Dynamic disaster assessment method and system based on iron tower data

Abstract

The invention provides a dynamic disaster assessment method and system based on iron tower data, and relates to the technical field of GIS data processing, and the method comprises the steps: obtaining the iron tower state data of a target region, collecting multi-dimensional basic information, determining a plurality of critical regions based on the target region, carrying out the classification according to disaster levels, adding disaster level identifiers, and carrying out the clustering analysis. The method comprises the following steps: drawing up a disaster aggregation map, carrying out risk correction on the disaster aggregation map through iron tower state data and multi-dimensional basic information, determining a disaster influence level and identifying a key risk area, and carrying out resource allocation management in the disaster aggregation map according to the disaster influence level and the key risk area. The technical problems of low resource allocation decision accuracy and low rescue efficiency caused by inaccurate disaster assessment results in the prior art are solved. The technical effects of improving disaster assessment accuracy, ensuring scientificity and pertinence of resource allocation work and further improving rescue efficiency are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a dynamic disaster assessment method and system based on tower data. BACKGROUND

[0002] After a disaster occurs, accurately and quickly assessing the impact range and degree of the disaster is the key to formulating effective rescue and response strategies. However, traditional disaster assessment methods mainly rely on ground observations and manual reports, such as on-site investigation, disaster victim interviews, satellite remote sensing image interpretation, etc. Although these methods provide basic data for disaster assessment to some extent, they have obvious lag in information acquisition and are limited by on-site conditions such as road interruption and communication interruption, making it difficult to fully cover the disaster-affected areas. In addition, manual reports also have problems in accuracy and consistency, such as being disturbed by subjective factors or distorted information, often making it difficult to reflect the actual situation of the disaster. This lag and limitation may lead to inaccurate resource allocation decisions, delay the best rescue time, and increase disaster losses.

[0003] There is a technical problem in the prior art that the disaster assessment result is not accurate, leading to low accuracy of resource allocation decision and low rescue efficiency. SUMMARY

[0004] The present application provides a dynamic disaster assessment method and system based on tower data, which solves the technical problem in the prior art that the disaster assessment result is not accurate, leading to low accuracy of resource allocation decision and low rescue efficiency.

[0005] In view of the above problems, the present application provides a dynamic disaster assessment method and system based on tower data.

[0006] In a first aspect of the present application, a dynamic disaster assessment method based on tower data is provided, the method comprising: acquiring tower state data in a target area and collecting multi-dimensional basic information, the multi-dimensional basic information including population distribution information and traffic condition information; determining a plurality of critical areas based on the target area, classifying the plurality of critical areas according to disaster levels, and adding disaster level identifiers to the plurality of critical areas, the disaster level identifiers being associated with a disaster monitoring and early warning model through a mapping relationship, the disaster monitoring and early warning model being used to identify the association between tower abnormal data and disaster levels and to predict the possibility and impact range of disaster occurrence; performing clustering analysis based on the disaster level identifiers added to the plurality of critical areas, and formulating a disaster aggregation atlas, the disaster aggregation atlas being used to centrally manage different areas; performing risk correction on the disaster aggregation atlas based on the tower state data and the multi-dimensional basic information, determining disaster impact levels and identifying key risk areas; and performing resource allocation management according to the disaster impact levels and key risk areas in the disaster aggregation atlas.

[0007] In a second aspect of the present application, a dynamic disaster assessment system based on tower data is provided, which comprises: an information collection module, configured to acquire tower state data in a target area and collect multi-dimensional basic information, the multi-dimensional basic information including population distribution information, traffic condition information and satellite remote sensing image information; a grade classification module, configured to determine a plurality of critical areas based on the target area, classify the plurality of critical areas according to disaster grade, and add a disaster grade identifier to the plurality of critical areas, the disaster grade identifier being associated with a disaster monitoring and early warning model through a mapping relationship, the disaster monitoring and early warning model being configured to fuse the satellite remote sensing image information, identify the association between tower abnormal data and disaster grade, and predict the possibility and impact range of disaster occurrence; a clustering atlas formulation module, configured to perform clustering analysis and formulate a disaster aggregation atlas according to the disaster grade identifier added to the plurality of critical areas, the disaster aggregation atlas being configured to centrally manage different areas; a risk correction module, configured to correct the disaster aggregation atlas based on the tower state data and multi-dimensional basic information, determine a disaster impact grade and identify a key risk area; and a resource allocation module, configured to perform resource allocation management according to the disaster impact grade and key risk area in the disaster aggregation atlas.

