Urban infrastructure risk assessment method for storm disaster chain

Abstract

The application discloses a kind of urban infrastructure risk assessment methods for storm disaster chain, specifically comprising the following steps: S1, the urban infrastructure element is uniformly coded;S2, the associated network between urban infrastructure nodes is constructed;S3, the urban storm disaster chain network is constructed;S4, the secondary disaster risk caused by storm disaster is simulated;S5, the infrastructure risk caused by storm disaster chain is simulated;S6, the cascading failure conduction of infrastructure node risk is evaluated;S7, real-time risk display is realized based on uniform identification code;The application relates to emergency technology and computer technology field.The urban infrastructure risk assessment method for storm disaster chain, by considering the infrastructure node risk caused by storm and its secondary disaster, provides support for the dynamic emergency decision of multi-disaster linkage, realizes the fine infrastructure node risk prediction of time and point, and helps to make specific emergency decision and targeted disaster risk prevention of residents.

Description

Technical Field

[0001] This invention relates to the fields of emergency technology and computer technology, specifically to a method for risk assessment of urban infrastructure in response to rainstorm disaster chains. Background Technology

[0002] In the field of emergency management, risk assessment of urban infrastructure is a fundamental and important research area. Accurate risk assessment of urban infrastructure is a crucial basis for emergency decision-making and resource allocation, and also a vital foundation for ensuring the safety of urban residents' lives and property and promoting sustainable urban development.

[0003] Traditional urban infrastructure risk assessments mostly operate at the macro-system level, neglecting to incorporate spatiotemporal factors and fail to provide detailed risk assessment results for specific infrastructure nodes. Existing micro-level infrastructure risk assessments often focus only on single types of disaster impacts or randomly set initial failure points, and only consider the cascading failure propagation between infrastructure nodes. Furthermore, existing infrastructure risk assessment methods lack integration with computer technology to intelligently display and apply risk forecast results. With urbanization, population growth, and the increasing complexity and diversity of disaster-causing factors and environments, these single-disaster, single-time-point, macro-level risk assessment and emergency response models will struggle to cope with multi-disaster, intelligent, and refined disaster prevention scenarios.

[0004] Based on the above data, a risk assessment method for urban infrastructure oriented towards rainstorm disaster chains is proposed, which considers not only the impact of rainstorms but also the impact of secondary disasters on infrastructure. based on The spatiotemporal dual-dimensional approach provides refined risk assessment results for infrastructure nodes at specific times and locations, offering effective references for advance planning of emergency rescue, refined allocation of emergency resources, dynamic emergency decision-making across multiple disasters, and the implementation of smart disaster prevention and mitigation. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for risk assessment of urban infrastructure oriented towards rainstorm disaster chains, thus solving the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for risk assessment of urban infrastructure oriented towards rainstorm disaster chains, specifically including the following steps:

[0007] S1. Unify the coding of urban infrastructure elements: Select urban infrastructure elements in the target area, assign a unique identity to each urban infrastructure element, and obtain a number of infrastructure nodes with unified identification codes.

[0008] S2. Constructing a network of connections between urban infrastructure nodes: In the GIS system, infrastructure nodes of the same type are constructed into the same layer. The connections between infrastructure nodes are analyzed from the perspectives of geographical location and function. Based on these connections, a complex network model is used to construct infrastructure connection networks within and between layers.

[0009] S3. Constructing an urban rainstorm disaster chain network: Obtain typical historical rainstorm disaster cases in the target area, analyze the occurrence mechanism, pattern and secondary disasters of rainstorm disasters, extract disaster events and basic disaster chains based on text analysis methods, and construct a disaster chain network formed by rainstorms and their secondary disasters using complex network models;

[0010] S4. Simulate the risk of secondary disasters caused by rainstorm disasters: Construct a calculation formula for the risk of secondary disasters caused by rainstorm disasters, measure the local rainstorm risk level, and substitute it into the formula to calculate the temporal evolution of the risk level of each secondary disaster.

[0011] S5. Simulate the infrastructure risks caused by rainstorm disaster chains: Based on the risk level of local rainstorms and the risk level of secondary disasters, construct calculation formulas for the impact of different disasters on different infrastructure nodes, obtain the temporal evolution of the risk level of each infrastructure node, set thresholds for the risk levels of different infrastructure nodes, and use them to assess the working status of infrastructure at fixed times and locations to determine whether it is normal or ineffective, and obtain the spatial distribution of infrastructure nodes that fail at multiple time points.

