Meteorological disaster early warning method, device, equipment and storage medium

Abstract

The invention discloses a meteorological disaster early warning method, device and equipment and a storage medium. The method comprises the following steps: determining a disaster category and an early warning level of meteorological early warning data; inputting the disaster category and the early warning level into a monitoring early warning model to obtain a defense guide corresponding to the meteorological early warning data; inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding early warning mode; and sending each disaster response task to a corresponding contact person in a corresponding early warning mode, so that the contact person allocates the disaster response task and executes the disaster response task. According to the method, the meteorological early warning data is classified, the disaster response tasks are generated through the model, and the disaster response tasks are sent to the corresponding contacts in the corresponding early warning modes, so that the responsiveness to meteorological disasters can be improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of disaster warning technology, and in particular to a meteorological disaster warning method, device, equipment and storage medium. Background Art

[0002] In recent years, extreme weather events have often developed rapidly over short periods of time, increasing the difficulty of prevention and response. Different types of disasters can also overlap, forming compound disasters, further exacerbating their complexity and destructive power.

[0003] Banking branches are widely distributed, from urban centers to remote villages. While this layout improves the coverage of financial services, it also exposes banks to greater risks from natural disasters such as earthquakes, floods, and typhoons. Natural disasters such as floods, typhoons, and earthquakes can damage banks' physical facilities, disrupting normal operations and even threatening the safety of customers and employees. Emergencies such as public security incidents can also disrupt bank operations. In these situations, banks must quickly activate emergency plans and implement effective response measures to mitigate losses, restore operations, and protect the safety of employees and property.

[0004] Traditional disaster response relies primarily on manual monitoring and delayed responses, making it difficult to obtain accurate early warning information and leverage multi-channel outbound calls. This lack of timeliness and precision leads to fragmented prevention and control measures and inefficient response. Therefore, strengthening risk management and improving emergency response capabilities are crucial for banks. Summary of the Invention

[0005] The present invention provides a meteorological disaster early warning method, device, equipment and storage medium to solve the problem in the prior art that banks lack emergency response capabilities to disasters.

[0006] According to one aspect of the present invention, a meteorological disaster early warning method is provided, the method comprising:

[0007] Determine the disaster category and warning level of meteorological warning data;

[0008] Inputting the disaster category and the warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data;

[0009] Inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding early warning method;

[0010] Each disaster response task is sent to a corresponding contact person in a corresponding early warning manner, so that the contact person is assigned the disaster response task and performs it.

[0011] According to another aspect of the present invention, a meteorological disaster warning device is provided, the device comprising:

[0012] A determination module is used to determine the disaster category and warning level of meteorological warning data;

[0013] A first input module is used to input the disaster category and the warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data;

[0014] A second input module is configured to input the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding warning method;

[0015] The execution module is used to send each emergency call task to the corresponding contact in a corresponding early warning manner, so that the contact is assigned the emergency call task and executes it.

[0016] According to another aspect of the present invention, there is provided an electronic device, comprising: at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the meteorological disaster warning method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the meteorological disaster warning method described in any embodiment of the present invention when executed.

[0020] The embodiments of the present invention provide a meteorological disaster warning method, apparatus, device, and storage medium. The method includes: determining the disaster category and warning level of meteorological warning data; inputting the disaster category and warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data; inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding warning method; and sending each disaster response task to a corresponding contact person in a corresponding warning method so that the contact person assigns the disaster response task and executes it. This method improves the ability to respond to meteorological disasters by classifying meteorological warning data and generating disaster response tasks through a model, and sending each disaster response task to a corresponding contact person in a corresponding warning method. This solves the problem of banks' lack of emergency response capabilities to disasters in the prior art.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A schematic diagram of a flow chart of a meteorological disaster early warning method provided in the first embodiment of the present invention;

[0024] Figure 2 A schematic diagram of a flow chart of a meteorological disaster early warning method provided by an embodiment of the present invention;

[0025] Figure 3 A schematic diagram of issuing a disaster response task provided by an embodiment of the present invention;

[0026] Figure 4 A schematic structural diagram of a meteorological disaster warning device provided in Embodiment 2 of the present invention;

[0027] Figure 5 Schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention. It should be understood that the various steps described in the method implementation mode of the present invention can be performed in different orders and / or in parallel. In addition, the method implementation mode may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0029] As used herein, the term "including" and its variations are open-ended, i.e., "including but not limited to." The term "based on" means "based, at least in part, on." The term "one embodiment" means "at least one embodiment," the term "another embodiment" means "at least one additional embodiment," and the term "some embodiments" means "at least some embodiments." Other terms are defined in the following description.