[0008] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0009] The method provided by the embodiments of the present application acquires tower state data in a target area and collects multi-dimensional basic information, the multi-dimensional basic information including population distribution information and traffic condition information, determines a plurality of critical areas based on the target area, classifies the plurality of critical areas according to disaster grade, adds a disaster grade identifier to the plurality of critical areas, the disaster grade identifier being associated with a disaster monitoring and early warning model through a mapping relationship, the disaster monitoring and early warning model being configured to identify the association between tower abnormal data and disaster grade and predict the possibility and impact range of disaster occurrence, performs clustering analysis and formulates a disaster aggregation atlas according to the disaster grade identifier added to the plurality of critical areas, the disaster aggregation atlas being configured to centrally manage different areas, corrects the disaster aggregation atlas based on the tower state data and multi-dimensional basic information, determines a disaster impact grade and identifies a key risk area, and performs resource allocation management according to the disaster impact grade and key risk area in the disaster aggregation atlas. The technical effect of improving disaster assessment accuracy is achieved, the scientificity and pertinence of resource allocation work are ensured, and the rescue efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A flowchart illustrating a dynamic disaster assessment method based on tower data provided in this application;

[0012] Figure 2 This is a schematic diagram of the structure of a dynamic disaster assessment system based on tower data provided in this application.

[0013] Figure labeling: Information collection module 11, grade classification module 12, cluster map formulation module 13, risk correction module 14, resource allocation module 15. Detailed Implementation

[0014] This application provides a dynamic disaster assessment method and system based on tower data to address the technical problems in existing technologies where inaccurate disaster assessment results lead to low accuracy in resource allocation decisions and low rescue efficiency. It achieves the technical effect of improving the accuracy of disaster assessment, ensuring the scientific and targeted nature of resource allocation, and thus improving rescue efficiency.

[0015] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be understood that the present invention is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention. It should also be noted that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, not all of them.

[0016] Example 1, as Figure 1 As shown, this application provides a dynamic disaster assessment method based on tower data, the method comprising:

[0017] The system acquires data on the status of towers in the target area and collects multidimensional basic information, including population distribution information, traffic conditions information, and satellite remote sensing imagery information.

[0018] Specifically, the iron tower as an important infrastructure of the communication network, by analyzing the damage of the iron tower, the target disaster area can be accurately identified, and the target guidance for precise rescue is provided. Through the network management system or the Internet of Things device, such as a sensor, a remote monitoring system, etc., all the iron tower state data in the target area are obtained, the iron tower state data includes power state, communication quality index and equipment running state, etc., the power state reflects whether the iron tower is in the power supply state, whether the power supply is stable, the communication quality index reflects whether the iron tower can normally transmit data, such as signal strength, network coverage, call drop rate, time delay, etc. The equipment running state reflects whether the base station equipment, antenna equipment, etc. work normally. At the same time of obtaining the iron tower state data, multi-dimensional basic information of different sources is collected, the multi-dimensional basic information includes population distribution information, traffic condition information and satellite remote sensing image information, wherein the population distribution information is obtained through public database, geographic information system data published by the statistical bureau or population census data, etc., which can reflect the disaster population density, age structure, etc. in the target area. The traffic condition information is obtained through the traffic monitoring system, navigation platform or vehicle-mounted GPS device, which is used to reflect the disaster traffic condition, damage condition and road network structure of the target area. The satellite remote sensing image information is obtained through the satellite platform or unmanned aerial vehicle aerial photography, which can provide large-scale, objective ground coverage change information, such as terrain damage, water range, vegetation damage, ground displacement, building collapse condition and secondary disaster (such as landslide, barrier lake) risk area, etc., which is a macro supplement to the iron tower state data and ground information. By obtaining the iron tower state data of the target area and collecting the multi-dimensional basic information such as population distribution, traffic condition and satellite remote sensing image, it is not only helpful to more comprehensively and stereoscopically understand the actual situation of the target area after the disaster, especially the ground physical damage and potential secondary disaster risk, but also provides accurate and reliable data support for disaster assessment and reasonable allocation of resources and rescue path planning.