[0012] S6. Assess the cascading failure propagation of infrastructure node risks: Analyze the cascading failure propagation mechanism between different infrastructure nodes, simulate the failure propagation process and impact range caused by the failure of infrastructure nodes at multiple time points in different disaster scenarios, and form the final timed, location-specific, refined and dynamic infrastructure node risk assessment results.

[0013] S7. Real-time risk display based on unified identification code: Import the risk level and working status information of each infrastructure node calculated in S5 and S6 into the urban infrastructure node information database, build an urban infrastructure node rainstorm disaster chain risk early warning and forecasting service platform, and realize real-time intelligent forecasting and early warning of any urban infrastructure node before disaster through unified identification code.

[0014] The present invention is further configured such that the urban infrastructure elements in S1 include power infrastructure elements, communication infrastructure elements, road infrastructure elements, and subway infrastructure elements.

[0015] The present invention is further configured such that the encoding rules of the unified identification code in S1 include: management subject + object category + individual code.

[0016] The present invention is further configured such that: the method for measuring the level of localized rainstorm risk in S4 includes:

[0017] Acquire rainfall data, urban elevation, drainage network and land use data published in weather forecasts for the target area, divide the target area into small grids, and simulate real-time water accumulation data in each small grid area as an indicator to measure the risk of local rainstorms.

[0018] The present invention is further configured such that the calculation formula for the risk of secondary disasters caused by rainstorm disasters in S4 includes:

[0019]

[0020] In the formula, S j→i (t) represents the node risk, α j→i ε represents the relationship strength factor, r represents disaster intervention capability, and ε represents the disaster intervention capability. j→i A value of 1 or 0 indicates whether there is government intervention between the two disaster nodes, respectively. j→i β represents the probability that a parent disaster event will trigger a child disaster event. t-tj→i This represents the delay coefficient, used to avoid tt when judging the function. j→i When the value is less than 0, the degree of disaster loss becomes negative, and t represents the time of disaster evolution.

[0021] The present invention is further configured such that the calculation formulas for the impact of different disasters on different infrastructure nodes in S5 include:

[0022]

[0023] In the formula, X i This refers to the damaged state of infrastructure node i. B ki H represents the connection strength between the disaster-causing factor k and infrastructure node i. k t represents the risk level of disaster-causing factor k. ki τ and β are the damping coefficient and time delay coefficient, respectively representing the time delay in the damage to infrastructure caused by the disaster-causing factor and the delay in the spread of loss. i ξ represents the self-recovery capability of infrastructure nodes. i (t) represents the random term, θ represents the threshold, and when the damage state of an infrastructure node exceeds the threshold, it is defined as a functional failure. The normal or failed working state of an infrastructure node is represented by F, respectively. i It can be represented by 1 or 0.

[0024] The present invention is further configured such that: the method for obtaining the spatial distribution of infrastructure nodes that fail at multiple time points in S5 includes:

[0025] By substituting the risk level of rainstorms and their secondary disasters in each grid of the target area into the calculation formula for the impact of different disasters on different infrastructure nodes, the degree of impact of rainstorm disaster chains on infrastructure nodes and the final working status of infrastructure nodes are obtained, forming a spatial distribution map of failed infrastructure nodes at multiple time points.

[0026] The present invention is further configured such that: the method for realizing real-time intelligent forecasting and early warning of disaster risks for any urban infrastructure node through a unified identification code in S7 includes:

[0027] Input the unified identification code into the urban infrastructure node rainstorm disaster chain risk early warning and forecasting service platform to obtain the risk level and working status of the infrastructure node corresponding to the unified identification code;

[0028] The urban infrastructure node rainstorm disaster chain risk early warning and forecast service platform is integrated with the emergency response system and public notification system of the city where the target area is located. When the disaster risk reaches the target threshold, an alarm is triggered, and forecast and early warning notices are sent to residents and relevant departments in the target area.