[0030] It should be noted that the terms "first," "second," and the like in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, any variations of the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0031] It should be noted that the modifications of "one" and "multiple" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0032] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0033] Traditional disaster management models face significant bottlenecks: First, access to pre-emptive warning information is delayed and fragmented, relying on manual monitoring and difficult to effectively link with the business data of grassroots banks. This leads to a disconnect between response decisions and the actual impact of risks. Second, emergency response relies on empirical, static plans that lack the ability to adapt to the dynamics of disasters, making it difficult to quickly adjust strategies and resulting in inefficient response. Third, in-flight resource allocation is inefficient, with information asymmetry and poor communication often delaying the deployment of emergency supplies and rescue teams, hindering rescue effectiveness. Fourth, in terms of organizational command, problems such as information decay, insufficient authority, and poor communication exist across the five levels of the bank (head office, first-tier branches, second-tier branches, first-tier sub-branches, and business outlets), hindering overall emergency response effectiveness. Fifth, post-disaster review and strategy optimization lack data support. Most banks rely solely on crude assessments of losses or disruption duration through manual calculations, failing to establish a closed-loop feedback loop of "warning-response-optimization." This makes it difficult to continuously improve prevention and control efficiency when similar disasters recur.

[0034] For example, the existing patent CN119539333A proposes a disaster emergency rescue command system and method, including: an information collection module: the information collection module collects information on the disaster site through various sensors and communication equipment, including the disaster situation, personnel distribution, and environmental conditions; the collected information is input into the system to obtain an accurate disaster assessment; a decision support module: the decision support module uses a linear programming model to determine the optimal allocation of rescue resources based on the disaster assessment provided by the information collection module, and uses a graph theory algorithm to plan the rescue path; a command and control module: the command and control module is used to convert the rescue decisions of the decision support module into specific rescue instructions. However, this method still has the following problems: information collection relies on sensors and other equipment. If the data is biased or erroneous, it will lead to inaccurate disaster assessment, affect the decision-making of the decision support module, and then affect the rationality of the rescue instructions; the system is highly complex, and the system contains multiple modules and complex algorithms. It is difficult to maintain and manage, and requires high technical skills from operators. Once the system fails, it may be difficult to repair it in time, affecting the rescue work; it lacks flexibility. In actual rescue, the situation may change rapidly. The decisions made by the system based on fixed models and algorithms may not be able to adapt to changes in time and lack sufficient flexibility to deal with emergencies; it cannot provide guidance and reference for emergency actions in the banking industry.

[0035] In response to the above-mentioned problems of the prior art, the present invention proposes an iterative meteorological disaster warning and disposal method for the banking industry to achieve "smart emergency response". First, in the pre-warning stage, based on real-time meteorological data and combined with bank branch distribution data, accurate positioning and second-level push of disaster warnings are achieved to ensure that warning information is promptly and accurately conveyed to relevant emergency contacts. Second, in the emergency response stage after the incident, a dynamic disposal rule base is established, and reinforcement learning algorithms are used to continuously optimize the disposal strategy. According to the type, intensity and dynamic evolution path of the disaster, adaptive emergency response plans are automatically generated to improve response efficiency and accuracy. Third, in terms of organization and command, a centralized and intelligent emergency command system is constructed, covering five levels from the head office to business outlets, clarifying the responsibilities of each level, optimizing the information transmission process, reducing information attenuation and misunderstanding, and improving overall emergency response efficiency. Fourth, in the post-event review and strategy optimization stage, a closed-loop feedback mechanism is established to conduct detailed assessments of disaster losses and emergency response processes, build a "warning-handling-optimization" link, and continuously improve prevention and control capabilities. The details are as follows:

[0036] Example 1

[0037] Figure 1This is a flow chart of a meteorological disaster warning method provided in Example 1 of the present invention. The method can be applied to the banking industry to warn and respond to meteorological disasters. The method can be executed by a meteorological disaster warning device, wherein the device can be implemented by software and / or hardware and is generally integrated on an electronic device. In this embodiment, the electronic device includes but is not limited to: computers and other devices.