[0019] Based on the target area, a plurality of critical areas are determined, classified according to the disaster level, and a disaster level identifier is added to the plurality of critical areas, the disaster level identifier is associated with a disaster monitoring and early warning model through a mapping relationship, the disaster monitoring and early warning model is used to fuse the satellite remote sensing image information, identify the association relationship between the iron tower abnormal data and the disaster level and predict the possibility and influence range of disaster occurrence.

[0020] Specifically, according to the target area, a plurality of critical areas are determined using a geographic information system (GIS) and satellite remote sensing images. The target area refers to the area where the disaster occurs, and the critical area refers to the surrounding or associated area that is affected by the disaster due to factors such as geographic location, topography, climate conditions, etc. The target area is divided into a plurality of critical areas by superimposing satellite remote sensing images (such as post-disaster immediate images, change detection maps) through a geographic information system (GIS). The plurality of critical areas can be administrative areas of a city, natural divisions (such as river basins, mountain units), population distribution hotspots, or small blocks divided according to physical damage boundaries identified by remote sensing images (such as landslide range, flood inundation line, fire spread edge). After determining the plurality of critical areas, the plurality of critical areas are classified according to the actual impact of the disaster. The disaster-affected level classification is based on a combination of factors such as the degree of physical damage to the ground surface (such as building collapse rate, infrastructure damage range, change in ground cover), the range and intensity of the disaster, the degree of communication and infrastructure damage (based on tower status data evaluation), the degree of population affected, and the traffic disruption (based on multi-dimensional basic information evaluation). For example, according to these indicators, the critical areas are automatically classified into lightly affected areas, severely affected areas, and extremely affected areas, and a disaster-affected level identifier is added to each critical area according to these classifications. The disaster-affected level identifier can represent the severity of the disaster through a number, color, or label. For example: the lightly affected area is identified as green, indicating that the disaster has a small impact and the physical damage to the ground surface is slight, the communication is basically unaffected, and the traffic is smooth. The severely affected area is identified as yellow, indicating that the disaster has a moderate impact, some buildings / infrastructure are damaged (visible in remote sensing images), some facilities are damaged, and some traffic is blocked. The extremely affected area is identified as red, indicating that the disaster is severe, the physical damage to the ground surface is extensive and severe (clearly reflected in remote sensing images), there are many casualties, infrastructure is destroyed, and traffic is completely blocked. By determining the critical areas and their disaster-affected levels, clear priority guidelines are provided, providing a reliable basis for resource allocation management. For example, the extremely affected area may require more supplies, medical support, and rescue personnel, while the lightly affected area may only require basic supplies.

[0021] To achieve accurate identification and dynamic assessment of disaster risk levels in different critical regions, a disaster monitoring and early warning model is constructed based on historical disaster data and expert knowledge. The disaster monitoring and early warning model takes the geographical environment characteristics of the target region, the infrastructure status, and the features extracted from satellite remote sensing images (such as spectral features, texture features, change detection indicators, and secondary disaster risk maps) as core input variables, and combines other multi-dimensional basic information to train and establish the corresponding relationship between disaster occurrence and disaster level. For example, first, collect the satellite remote sensing image features (such as NDVI change, land surface change, water body index anomaly) of multiple critical regions in the target region under different disaster levels in typical disaster events such as earthquakes, typhoons, and floods, as well as geographical data anomalies, communication interruptions, and other state information as training samples for the model. At the same time, introduce the field experts' grading judgment of disaster impact (usually also referring to the remote sensing image interpretation results) as knowledge constraints or model correction basis. In the modeling process, classification or regression algorithms such as decision tree, random forest, and neural network can be used to enable the disaster monitoring and early warning model to have the ability to assess the current state of multiple critical regions (fusing remote sensing, tower, and other surface data). The disaster possibility value and predicted impact range output by the disaster monitoring and early warning model will be mapped to the predefined disaster level identifiers through a set mapping mechanism, such as green for light disaster, yellow for moderate disaster, and red for severe disaster. By identifying and determining multiple critical regions in the target region based on the data of the target region and the disaster monitoring and early warning model, not only can static disaster level identifiers be added to these regions, but also can help emergency management personnel quickly and objectively (especially with the wide coverage and intuitiveness of remote sensing) assess the impact of disasters after the disaster, determine the priority of disaster areas, and effectively plan subsequent resource allocation, rescue tasks, and risk management, greatly improving the decision-making efficiency of emergency response.

