in the disaster meteorological correlation data, and build a database of meteorological precursor patterns for disaster prevention, mitigation, and disaster relief.

[0142] The risk monitoring module is used to collect current meteorological and environmental data in real time, assess the progressive evolution of the vulnerability of disaster-bearing bodies corresponding to various disasters, and generate a progressive curve of three-defense risk.

[0143] The disaster assessment module is used to acquire meteorological forecast data, calculate the meteorological similarity between the meteorological forecast data and various meteorological evolution models in the disaster prevention and mitigation meteorological precursor model library, and combine the disaster prevention and mitigation risk progression curve to predict the disaster risk factors of various types of disasters, and determine the disaster intervention based on the disaster risk factors.

[0144] The intervention identification module is used to analyze the feasible time periods and expected effects of different artificial weather modification operations corresponding to the disaster intervention based on the disaster risk factors of the disaster intervention, and intelligently identify the best operation type and the best time window based on the feasible time periods and expected effects.

[0145] The emergency command module is used to intelligently dispatch NBC (nuclear, biological, and chemical) emergency resources based on the optimal operation type and time window for disaster intervention, generate preventive resource deployment plans, and command various disaster prevention forces to execute the preventive resource deployment plans.

[0146] Example 3

[0147] This application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code, which, when executed by the one or more processors, can perform the emergency command method for disaster prevention and mitigation based on meteorological big data as described above.

[0148] The method or system according to the embodiments of this application can also be implemented using the architecture of the electronic device shown in this application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as a ROM or hard disk, may store the NBC emergency command method based on meteorological big data provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in this application is merely exemplary; when implementing different devices, one or more components of the electronic device shown in this application may be omitted according to actual needs.

[0149] Example 4

[0150] One embodiment of this application discloses a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the emergency command method for disaster prevention and mitigation based on meteorological big data, as described in the above-described embodiments of this application, can be executed. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0151] Furthermore, according to embodiments of this application, the processes described in the above-referenced flowcharts can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to perform instructions corresponding to the method steps provided in this application, such as a disaster prevention and mitigation emergency command method based on meteorological big data. When this computer program is executed by a central processing unit (CPU), it performs the functions defined in the method of this application.

[0152] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0153] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0154] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for emergency command of disaster prevention, mitigation, and hazard mitigation based on meteorological big data, characterized in that: include: Collect meteorological data related to disasters, identify meteorological evolution patterns corresponding to various disasters in the meteorological data related to disasters, and build a database of meteorological precursor patterns for disaster prevention, mitigation, and disaster relief. Real-time collection of current meteorological and environmental data; assessment of the progressive evolution of vulnerability of disaster-bearing bodies corresponding to various disasters; generation of three-defense risk progression curves. Methods for generating a three-proof risk progression curve include: Current meteorological data includes meteorological observation records collected in the target area at the current moment; environmental status data includes fire status data, flood status data, and wind disaster status data. Among them, fire status data includes combustible material moisture content and surface dryness; flood status data includes soil moisture content, river water level and drainage network load rate; wind disaster status data includes building wind resistance rating, number of temporary facilities and tree fall risk index; Based on current meteorological and fire status data, assess the progressive evolution of the vulnerability of the fire-affected body, form a progressive sequence of fire vulnerability, and plot a progressive curve of fire risk. Based on current meteorological and flood status data, assess the progressive evolution of the vulnerability of flood-affected bodies, form a progressive sequence of flood vulnerability, and plot a progressive curve of flood risk. Based on current meteorological data and wind disaster status data, assess the progressive evolution of the vulnerability of the corresponding disaster-bearing bodies, form a progressive sequence of wind disaster vulnerability, and plot a progressive curve of wind disaster risk. The fire risk progression curve, the flood risk progression curve, and the wind risk progression curve are integrated to generate the three-defense risk progression curve. Acquire meteorological forecast data, calculate the meteorological similarity between the meteorological forecast data and various meteorological evolution models in the disaster prevention and mitigation meteorological precursor model library, and combine the disaster prevention and mitigation risk progression curve to predict the disaster risk factors of various disasters, and determine the disaster intervention based on the disaster risk factors; Methods for predicting disaster risk factors for various types of disasters include: The meteorological forecast data includes meteorological forecast records for the target area at various future times. From the meteorological forecast data, fire forecast vectors, flood forecast vectors, and wind forecast vectors are extracted in sequence, and the meteorological similarity of fire forecast vectors, flood forecast vectors, and wind forecast vectors is calculated. The fire risk progression curve is extrapolated to predict the progression value at each future moment, forming a fire prediction progression curve. The moment when the progression value in the fire prediction progression curve first reaches the preset fire progression threshold is obtained and marked as the fire critical moment. The remaining time of the fire is calculated based on the fire critical moment and the current moment. The risk evolution rate is calculated based on the fire risk progression curve. The fire meteorological similarity and risk evolution rate are weighted and summed based on the preset fire risk weights to obtain the fire occurrence probability. The fire impact range is obtained based on the fire precursor pattern corresponding to the fire meteorological similarity. The remaining time of the fire, the fire occurrence probability, and the fire impact range are integrated to form fire risk elements. Based on the prediction method of fire risk factors, the disaster risk factors of floods and wind disasters are predicted respectively, and used as the disaster risk factors of floods and wind disasters respectively. Based on the disaster risk factors of disaster intervention, the feasible time periods and expected effects of different artificial weather modification operations corresponding to disaster intervention are analyzed, and the best operation type and the best time window are intelligently identified based on the feasible time periods and expected effects. Based on the optimal operation type and time window for disaster intervention, the system intelligently schedules emergency resources for disaster prevention, mitigation, and hazard mitigation, generates preventive resource deployment plans, and directs various disaster prevention forces to execute these plans.