[0038] like Figure 1 As shown, a meteorological disaster early warning method provided by the first embodiment of the present invention includes the following steps:

[0039] S110. Determine the disaster category and warning level of the meteorological warning data.

[0040] Meteorological warning data is a standardized collection of information released by meteorological departments based on monitoring, analysis, and assessment of potential or ongoing meteorological disasters. Meteorological warning data can be obtained from meteorological observatories. Disaster categories and warning levels can be set based on actual circumstances.

[0041] In this embodiment, the disaster category and warning level of the acquired meteorological warning data may be determined.

[0042] In one embodiment, determining the disaster category and warning level of meteorological warning data includes: obtaining meteorological warning data; classifying the meteorological warning data based on preset classification and grading standards to obtain the disaster category and warning level of the meteorological warning data.

[0043] Among them, the preset classification and grading standards can be pre-set standards for classifying meteorological warning data.

[0044] In this embodiment, meteorological warning data can be obtained from various meteorological stations, and the meteorological warning data can be classified based on preset classification and grading standards to obtain the disaster category and warning level of the meteorological warning data.

[0045] In one embodiment, the preset classification and grading standards include preset classifications and preset grades. The preset classifications include heavy rain, hail, lightning, high temperature, drought, blizzard, cold wave, frost, icy roads, typhoon, strong wind, heavy fog, sandstorm and haze. The preset grades include first grade, second grade, third grade and fourth grade.

[0046] In this embodiment, the preset classification of meteorological disasters is divided into 14 categories: heavy rain, hail, lightning, high temperature, drought, blizzard, cold wave, frost, icy roads, typhoon, strong wind, heavy fog, sandstorm, and haze. The preset classification of meteorological disasters is divided into four levels: first level, second level, third level, and fourth level. The first level, second level, third level, and fourth level can be red, orange, yellow, and blue from high to low.

[0047] S120: Input the disaster category and the warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data.

[0048] The monitoring and early warning model can be a model that issues early warnings based on meteorological early warning data. The defense guide can be formulated for specific meteorological disasters, early warning levels, and different business locations.

[0049] In this embodiment, a monitoring and early warning model can be pre-built, allowing disaster categories and warning levels to be input into the model to generate defense guidelines corresponding to meteorological warning data. Through monitoring and early warning, various factors that may trigger emergencies can be continuously observed, detected, and analyzed, and alerts can be issued promptly when risks reach a certain threshold. This integration of data collection, analysis, and information dissemination aims to achieve early detection, accurate assessment, and timely response to potential risks, providing strong support for disaster prevention and mitigation efforts.

[0050] For example, the system can classify and grade various meteorological disasters by using customized risk assessment methods, establish monitoring and early warning rule models, accurately classify different meteorological disasters, and realize personalized defense guidelines for various disasters (heavy rain, hail, lightning, high temperature, drought, blizzard, cold wave, frost, road icing, typhoon, strong wind, heavy fog, sandstorm, haze), various levels of warnings (red, orange, yellow, blue), and various locations of the bank (business outlets, office buildings, vaults, self-service banks, etc.).

[0051] S130: Input the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding early warning method.

[0052] A disaster response task can refer to a specific disaster prevention action instruction generated based on a preparedness guideline and requiring execution by a specific responsible party when a meteorological disaster is about to occur or is nearing its impact. A disaster response task can include task objectives, executors, timelines, task priorities, and associated criteria. An early warning method can refer to the method by which a disaster response task is delivered. For example, this can be via one or more of a mobile phone, landline, computer, email, text message, and robocall.

[0053] In this embodiment, a disaster response model can be pre-built and defense guidelines can be input into the model to generate at least one disaster response task and corresponding warning method. Disaster response can refer to a mechanism that links disaster warning and emergency response. When a disaster is about to occur or in its early stages, an intelligent system automatically triggers pre-set emergency instructions and tracks in real time the receipt of feedback from key personnel, the availability of resources, and the execution status of response actions. This creates a closed-loop control system of "warning-instruction-execution-feedback" to minimize losses caused by disasters.

[0054] In one embodiment, the step of inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding warning method includes: inputting the defense guide into a disaster response model, wherein the defense guide includes the disaster type, warning level, warning location, response area, and response channel of meteorological warning data; matching a corresponding response template based on the disaster type, warning level, warning location, and response area; generating at least one disaster response task based on the response template; and determining the warning method corresponding to the disaster response task based on the response channel.