[0022] According to the disaster level identifiers added to the multiple critical regions, cluster analysis is performed to develop a disaster aggregation map, which is used for centralized management of different regions.

[0023] Specifically, a plurality of critical regions are determined according to the disaster impact of the target region, and a disaster level identifier is added to the plurality of critical regions, which reflects the disaster degree of different regions after the disaster (the identifier has integrated the ground physical damage information interpreted from satellite remote sensing images). After completing the addition of the disaster level identifier, clustering analysis methods such as K-means clustering, hierarchical clustering, etc. are used for clustering analysis, and regions with similar disaster characteristics (including the comprehensive severity represented by the disaster level identifier and the similarity of the ground damage pattern revealed by satellite remote sensing images) are grouped. For example, severely damaged areas with large areas of severe physical damage (such as dense building collapse, severe landslide) are clustered into a group. The results of clustering analysis form different aggregation regions, each of which represents a similar disaster situation. On this basis, according to the clustering analysis results, a disaster aggregation atlas is prepared, which is usually based on a GIS platform, superimposes satellite remote sensing image base maps or key feature layers (such as damage range maps, secondary disaster risk maps), and visualizes the aggregation results of each disaster region. The disaster aggregation atlas not only clearly shows the disaster situation of different regions after the disaster, but also can intuitively distinguish between light, moderate, and severe disaster areas according to different colors, symbols, or markers, and can intuitively present the correspondence between aggregation regions and actual physical damage regions on the remote sensing image background. By preparing the disaster aggregation atlas, it is not only helpful to concentrate management of different regions, but also can intuitively show the overall situation of the disaster situation by combining the macro view provided by remote sensing, and provide intuitive and comprehensive data information. Through clustering analysis and preparation of the disaster aggregation atlas, post-disaster emergency management can more efficiently concentrate on handling disaster conditions in different regions. Different disaster level regions can be allocated targeted rescue resources, thereby improving the efficiency of resource use and post-disaster response speed.

[0024] The risk of the disaster aggregation atlas is corrected by the iron tower state data and the multi-dimensional basic information, the disaster impact level is determined, and the key risk region is identified.

[0025] Specifically, after obtaining the disaster aggregation map, the tower state data, satellite remote sensing image information (providing real-time / near real-time ground physical damage and secondary disaster risk dynamics) and multi-dimensional basic information are comprehensively analyzed, and the disaster aggregation map data is risk corrected. Risk correction refers to combining real-time tower state data (reflecting the functional state of communication infrastructure), satellite remote sensing image information (reflecting changes in ground physical state and potential risks), and multi-dimensional basic information (such as population, traffic dynamics), dynamically updating and adjusting the disaster impact degree of each region in the disaster aggregation map data, and re-evaluating and marking the disaster degree of each region. Through risk correction, the disaster impact level of different critical regions is more accurately determined, and it is identified which regions are key risk areas, which are places where the disaster impact is the most serious and resources need to be allocated first, especially those regions that show serious physical damage on remote sensing images, have high secondary disaster risk, and tower state data shows that the communication function is severely damaged. Through risk correction of the disaster aggregation map, disaster assessment is changed from static analysis to dynamic adjustment, ensuring that the assessment results are consistent with the actual disaster situation. At the same time, by identifying key risk areas, resource allocation and rescue plans are more accurate, and the accuracy and efficiency of disaster emergency response are improved.

[0026] In the disaster aggregation map, according to the disaster impact level and the key risk area, resource allocation management is carried out.