[0029] This invention provides a method for risk assessment of urban infrastructure oriented towards rainstorm disaster chains. It has the following beneficial effects:

[0030] This invention provides support for dynamic emergency decision-making in the context of multiple disasters by considering the risks to infrastructure nodes caused by rainstorms and their secondary disasters. It also enables precise, time- and location-specific risk prediction of infrastructure nodes, which helps in the formulation of micro-level emergency decision-making strategies and targeted disaster risk prevention for residents. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the process of the present invention;

[0032] Figure 2 This is a schematic diagram of the infrastructure network associated with this invention;

[0033] Figure 3 This is a schematic diagram of the fault propagation in the infrastructure nodes of this invention. Detailed Implementation

[0034] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0035] Please see Figure 1-3 The present invention provides the following technical solution: a method for risk assessment of urban infrastructure oriented towards rainstorm disaster chains, specifically including the following steps:

[0036] S1. Unified coding of urban infrastructure elements: Select urban infrastructure elements in the target area, including power infrastructure elements, communication infrastructure elements, road infrastructure elements and subway infrastructure elements, assign a unique identity to each urban infrastructure element, and obtain a number of infrastructure nodes with unified identification codes. The unified identification code system for urban infrastructure is constructed using MA codes based on the ISO / IEC 15459 international standard. The coding rules of the unified identification code include: management entity + object category + individual code.

[0037] As explained in detail, the management entities mainly include four nodes: the root node in the ISO standard, the industry root node code, the identifier registration service node code, and the identifier code version. ； The object categories mainly include three nodes: primary and secondary classifications of infrastructure objects, industry classification, and industry standard number. ； Individual coding mainly includes three nodes: industry-unique code, spatial coordinate code, and associated extension code, enabling precise positioning and information management of elements.

[0038] S2. Constructing a network of connections between urban infrastructure nodes: In the GIS system, infrastructure nodes of the same type are constructed into the same layer. The connections between infrastructure nodes are analyzed from the perspectives of geographical location and function. Based on these connections, a complex network model is used to construct infrastructure connection networks within and between layers.

[0039] The specific process includes:

[0040] S201: Collect relevant data on urban infrastructure nodes, including but not limited to detailed information on power plants, substations, charging stations, base stations, primary roads, secondary roads, tertiary roads, subway stations, and rail lines. The relevant data includes but is not limited to the name, management entity, industry classification, geographical location, and functional attributes of the objects. The geographical location is displayed in latitude and longitude. The collected data is sorted and cleaned to ensure the accuracy and consistency of the data.

[0041] S202: In the GIS system, based on the classification of infrastructure nodes, the geographical location and related attribute information of each type of infrastructure node are mapped to the corresponding layer. Each layer is a visual representation of a specific type of infrastructure node.

[0042] S203: Infrastructure nodes within a layer are connected based on real-world relationships, such as connecting road junctions or subway stations based on the direction of roads or subway lines; for infrastructure nodes within the same layer lacking real-world data, an approximate complex network model is used for simulation; infrastructure nodes between layers are connected based on real-world relationships and geographical and functional analysis, such as the interdependence between power and communication systems in a disaster context determining their interconnection; the unidirectional dependency between power and subway systems determining their unidirectional connection; the unidirectional dependency between subway and highway systems determining the unidirectional connection between road nodes within a certain range around subway stations. Integrating intra-layer and inter-layer relationships forms a complete infrastructure node connection network, such as... Figure 2 As shown.

[0043] S3. Constructing an urban rainstorm disaster chain network: Obtain typical historical rainstorm disaster cases in the target area, analyze the occurrence mechanism, pattern and secondary disasters of rainstorm disasters, extract disaster events and basic disaster chains based on text analysis methods, and construct a disaster chain network formed by rainstorms and their secondary disasters using complex network models.

[0044] S4. Simulate the risk of secondary disasters caused by rainstorms: Construct a calculation formula for the risk of secondary disasters caused by rainstorms, measure the level of local rainstorm risk, and substitute it into the formula to calculate the temporal evolution of the risk level of each secondary disaster.

[0045] The specific process includes:

[0046] S401: Obtain the time series data of rainfall forecast for the target area based on weather forecast data, collect urban elevation, land use and drainage network data of the target area, divide the target area into a 200*200 meter grid in Mike software, simulate the real-time water depth in each grid, and measure the rainstorm risk level of the area.