[0055] Warning locations can include business outlets, office buildings, vaults, and self-service banks. Response regions can refer to provinces in different regions. Response channels can include text messages, emails, intelligent robots, system networks (websites), and mobile applications (APPs). Response templates can refer to defense guidelines corresponding to meteorological disaster warnings.

[0056] In this embodiment, the defense guide can be input into the disaster response model, and the corresponding response template can be matched based on the disaster type, warning level, warning location and response area. At least one disaster response task is generated based on the response template, and the warning method corresponding to the disaster response task is determined based on the response channel, thereby determining how to allocate the disaster response tasks and how to issue warnings for the disaster response tasks.

[0057] For example, the system establishes a disaster response model for 14 disaster types, four warning levels, four warning locations, 34 provincial regions, and five response channels. This model automatically configures meteorological disaster response based on disaster type, warning level, location, response region, and response channel, meeting the specific disaster prevention priorities in different regions. It also automatically matches relevant prevention guidelines and flexibly generates disaster response tasks. Customized response templates for different channels, including SMS, email, and intelligent robot outbound calls, transform warning information from a push-based response to a targeted response.

[0058] S140: Send each disaster response task to a corresponding contact person in a corresponding early warning manner, so that the contact person is assigned the disaster response task and executes it.

[0059] In this embodiment, the emergency response task can be sent to the corresponding contact in a corresponding early warning manner, so that the contact can assign the received emergency response task and notify the corresponding personnel to execute it.

[0060] A meteorological disaster warning method provided in a first embodiment of the present invention includes: determining the disaster category and warning level of meteorological warning data; inputting the disaster category and warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data; inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding warning method; and sending each disaster response task to a corresponding contact person in a corresponding warning method so that the contact person assigns the disaster response task and executes it. This method improves the ability to respond to meteorological disasters by classifying meteorological warning data and generating disaster response tasks through a model, and sending each disaster response task to a corresponding contact person in a corresponding warning method, thereby solving the problem of banks' lack of emergency response capabilities to disasters in the prior art.

[0061] Based on the above embodiment, a modified embodiment of the above embodiment is proposed. It should be noted that, in order to simplify the description, only the differences from the above embodiment are described in the modified embodiment.

[0062] In one embodiment, after the disaster response task is executed, the method further includes: determining whether to conduct on-site disposal based on the disaster type at the disaster site, the on-site conditions at the warning site, and on-site feedback; if so, generating an on-site disposal plan based on an on-site processing task allocation model; atomically splitting the on-site disposal plan to obtain different on-site disposal tasks; and distributing the on-site disposal tasks to different staff members so that the staff members can execute them.

[0063] The disaster site can be the location of the site warned. On-site conditions can refer to the real-time situation at the warning site, and on-site feedback can be information provided by on-site staff. An on-site response plan can refer to specific on-site response measures and operational procedures developed for a specific location or type of emergency. These plans detail the emergency actions, personnel division of labor, and material usage to be taken at the scene of the emergency. These plans can provide clear guidance to on-site emergency personnel and mitigate the further development and escalation of the incident.

[0064] In this embodiment, whether to conduct on-site disposal can be determined based on the disaster type at the disaster site, the on-site conditions at the warning site and on-site feedback. If so, an on-site disposal plan can be generated based on the on-site processing task allocation model, and the on-site disposal plan can be atomically split to obtain different on-site disposal tasks, and the on-site disposal tasks can be distributed to different staff members for execution.

[0065] In one embodiment, distributing the on-site disposal tasks to different staff members includes: dividing the on-site disposal tasks into different task lists according to different work groups; and for each work group, allocating the on-site disposal tasks in the task list to the staff members in the work group.

[0066] Each task list may include multiple on-site disposal tasks.

[0067] In this embodiment, the on-site disposal tasks can be divided into different task lists according to different working groups. For each working group, the on-site disposal tasks in the task list corresponding to the working group can be assigned to the staff in the working group to clarify the division of labor of each staff member.