[0027] Specifically, on the basis of the disaster aggregation map obtained after the risk correction of satellite remote sensing image information, according to the evaluation results of disaster influence level and key risk area (the determination of key risk area has included the serious physical damage and high secondary disaster risk information revealed by satellite remote sensing), the type and quantity of resource demand of each area are determined, including basic materials, technical support and human resources, etc. Basic materials, such as food, drinking water, medical supplies, etc. Technical support, such as communication equipment, emergency communication base station, is used to restore the communication network of the disaster area. Human resources, such as rescue personnel, medical team. After completing the resource demand analysis, according to the priority principle, the balanced principle and the dynamic adjustment principle, the resource allocation rules are formulated, the priority principle means that the key risk area (especially the area with serious physical damage, high secondary disaster risk and communication interruption revealed by remote sensing) is given priority to obtain resources, the balanced principle means that according to the total amount of resources and regional demand, it is avoided that the over-concentration of resources leads to the lack of support for other areas, and the dynamic adjustment principle means that the real-time monitoring of regional disaster changes (including the development of secondary disasters, new damage areas and the recovery of iron tower / traffic monitored by remote sensing) is carried out, and the resource allocation plan is adjusted flexibly. Further, according to the resource allocation rules, the resource allocation management is carried out, the rescue forces and resources are reasonably allocated, the scientificity and pertinence of rescue work are ensured, and the disaster loss is reduced. Under the support of the disaster aggregation map (which has integrated remote sensing base map and risk information), based on the disaster influence level and the key risk area integrated with remote sensing criteria, the resource allocation management is carried out to realize the precise allocation of rescue resources, improve the rescue efficiency and reduce the disaster loss.

[0028] Further, according to the disaster influence level and the key risk area, resource allocation management is carried out, and the method comprises: acquiring real-time traffic condition information; identifying road congestion sections based on the real-time traffic condition information; according to the road congestion sections, pushing to a traffic navigation platform through an API interface, and the traffic navigation platform uses a message queue and a service bus to plan a resource allocation path.

[0029] Specifically, real-time road traffic information of key risk areas is collected through navigation platforms and traffic monitoring systems, as well as road damage analysis of satellite remote sensing images (to assist in identifying institutional blockades), to obtain real-time traffic condition information. By analyzing real-time traffic condition information, the current road congestion section is accurately identified, providing optimization suggestions for subsequent path planning, ensuring that rescue resources are not delayed due to traffic problems. According to the road congestion section, the road congestion section information is pushed to the traffic navigation platform through the API interface, connecting real-time traffic conditions with the resource allocation system to achieve real-time data flow and updates. The API interface ensures the efficiency and real-time nature of data exchange. After receiving the road congestion section, the traffic navigation platform uses message queues and service buses to combine congestion section information with route planning for resource allocation path planning. Message queues are used for asynchronous communication between platforms, enabling message transmission between different systems or platforms to ensure smooth collaboration between various links. The traffic navigation platform sends the results of path planning to relevant platforms, such as emergency command centers and rescue teams, via message queues, which can avoid message congestion caused by a large number of requests while ensuring smooth information transmission. The service bus is an integrated platform that coordinates and manages message exchange between different platforms. During resource allocation path planning, the service bus can help integrate information from different platforms, such as traffic conditions and resource requirements (prioritized by disaster level and key risk areas, including remote sensing information), to form a comprehensive dispatch plan for decision-makers and operators. Using message queues and service buses ensures efficient data flow and high-concurrency processing capabilities throughout the dispatch process, improving the real-time nature and accuracy of resource allocation. By obtaining real-time traffic condition information (supplemented by remote sensing road condition analysis), identifying road congestion sections, and pushing this information to the traffic navigation platform through API interfaces, combined with message queue and service bus technologies for path planning, resource transportation paths can be optimized, improving the real-time nature and decision support capabilities of resource allocation.

[0030] Further, the risk of the disaster-affected cluster map is corrected based on the tower state data and multi-dimensional basic information. The method further includes: positioning the traffic hotspot coordinates based on the multi-dimensional basic information; analyzing the communication quality indicators in the tower state data, combining the traffic hotspot coordinates to evaluate the network coverage quality and signal strength; determining the potential communication interruption area through the network coverage quality and signal strength, and performing risk level verification.