[0047] S402: Constructing the formula for calculating secondary disaster risk:

[0048]

[0049] In the formula, S j→i (t) represents the node risk, α j→i ε represents the relationship strength factor, r represents disaster intervention capability, and ε represents the disaster intervention capability. j→i A value of 1 or 0 indicates whether there is government intervention between the two disaster nodes, respectively. j→i β represents the probability that a parent disaster event will trigger a child disaster event. t-tj→i This represents the delay coefficient, used to avoid tt when judging the function. j→i When the value is less than 0, the degree of disaster loss becomes negative, where t represents the time of disaster evolution;

[0050] Substituting the time series data of rainstorm disaster risk within each grid into the calculation formula for secondary disaster risk, the time series data of the risk level of each secondary disaster in all local spaces are calculated.

[0051] S5. Simulate the infrastructure risks caused by rainstorm disaster chains: Based on the local rainstorm risk level and the risk level of its secondary disasters, construct calculation formulas for the impact of different disasters on different infrastructure nodes, obtain the temporal evolution of the risk level of each infrastructure node, set thresholds for the risk levels of different infrastructure nodes, and use them to assess the working status of infrastructure at fixed times and locations to determine whether it is normal or ineffective. Substitute the risk levels of rainstorms and their secondary disasters in each grid of the target area into the calculation formulas for the impact of different disasters on different infrastructure nodes, obtain the degree of impact of rainstorm disaster chains on infrastructure nodes and the final working status of infrastructure nodes, and form a spatial distribution map of failed infrastructure nodes at multiple time points.

[0052] The calculation formulas for the impact of different disasters on different infrastructure nodes include:

[0053]

[0054] In the formula, X i This refers to the damaged state of infrastructure node i. B ki H represents the connection strength between the disaster-causing factor k and infrastructure node i. k t represents the risk level of disaster-causing factor k. ki τ and β are the damping coefficient and time delay coefficient, respectively representing the time delay in the damage to infrastructure caused by the disaster-causing factor and the delay in the spread of loss. i ξ represents the self-recovery capability of infrastructure nodes. i (t) represents the random term, θ represents the threshold, and when the damage state of an infrastructure node exceeds the threshold, it is defined as a functional failure. The normal or failed working state of an infrastructure node is represented by F, respectively. i It can be represented by 1 or 0.

[0055] S6. Assess the cascading failure propagation of infrastructure node risks: Analyze the cascading failure propagation mechanism between different infrastructure nodes, simulate the failure propagation process and impact range caused by the failure of infrastructure nodes at multiple time points in different disaster scenarios, and form the final timed, location-specific, refined and dynamic infrastructure node risk assessment results.

[0056] The specific process is as follows:

[0057] S601: Analyze the cascading fault propagation mechanism between different infrastructure nodes, such as Figure 3As shown, taking four types of infrastructure layers—power network, communication network, subway network, and road network—as examples, this paper demonstrates the cascading fault propagation mode between related infrastructure layers at a micro level. There is an interdependence between the power and communication layers. When a disaster-induced impact causes a communication node to fail, the failure will propagate to the relevant power node. Conversely, when a power node fails due to a disaster, the failure will propagate to the connected communication node. Specifically, in the communication network, when node C0 fails due to a disaster, it triggers the propagation of cascading faults. In the subsequent stage, the power node E0 that depends on C0 and the edge connecting C0 and E0 are removed simultaneously. In the second stage, neighboring nodes in the network may fail due to overload, so the edges connecting other power nodes such as E1 and E2 to the isolated communication node are also removed. The third stage performs similar processing on the communication network nodes, forming the final working state of the network nodes.

[0058] There is a one-way dependency between power and subway. When a power node fails due to a disaster, the operation of the subway node connected to it will also fail. Conversely, when a subway node fails due to a disaster, the failure will not propagate to the power node. Specifically, when power node E0 fails due to a disaster, the connected subway node M0 will fail, and the edge connecting E0 and M0 will be removed. Conversely, when subway node M0 fails due to a disaster, only the edge connected to it in the subway network is removed, and the failure will not propagate to the power node.