2. The emergency command method for disaster prevention, mitigation, and hazard mitigation based on meteorological big data according to claim 1, characterized in that, Methods for collecting disaster meteorological correlation data include: Acquire historical meteorological data covering the target area, including meteorological observation records collected at various historical moments in the target area; acquire disaster archive data corresponding to the same time period as the historical meteorological data, including records of various disaster events that have occurred in the target area in the past; Based on the occurrence time in each disaster event record, the corresponding associated meteorological sequence is extracted from historical meteorological data; each associated meteorological sequence is decomposed into meteorological elements to obtain a meteorological element sequence corresponding to each meteorological element; each meteorological element sequence is associated with the corresponding disaster event record to form disaster meteorological association data.

3. The emergency command method for three-defense based on meteorological big data according to claim 2, characterized in that, Methods for constructing a database of meteorological precursor models for three-defense systems include: Based on the disaster type of each disaster event record in the disaster meteorological association data, the disaster meteorological association data is divided into fire association dataset, flood association dataset and wind disaster association dataset; For fire-related datasets, meteorological evolution patterns of fires are extracted and used as fire precursor patterns. For flood-related datasets, meteorological evolution patterns of floods are extracted and used as precursor patterns for floods. For wind disaster-related datasets, meteorological evolution patterns of wind disasters are extracted and used as precursor patterns for wind disasters. By summarizing the precursor patterns of fires, floods, and winds, a database of meteorological precursor patterns for disaster prevention, mitigation, and hazard mitigation is constructed.

4. The emergency command method for three-defense based on meteorological big data according to claim 3, characterized in that, Methods for determining disaster intervention based on disaster risk factors include: For each group of disaster risk factors, the necessity of intervention is determined sequentially. A fire intervention threshold is preset, and the fire risk factors are compared with the fire intervention threshold. The fire intervention threshold includes a fire time threshold and a fire probability threshold. If the remaining time of the fire is less than or equal to the fire time threshold, and the probability of the fire occurring is greater than or equal to the fire probability threshold, then the fire is determined to require intervention. If the remaining time of the fire is greater than the fire time threshold, or the probability of the fire occurring is less than the fire probability threshold, then the fire is determined not to require intervention. Using the same determination method, the necessity of intervention for floods and winds is determined separately based on the flood risk factors and wind risk factors. All disaster types identified as requiring intervention are compiled into an intervention disaster list. If multiple disaster types exist in several disaster prevention lists, the intervention urgency corresponding to each disaster type in the intervention disaster list is calculated. The disaster types in the intervention disaster list are sorted in descending order according to the intervention urgency to generate an intervention disaster sequence. According to the ascending order of the intervention disaster sequence, each disaster type is marked as an intervention disaster and assigned a decreasing numerical label as the disaster priority for each disaster type.

5. The emergency command method for three-defense based on meteorological big data according to claim 4, characterized in that, Methods for analyzing the feasible timing and expected effects of different weather modification operations in response to disaster interventions include: Based on the disaster type of the intervention, a set of candidate operation types is obtained from a pre-built operation type library. The set of candidate operation types includes the candidate operation type and the operation implementation conditions. The candidate operation type is the artificial weather modification operation type. Meteorological forecast records for each future time are extracted from the meteorological forecast data. Each future time meteorological forecast record is compared with each operation implementation condition. If the parameters in the meteorological forecast record all meet the same operation implementation conditions, the corresponding future time is taken as the feasible time for the corresponding candidate operation type. The feasible times corresponding to each candidate operation type are summarized to form the feasible time period corresponding to each candidate operation type. Different numerical labels are assigned to different candidate operation types and marked as operation labels. The disaster risk factors of the disaster to be intervened for each candidate operation type, the meteorological forecast record at a feasible time, and the operation label are combined to obtain multiple sets of change prediction sets. Each set of change prediction sets is input into the trained change prediction model to predict the meteorological change set after each intervention disaster is applied by each candidate operation type at different feasible times, and this is used as the expected effect of each candidate operation type. The meteorological change set includes temperature change value, relative humidity change value, wind speed change value, and precipitation change value.