[0031] Specifically, based on the population distribution information, traffic condition information and multi-dimensional basic information of the temporary population gathering area (such as the shelter, traffic hub) displayed by the satellite remote sensing image, the traffic hotspot coordinates are located, the traffic hotspot refers to the area with intensive data traffic or user activity in a specific time, the traffic hotspot coordinate positioning implementation manner includes: through the historical traffic data, sensor data, GPS data and remote sensing heat map analysis, identifying which areas have larger communication load before or during the disaster, especially the key areas such as densely populated areas, traffic arteries, high-value rescue targets (such as hospitals, schools) or temporary gathering points identified by remote sensing, etc., and the traffic hotspot coordinate positioning is carried out. With the help of the traffic monitoring system or third-party API, such as communication operator data, Wi-Fi access point traffic information, etc., the communication traffic data of each area is obtained to locate the traffic hotspot coordinates. Through locating the traffic hotspot coordinates, not only the areas with concentrated demand are reflected, but also the basis for subsequent network quality evaluation and communication recovery is provided. The communication quality indicators in the tower state data are analyzed, the communication quality indicators directly affect the communication quality and service stability in the area, such as signal strength, network coverage range, call drop rate, time delay, etc. Through the analysis of the communication quality indicators of the tower, the stability of the network and the reliability of the signal can be evaluated. The located traffic hotspot coordinates are combined with the analyzed tower state data, through spatial analysis and clustering algorithm, the network coverage quality and signal strength are evaluated, the traffic hotspot represents higher communication demand, combined with the coordinates of the traffic hotspot and the communication quality data of the tower, the network service capability of multiple critical areas is evaluated. For example, if the signal quality of the tower in a certain hotspot area is poor or there is a base station failure, the communication quality of the area may be affected, by aggregating the traffic hotspot and tower data, using the weighted average method to evaluate the comprehensive coverage quality of the network in the traffic hotspot area, further determine whether there are areas with weak signal and unstable network. Key, compare the network coverage quality and signal strength, and superimpose the secondary disaster risk areas (such as high-risk areas of landslides, flood inundation areas, fire spread reverse, barrier lake threat areas, etc.) identified by satellite remote sensing image information, determine the potential communication interruption area, if the signal strength of a certain traffic hotspot area is weak and the communication demand is extremely high, especially when it is located or adjacent to the secondary disaster high-risk area identified by remote sensing, then this area has a high risk of communication interruption. Similarly, according to the state data of the tower, if the operation state of some towers is not good and the location is in the direct threat path of the secondary disaster (such as under the landslide body, flood prone area), there is also a risk of network interruption in its coverage area.Finally, according to the identified potential communication interruption area (combined with communication quality, demand heat and secondary disaster risk), the risk level of the disaster-affected cluster map is checked, that is, the risk level of each area is adjusted according to the communication quality, signal strength, traffic demand and secondary disaster threat level and other factors, which effectively improves the stability of post-disaster communication and the accuracy of emergency response, and avoids secondary communication interruption caused by secondary disasters in advance.

[0032] Further, the network coverage quality and signal strength are used to determine the potential communication interruption area and perform risk level checking. The method comprises: establishing a disaster history database and setting risk level evaluation standards; checking the risk level of the potential communication interruption area through the risk level evaluation standards, and establishing a linkage deployment mechanism, which uses an API interface to share resource demand information of each risk level in real time.

[0033] Specifically, a disaster history database is established, which records past disaster events and their impacts in the target area, including various types of disasters such as earthquakes, floods, typhoons, etc. The impact on the communication network, the geographical location of the communication interruption area, the recovery time, and the corresponding satellite remote sensing image features (such as the association of specific spectral / texture patterns with communication recovery time) are included. According to multiple factors, risk level assessment criteria are set, including the duration of communication interruption, the population density and importance of the interruption area, network recovery capability, and satellite remote sensing image interpretation of secondary disaster risk level (high / medium / low) and its potential damage to communication facilities. For example, if the communication interruption lasts for a certain period of time, the risk level is high. If the interruption area is located in a densely populated area or near critical infrastructure or in a high-risk core area of secondary disasters, the risk level is significantly higher. If the network recovery speed of the area is slow, there is no backup communication scheme, or the secondary disaster risk makes the recovery conditions worse (such as continuous road interruption, unstable geology), the risk is greater. Then, using the risk level assessment criteria, according to the real-time network coverage quality, signal strength data, and the latest satellite remote sensing secondary disaster risk assessment results, combined with the reference of the disaster history database, the identified potential communication interruption areas are checked for risk level to ensure that each area is accurately classified into the corresponding risk level. For example, if the signal strength and network coverage quality of a certain area are poor, and the remote sensing image shows that it is within the threat range of a high-activity landslide, even if the historical disaster data shows that the recovery of the area is acceptable, the risk level of the area is rated as extremely high. Conversely, if the communication quality of an area is temporarily poor, and the remote sensing shows that the surrounding secondary disaster risk is extremely low and the recovery path is smooth, the risk level can be appropriately lowered. After completing the risk assessment, a linkage deployment mechanism is established, which refers to the real-time sharing of resource demand information to quickly respond to the needs of each risk level area. The linkage deployment mechanism uses API interface to share resource demand information of each risk level in real time, such as emergency communication equipment, power supply, rescue personnel, and specific protection or engineering resources for secondary disasters (daily landslide control equipment, flood control materials), etc. to ensure that rescue resources and recovery equipment can quickly and accurately reach high-risk areas, and priority is given to those areas that face both communication interruption and high secondary disaster risk for emergency communication recovery and safety. The risk level verification and linkage deployment mechanism provides real-time and accurate decision-making basis for emergency management personnel, optimizes resource allocation, and improves the overall emergency response capability, ensuring smooth post-disaster recovery and effectively reducing secondary losses caused by secondary disasters.