[0059] There is a one-way dependency between the subway and highway systems. When all the highway traffic nodes within a certain range around the subway fail, it may affect the normal entry and exit of passengers at the subway station, and the subway station will also fail. Conversely, when the subway station around the highway fails, it will not affect the normal use of the highway. Specifically, when the road node R0 fails within a certain range of the subway node M0, the normal passage for passengers to enter and exit the subway station may be blocked, causing the subway station to fail. At the same time, the edge connecting R0 and M0 is removed. Conversely, when the subway node M0 fails due to a disaster, only the edge connected to it in the subway network is removed, and this failure will not propagate to the road node.

[0060] S602: Based on the spatial distribution of multi-time-point failure infrastructure nodes calculated in step S5, the cascading fault propagation mechanism described in S601 is used to simulate the fault propagation caused by these failed infrastructure nodes, forming a final, dynamic, and refined assessment result of the infrastructure node risk or working status.

[0061] S7. Real-time risk display based on unified identification code: Import the risk level and working status information of each infrastructure node calculated in S5 and S6 into the urban infrastructure node information database, build an urban infrastructure node rainstorm disaster chain risk early warning and forecasting service platform, and realize real-time intelligent forecasting and early warning of any urban infrastructure node before disaster through unified identification code.

[0062] The specific process is as follows:

[0063] S701: Import the real-time forecast information of risk level and working status of each infrastructure node calculated by S5 and S6 into the urban infrastructure node information database corresponding to each unified identification code in S1, to ensure accurate correspondence and real-time updates of information.

[0064] S702: Develop and deploy a user-friendly mobile application or web platform for urban infrastructure node risk forecasting. As a service platform for early warning and forecasting of rainstorm disaster chains for urban infrastructure nodes, the platform should be able to receive and display data from the aforementioned database. The platform should provide a clear and intuitive interface, enabling users to easily query relevant information by scanning or manually entering the code of the infrastructure node or the corresponding digital QR code of the infrastructure node. The platform should display the risk level and working status of the infrastructure nodes in real time in the form of maps, charts, etc., so that users can intuitively understand the current disaster risk situation.

[0065] To further explain, the platform can also provide personalized disaster early warning information based on the user's location and the infrastructure nodes they are interested in, including the expected downtime and potential impact range. Based on the assessment results and forecast information, it can generate and provide targeted disaster prevention strategy suggestions to help city managers and residents take effective prevention and response measures.

[0066] S703: By cooperating with urban management departments, the platform is integrated with the city's emergency response system and public notification system. When the disaster risk reaches a certain threshold, an alarm is automatically triggered, and an emergency notification is sent to residents and relevant departments in the relevant area to ensure timely information transmission and effective response.

[0067] S704: Utilizing big data analytics and intelligent algorithms, analyze user behavior data and feedback to continuously optimize platform functionality and the accuracy of alerts. For example, adjust the sending time and content of alerts based on user responses to alert information to improve alert effectiveness and user satisfaction.