[0068] This embodiment establishes an on-site processing task allocation model (such as an on-site disposal rule base) based on a uniformly formulated on-site disposal plan template. Grassroots organizations can customize the content of their own on-site disposal plans based on their own emergency drills and actual emergency disposal practices. Based on this rule base, in the subsequent disaster disposal process, the reinforcement learning algorithm can be used to iteratively update the rule base. At the same time, combined with the historical problem solving situation and the manual supplementation of emergency disposal personnel, the disposal rule base can be continuously upgraded and optimized. When a meteorological disaster occurs and requires grassroots emergency contacts to conduct on-site disposal, the system will retrieve the rule base in real time to generate the corresponding on-site disposal plan for this organization. After the on-site disposal plan is atomically split, it is split into task lists according to different working groups, and the work tasks are clearly assigned to the people in the group.

[0069] For example, Figure 2 A schematic diagram of a meteorological disaster warning method provided by an embodiment of the present invention is shown in FIG. Figure 2 As shown, the system automatically pushes corresponding defense guidelines for each scenario (such as business outlets, office buildings, vaults, and off-site self-service banks) for 14 types of high-level meteorological disaster warnings (red and orange). The defense tasks in the guidelines are linked to the meteorological disaster, automatically matching and generating immediate disaster response tasks. Based on the meteorological disaster warning signal, the system automatically analyzes the affected area, identifies the personnel to be notified, and automatically completes targeted push notifications based on configured templates.

[0070] After receiving a disaster warning, the pre-disaster response model can be used to automatically splice and modify SMS templates, robot outbound call templates, and email templates to generate personalized reminder content. According to the configured call rules, the emergency contact corresponding to the task can be called. For example, through the monitoring and early warning model and the pre-disaster response model, the red warning is linked with the "pre-disaster response" module, supporting the review of warning messages and defense work tasks through SMS, robot outbound calls, system web terminals, mobile APPs, emails, and other methods, and feedback on the completion of defense execution. For orange warnings, support is provided for reviewing warning messages and defense work tasks through the system web terminals, mobile APPs, emails, etc., which mainly serve as prompts and guidance, and grassroots banks do not need to provide feedback. Figure 3 A schematic diagram of issuing a disaster response task provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, for yellow alerts, the system can send warnings through channels such as the web and mobile app. For red alerts, the system can send warnings via SMS, robot calls, the system web, mobile app, and email. After receiving the disaster response task, the warning site can handle it independently, notify the emergency contact of the superior agency, and log in to the system web or mobile app to report the completion status of the task. The superior agency's emergency contact can track the progress of the task and promptly remind any unanswered questions by the deadline.

[0071] After receiving the task dispatch notification, the emergency contact at the warning site can log in to the system through multiple channels to handle the task. The system automatically generates the corresponding disaster response task based on the disaster response model. Operators must operate according to the corresponding task requirements and complete the disaster response task feedback. The system then automatically identifies the site type and environment of the on-site agency. Combined with the work feedback of specific on-site personnel, it determines whether on-site disposal tasks need to be generated and distributed based on the on-site disposal task allocation model. On-site personnel will receive pending tasks and can perform on-site response operations based on the corresponding on-site disposal tasks. At the same time, task monitoring personnel can track and view the execution status of on-site tasks at any time. The system's series of automatic operations can greatly shorten the emergency response time of grassroots agency personnel. Statistical analysis shows that the emergency response efficiency of grassroots emergency personnel can be improved by 90%.

[0072] In one embodiment, after the on-site handling task is executed, the method also includes: obtaining the on-site handling task to set the execution status after the execution action is executed; analyzing the execution status based on the reinforcement learning algorithm to obtain the reward value of the execution action; and adjusting the on-site processing task allocation model based on the reward value.

[0073] The set execution action can refer to the method for completing an on-site disposal task. The execution status can refer to the situation that occurs after the on-site disposal task is executed. Reinforcement learning is a branch of machine learning. Its core concept is to enable an intelligent agent (Agent) to learn how to take the optimal action (Action) in a specific scenario through interaction with the environment to maximize long-term accumulated rewards (Reward). Its core characteristics are "trial and error learning" and "dynamic decision-making."

[0074] In this embodiment, an on-site disposal task can be obtained to set the execution status after the execution action is executed. The execution status can be analyzed based on the reinforcement learning algorithm to obtain the reward value of the execution action, thereby adjusting the on-site processing task allocation model based on the reward value to improve the accuracy of the on-site processing task allocation model in allocating tasks.