6. The emergency command method for three-defense based on meteorological big data according to claim 5, characterized in that, Methods for intelligently identifying the optimal job type and optimal time window include: The feasible time periods for each candidate operation type are compared with the remaining time of the corresponding disaster; the remaining time of the disaster includes the remaining time of fire, flood and wind disasters; if at least one feasible moment in the feasible time period is within the remaining time of the corresponding disaster, the corresponding candidate operation type is taken as the valid operation type. Based on the expected effects of each effective operation type, calculate the risk reduction rate of each effective operation type for the corresponding disaster at different feasible times; calculate the time period score of each effective operation type based on the feasible time period of each effective operation type; calculate the benefit score of each effective operation type based on the risk reduction rate, and combine it with the corresponding time period score to calculate the comprehensive operation score of each effective operation type; calculate the weighted sum of the benefit score and the time period score based on the preset scoring weights to obtain the comprehensive operation score; compare the comprehensive operation scores of the effective operation types corresponding to the same disaster, and select the effective operation type with the highest comprehensive operation score as the best operation type for the corresponding disaster. Based on the feasible time periods for each optimal job type, identify the time intervals consisting of consecutive feasible moments and mark them as candidate time windows; for each candidate time window, calculate the corresponding window optimization score in sequence; compare the window optimization scores of the candidate time windows corresponding to the same optimal job type, and select the candidate time window with the highest window optimization score as the optimal time window for the corresponding optimal job type.

7. The emergency command method for disaster prevention, mitigation, and hazard mitigation based on meteorological big data according to claim 6, characterized in that, Methods for generating preventative resource deployment plans include: Based on the disaster type and optimal operation type of each intervention disaster, obtain the emergency resource demand list corresponding to each intervention disaster from the pre-built resource allocation library; obtain the current schedulable emergency resource status information; and match resource supply and demand according to the disaster priority of the intervention disaster based on the emergency resource demand list and emergency resource status information; determine the schedulable resource list for each intervention disaster based on the matching results. The schedulable resource list includes the operational equipment, personnel, and materials to be scheduled, and is collectively referred to as schedulable resources. Determine the center location of the target area and obtain the current location and dispatch speed of each dispatch resource from the emergency resource status information; calculate the resource dispatch path corresponding to each dispatch resource based on the current location of each dispatch resource and the center location of the target area; calculate the dispatch departure time of each dispatch resource based on the length of the corresponding resource dispatch path and the dispatch speed; summarize the list of dispatch resources for the same intervention disaster and the dispatch departure time of each dispatch resource to generate an emergency resource dispatch plan for each intervention disaster; summarize the resource dispatch plans for each intervention disaster to generate a preventive resource deployment plan.

8. A flood control and drought relief emergency command system based on meteorological big data, implementing the flood control and drought relief emergency command method based on meteorological big data as described in any one of claims 1-7, characterized in that, include: The meteorological mining module is used to collect disaster meteorological correlation data, identify meteorological evolution patterns corresponding to various disasters in the disaster meteorological correlation data, and build a database of meteorological precursor patterns for disaster prevention, mitigation, and disaster relief. The risk monitoring module is used to collect current meteorological and environmental data in real time, assess the progressive evolution of the vulnerability of disaster-bearing bodies corresponding to various disasters, and generate a progressive curve of disaster prevention, mitigation, and disaster relief risks. The disaster assessment module is used to acquire meteorological forecast data, calculate the meteorological similarity between the meteorological forecast data and various meteorological evolution models in the disaster prevention and mitigation meteorological precursor model library, and combine the disaster prevention and mitigation risk progression curve to predict the disaster risk factors of various types of disasters, and determine the disaster intervention based on the disaster risk factors. The intervention identification module is used to analyze the feasible time periods and expected effects of different artificial weather modification operations corresponding to the disaster intervention based on the disaster risk factors of the disaster intervention, and intelligently identify the best operation type and the best time window based on the feasible time periods and expected effects. The emergency command module is used to intelligently dispatch NBC (nuclear, biological, and chemical) emergency resources based on the optimal operation type and time window for disaster intervention, generate preventive resource deployment plans, and command various disaster prevention forces to execute the preventive resource deployment plans.

Images