[0068] Although already described and displayThe embodiments of the present invention have been described. It will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A storm disaster chain oriented urban infrastructure risk assessment method, characterized in that: Specifically comprising the following steps: S1, uniformly coding city infrastructure elements: selecting city infrastructure elements in a target area, assigning a unique identity to each city infrastructure element, and obtaining a number of infrastructure nodes with uniform identification codes; The city infrastructure elements include power infrastructure elements, communication infrastructure elements, road infrastructure elements, and subway infrastructure elements; S2, constructing an associated network between city infrastructure nodes: constructing the same type of infrastructure nodes into the same layer in the GIS system, analyzing the association between infrastructure nodes from the geographical location and functional level, and constructing the infrastructure association network within and between layers based on the complex network model according to the association; S3, constructing a city rainstorm disaster chain network: obtaining historical rainstorm disaster typical cases in the target area, analyzing the occurrence mechanism, law and secondary disasters of rainstorm disasters, extracting disaster events and disaster chains based on text analysis method, and constructing a disaster chain network formed by rainstorm and secondary disasters by using a complex network model; S4, simulating secondary disaster risks caused by rainstorm disasters: constructing a calculation formula of secondary disaster risks caused by rainstorm disasters, measuring the local rainstorm risk level and substituting the formula to calculate the time sequence evolution of each secondary disaster risk level; S5, simulating infrastructure risks caused by rainstorm disaster chains: according to the local rainstorm risk level and the risk level of secondary disasters, constructing a calculation formula of the influence of different disasters on different infrastructure nodes, obtaining the time sequence evolution of the risk level of each infrastructure node, setting the threshold of the risk level of different infrastructure nodes to evaluate the working state of the infrastructure, that is, normal or failure, and obtaining the spatial distribution of infrastructure nodes that fail at multiple time points; S6, evaluating the cascading failure transmission of infrastructure node risks: analyzing the cascading failure transmission mechanism between different infrastructure nodes, simulating the failure diffusion process and influence range of failed infrastructure nodes in the corresponding period at different disaster scenarios, and forming the infrastructure node risk evaluation result; S7, realizing real-time risk display based on uniform identification code: importing the risk level and working state information of each infrastructure node calculated in S5 and S6 into the city infrastructure node information library, constructing a city infrastructure node rainstorm disaster chain risk early warning and prediction service platform, and realizing real-time risk intelligent prediction and early warning of any city infrastructure node before disaster through uniform identification code. 2.The urban infrastructure risk assessment method for storm disaster chain according to claim 1, wherein: The coding rules of the uniform identification code in S1 include: management subject + object category + individual code. 3.The urban infrastructure risk assessment method for storm disaster chain according to claim 1, wherein: The way of measuring the local rainstorm risk level in S4 includes: Obtaining rainfall data, city elevation, drainage pipe network and land use data published by weather forecast in the target area, dividing the target area into small grids, and simulating real-time water accumulation data in each small grid area as an index for measuring local rainstorm risk.

4. The urban infrastructure risk assessment method for storm disaster chain according to claim 1, characterized in that: The calculation formula of secondary disaster risks caused by rainstorm disasters in S4 includes: ； ； wherein, represents the node risk, represents the relationship strength factor, r represents the disaster intervention ability, equals 1 or 0, respectively indicating that there is or is not government intervention between two disaster nodes, represents the probability of the parent disaster event triggering the child disaster event, represents the delay coefficient, when judging the function, in order to avoid negative values of disaster loss degree, t represents the time of disaster evolution. 5.The urban infrastructure risk assessment method for storm disaster chain according to claim 1, wherein: The calculation formula of the influence of different disasters on different infrastructure nodes in S5 includes: ; wherein, X i is the damaged state of the infrastructure node i , B ki denotes the disaster-causing factor k , i denotes the connection strength between the disaster-causing factor H , k denotes the risk level of the disaster-causing factor k , t ki and β are the damping coefficient and the time delay coefficient, respectively, representing the time delay of the disaster-causing factor leading to the infrastructure damage and the delay of the loss spreading, τ i denotes the self-recovery capability of the infrastructure node ξ i (t) denotes the random term, θ denotes the threshold value, when the damaged state of the infrastructure node exceeds the threshold value, the infrastructure node is defined as its functional failure, and the infrastructure node working state is normal or failure, respectively, represented by F i equal to 1 or 0.

6. The urban infrastructure risk assessment method for storm disaster chain according to claim 5, characterized in that: The way of obtaining the spatial distribution of infrastructure nodes that fail at multiple time points in S5 includes: The risk level of the storm and its secondary disasters in each grid of the target area is substituted into a calculation formula of the influence of different disasters on different infrastructure nodes to obtain the influence degree of the storm disaster chain on the infrastructure nodes and the final working state of the infrastructure nodes, and a spatial distribution diagram of the failed infrastructure nodes at multiple time points is formed.

7. The urban infrastructure risk assessment method for storm disaster chain according to claim 1, characterized in that: The way of realizing the pre-disaster real-time risk intelligent prediction and early warning of the infrastructure nodes in any city through the unified identification code in S7 comprises: inputting the unified identification code into the storm disaster chain risk early warning and prediction service platform of the city infrastructure nodes to obtain the risk level and working state of the infrastructure nodes corresponding to the unified identification code; integrating the storm disaster chain risk early warning and prediction service platform of the city infrastructure nodes with the emergency response system and the public notification system of the city where the target area is located, triggering an alarm when the disaster risk reaches a target threshold, and sending the prediction and early warning notification to the residents and relevant departments in the target area.

Images