[0075] For example, this embodiment can use the Q-learning algorithm as a reinforcement learning algorithm. The Q-learning algorithm is a model-free reinforcement learning algorithm that aims to learn an optimal strategy to maximize cumulative rewards through interaction with the environment. This method maintains a Q table (state-action value table) to record the expected benefits of taking actions in each state. The formula is as follows:

[0076]

[0077] In this embodiment, the agent represents the disaster warning system and is responsible for selecting the optimal disposal action (such as closing outlets and dispatching supplies) based on real-time data. (Environment) includes the external disaster environment (such as typhoon path, rainfall changes) and the internal business environment (such as outlet operation status and resource inventory). State refers to the comprehensive disaster parameters (wind speed, rainfall, water depth) and system parameters (equipment status, customer flow), which constitute a multi-dimensional state vector. Action refers to the set of disposal instructions that can be executed by the system. Reward refers to the feedback value dynamically calculated based on the disposal effect, which quantifies the pros and cons of the strategy.

[0078] Table 1 Mapping relationship between algorithm parameters and dimensions

[0079]

[0080] As shown in Table 1, the table shows the mapping relationship between the various parameters of the above formula and the disaster mapping dimensions. For these dimensions, the strategy can be updated through the following steps:

[0081] (1) Observation status: The agent perceives the current environmental status (such as typhoon path prediction and network device status).

[0082] (2) Select an action: Select an action (such as starting the backup power supply) based on the policy.

[0083] (3) Execution of action: The action acts on the environment, and the state of the environment changes.

[0084] (4) Obtaining rewards: environmental feedback reward value (e.g., reducing losses by 500,000 yuan → high reward).

[0085] (5) Update strategy: Adjust the strategy based on the reward to improve the quality of future decisions.

[0086] By using the reinforcement learning Q-learning algorithm, accessing sensor information from relevant equipment, and performing intelligent iteration, we can improve the accuracy of task allocation in the on-site task allocation model. Taking a typhoon disaster scenario as an example:

[0087] (1) Initial status: The typhoon path is predicted to land around network point A, with a wind speed of level 10 and a rainfall of 60 mm / h. The wind resistance level of the building at network point A is level II.

[0088] (2) The agent selects an action: According to the current Q table, it selects the action of “closing network A”.

[0089] (3) Environmental feedback reward: After closing the outlet, the actual loss is reduced by 300,000 yuan → reward R = 30*10 = 300 points.

[0090] (4) The environment shifted to a new state: After the typhoon actually made landfall, the wind speed increased to level 12, and the depth of water accumulation around site A reached 0.5 meters.

[0091] (5) The agent evaluates the next state: calculation When the water level is 0.5 meters, the optimal action is to "transfer personnel and equipment" (assuming the maximum Q value is 250).

[0092] (6) Update Q value: For example, if the original Q value Q(s,a) = 200, the learning rate α = 0.2, and the discount factor γ = 0.9, then the calculation is based on the formula:

[0093]

[0094] Q(s,a)=200+0.2*[300+0.9*250-200]=200+0.2*(300+225-200)=200+65=265;

[0095] The new Q value of 265 indicates that the value of "closing outlet A" in this state has increased, making this action more likely to be chosen in the future. Through continuous iteration of the above algorithm, an enhanced closed loop of "warning-action-optimization" can be formed. Compared to the Deep Q-Network (DQN) algorithm, the Q-learning algorithm in this embodiment requires fewer computing resources, does not require a dedicated computing cluster, and has low latency. By designing a set of quantifiable indicators for scoring and weighting, system optimization can be incorporated into departmental assessments for subsequent incentive effects. By updating and optimizing the rule base based on historical problem resolution, preventive measures can be improved. For example, if the original rule stipulates "close outlets when rainfall ≥ 80 mm / h," and if rainfall during a disaster falls below this threshold, and if reports from outlet sensors or emergency response personnel indicate that there are problems such as flooding or short circuits, the system will iterate based on the rainfall information obtained for this location, lowering the rainfall warning threshold for this location.

[0096] The method of this embodiment utilizes multimodal meteorological data fusion and assimilation technology for real-time monitoring of 14 natural disasters, including rainstorms, typhoons, and cold waves. This strengthens the overlay analysis of meteorological observation data, forecast data, and disaster data with over 20,000 business outlets, branches, and key regions nationwide, improving the timeliness and accuracy of early warning monitoring and rapidly gathering and grasping on-site disaster information for emergencies. Multimodal meteorological data fusion and assimilation technology standardizes and integrates meteorological data (of different modalities) from multiple sources, including satellites, radar, ground observation stations, and numerical forecasts. Through spatiotemporal calibration, quality verification, and dynamic weighting, conflicts in format, accuracy, and time differences between data are eliminated, generating a highly confident unified meteorological dataset to support accurate disaster early warning. Natural language processing technology can also be used simultaneously to implement an automatic response mechanism directly to grassroots emergency personnel through intelligent data analysis. By atomically decomposing on-site response tasks, grassroots emergency personnel can clearly understand their work in emergencies, significantly improving emergency response efficiency. Furthermore, reinforcement learning algorithms are used to iterate the response rule base, continuously upgrading and optimizing the response rules and improving their ability to perceive, predict, and prevent risk factors.

[0097] This method is based on multimodal meteorological data analysis to establish a monitoring and early warning rule model. According to the disaster type, response area, and response channel, it realizes personalized configuration of automatic disaster response, meets the focus of disaster prevention in different regions, and automatically matches relevant defense guidelines to generate differentiated early warning plans and execute push. By establishing a disaster response rule model, integrating multi-source meteorological data with the banking business system, disaster warnings can be parsed and distributed at the millisecond level, and dynamically pushed to emergency contacts of affected outlets. At the same time, relevant personnel are called in different time periods and channels, and response efficiency is improved by 90%. By collecting disaster loss and disposal data in real time, based on a unified on-site disposal plan template, the reinforcement learning algorithm is used to iterate the disposal rule base. At the same time, combined with self-built rules, through the analysis of historical problem solving and the manual supplementation of emergency disposal personnel, the disposal rules are continuously upgraded and optimized.

[0098] Example 2

[0099] Figure 4 This is a structural schematic diagram of a meteorological disaster warning device provided in Example 2 of the present invention. The device can be used in the banking industry to warn and respond to meteorological disasters. The device can be implemented by software and / or hardware and is generally integrated into an electronic device.

[0100] like Figure 4 As shown, the device includes:

[0101] Determination module 210, for determining the disaster category and warning level of the meteorological warning data;

[0102] The first input module 220 is used to input the disaster category and the warning level into the monitoring and warning model to obtain the defense guide corresponding to the meteorological warning data;

[0103] The second input module 230 is used to input the defense guide into the disaster response model to obtain at least one disaster response task and a corresponding warning method;

[0104] The execution module 240 is used to send each disaster response task to a corresponding contact in a corresponding early warning manner, so that the contact is assigned the disaster response task and executes it.

[0105] This embodiment provides a meteorological disaster warning device, including: a determination module for determining the disaster category and warning level of meteorological warning data; a first input module for inputting the disaster category and the warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data; a second input module for inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding warning method; an execution module for sending each disaster response task to a corresponding contact person in a corresponding warning method, so that the contact person assigns the disaster response task and executes it. By classifying meteorological warning data and generating disaster response tasks through a model, and sending each disaster response task to a corresponding contact person in a corresponding warning method, the response capability to meteorological disasters can be improved, solving the problem of banks' lack of emergency response capability to disasters in the prior art.

[0106] Furthermore, the determination module 210 includes:

[0107] Obtain weather warning data;

[0108] The meteorological warning data is classified based on a preset classification and grading standard to obtain the disaster category and warning level of the meteorological warning data.

[0109] Furthermore, the preset classification and grading standards include preset classifications and preset grades. The preset classifications include heavy rain, hail, lightning, high temperature, drought, blizzard, cold wave, frost, icy roads, typhoon, strong wind, heavy fog, sandstorm and haze. The preset grades include first grade, second grade, third grade and fourth grade.

[0110] Furthermore, the second input module 230 includes:

[0111] Inputting the defense guide into a disaster response model, wherein the defense guide includes the disaster type, warning level, warning location, response area, and response channel of the meteorological warning data;

[0112] Matching a corresponding response template based on the disaster type, warning level, warning location, and response area;

[0113] generating at least one disaster response task based on the response template;

[0114] An early warning method corresponding to the disaster response task is determined based on the response channel.

[0115] Furthermore, the execution module 240 further includes:

[0116] Determine whether to conduct on-site disposal based on the disaster type at the disaster site, the on-site conditions at the warning site, and on-site feedback;

[0117] If so, an on-site disposal plan is generated based on the on-site processing task allocation model;

[0118] Atomizing and splitting the on-site disposal plan to obtain different on-site disposal tasks;

[0119] The on-site handling tasks are distributed to different workers so that the workers can perform them.

[0120] Furthermore, distributing the on-site handling tasks to different staff members includes:

[0121] Divide the on-site disposal tasks into different task lists according to different working groups;

[0122] For each work group, the on-site handling tasks in the task list are assigned to the staff members in the work group.

[0123] Furthermore, after the on-site disposal task is executed, the device further includes:

[0124] Obtaining the on-site disposal task to set the execution status after the execution action is executed;

[0125] Analyzing the execution status based on a reinforcement learning algorithm to obtain a reward value for the execution action;

[0126] The field processing task allocation model is adjusted based on the reward value.

[0127] The above-mentioned meteorological disaster warning device can execute the meteorological disaster warning method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0128] Example 3

[0129] Figure 5 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0130] like Figure 5As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0131] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0132] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the meteorological disaster early warning method.

[0133] In some embodiments, the meteorological disaster warning method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the meteorological disaster warning method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the meteorological disaster warning method by any other appropriate means (e.g., by means of firmware).

[0134] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0135] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0136] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0137] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0138] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0139] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0140] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0141] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A meteorological disaster early warning method, characterized in that: The method comprises: Determine the disaster category and warning level of meteorological warning data; Inputting the disaster category and the warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data; Inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding early warning method; Each disaster response task is sent to a corresponding contact person in a corresponding early warning manner, so that the contact person is assigned the disaster response task and performs it.

2. The method according to claim 1, characterized in that Determining the disaster category and warning level of meteorological warning data includes: Obtain weather warning data; The meteorological warning data is classified based on a preset classification and grading standard to obtain the disaster category and warning level of the meteorological warning data.

3. The method according to claim 2, characterized in that The preset classification and grading standards include preset classifications and preset grades. The preset classifications include heavy rain, hail, lightning, high temperature, drought, blizzard, cold wave, frost, icy roads, typhoon, strong wind, heavy fog, sandstorm and haze. The preset grades include first grade, second grade, third grade and fourth grade.

4. The method according to claim 1, wherein Inputting the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding early warning method includes: Inputting the defense guide into a disaster response model, wherein the defense guide includes the disaster type, warning level, warning location, response area, and response channel of the meteorological warning data; Matching a corresponding response template based on the disaster type, warning level, warning location, and response area; generating at least one disaster response task based on the response template; An early warning method corresponding to the disaster response task is determined based on the response channel.

5. The method according to claim 1, wherein After the disaster response task is executed, the method further includes: Determine whether to conduct on-site disposal based on the disaster type at the disaster site, the on-site conditions at the warning site, and on-site feedback; If so, an on-site disposal plan is generated based on the on-site processing task allocation model; Atomizing and splitting the on-site disposal plan to obtain different on-site disposal tasks; The on-site handling tasks are distributed to different workers so that the workers can perform them.

6. The method according to claim 5, characterized in that The on-site handling tasks are distributed to different staff members, including: Divide the on-site disposal tasks into different task lists according to different working groups; For each work group, the on-site handling tasks in the task list are assigned to the staff members in the work group.

7. The method according to claim 5, characterized in that After the on-site disposal task is executed, the method further includes: Obtaining the on-site disposal task to set the execution status after the execution action is executed; Analyzing the execution status based on a reinforcement learning algorithm to obtain a reward value for the execution action; The field processing task allocation model is adjusted based on the reward value.

8. A meteorological disaster warning device, characterized in that: The device comprises: A determination module is used to determine the disaster category and warning level of meteorological warning data; A first input module is used to input the disaster category and the warning level into a monitoring and warning model to obtain a defense guide corresponding to the meteorological warning data; A second input module is configured to input the defense guide into a disaster response model to obtain at least one disaster response task and a corresponding warning method; The execution module is used to send each emergency call task to the corresponding contact in a corresponding early warning manner, so that the contact is assigned the emergency call task and executes it.

9. An electronic device, characterized in that: The device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the meteorological disaster warning method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the meteorological disaster warning method according to any one of claims 1 to 7 when executed.

Images