[0034] Further, the linkage allocation mechanism uses an API interface to share resource demand information corresponding to each risk level in real time, and the method comprises: establishing a communication network using an API interface; tracking resource transfer status using an Internet of Things device; and formulating a closed-loop feedback loop based on the resource transfer status, the closed-loop feedback loop being used to optimize resource allocation paths.

[0035] In particular, a communication network is established using API interfaces for passing and sharing information between different systems or platforms. Through the API interfaces, the linkage and deployment mechanism can pass the resource demand information of each region, such as medical supplies, power supply, communication recovery equipment, etc. (especially special protection or engineering resources for high secondary disaster risk areas), and the risk level corresponding to these demands (including secondary disaster risk level information determined by satellite remote sensing) to the resource deployment center, material suppliers, transportation scheduling platform and on-site rescue / engineering nodes in real time after the disaster. By establishing a unified communication network, all related systems or platforms can quickly receive changes in resource demand and its priority (based on the key risk level of fused remote sensing information), and respond according to real-time data. Each transported resource and material, especially the material transported to high secondary disaster risk areas, is equipped with Internet of Things devices such as RFID tags, GPS devices, and environmental sensors (such as temperature and humidity, vibration detection), which can provide real-time information such as transportation path, current location, estimated time of arrival, and transportation environment status (especially important for perishable or critical materials). Using Internet of Things devices can continuously monitor the transportation status throughout the transportation process. For example, through the monitoring of rescue materials by Internet of Things devices, the current location, transportation status, and whether the materials enter or approach the secondary disaster risk area updated by remote sensing monitoring can be known in real time. Based on the resource transportation status and combined with the dynamic changes of secondary disaster risks (such as new landslide warning, flood range expansion) and traffic network physical damage information (such as bridge collapse, road burial) provided by satellite remote sensing images, a closed-loop feedback loop is developed. The closed-loop feedback loop is a self-regulating mechanism that dynamically adjusts the scheduling path, transportation priority, and safety strategy of resources by continuously collecting and feeding back real-time data of resource transportation status and external risk environment changes (mainly from remote sensing monitoring). Based on the closed-loop feedback loop and real-time tracking data and remote sensing risk intelligence, the deployment path of resources is dynamically optimized. For example, when a delay or other unforeseen obstacles (such as traffic congestion, local weather deterioration) are found in the transportation of materials, the feedback mechanism will immediately link the deployment mechanism, use path optimization algorithms such as the shortest path algorithm or dynamic path adjustment based on congestion, recalculate and adjust the transportation route of resources, and adjust the transportation path in real time to ensure that rescue materials reach the places that need them most in time. When remote sensing monitoring shows that the transportation destination or its route has a sudden high secondary disaster risk (such as dam-break warning of barrier lake or approaching of forest fire spread), the feedback mechanism will immediately trigger path re-planning and safety warning, and preferentially guide the transportation team to bypass the high-risk area or go to a safe gathering point. When remote sensing identifies new road structural damage (such as bridge collapse), the feedback mechanism will immediately update the road network model and re-plan the path to avoid vehicles entering dangerous sections.Through the combination of API interface, Internet of Things device, satellite remote sensing risk monitoring and closed-loop feedback loop, more intelligent and safer resource allocation decisions are realized, the efficiency and safety of resource allocation are optimized, it is ensured that resources (especially critical emergency resources) can efficiently and accurately reach the target area in the entire allocation process, resource waste and delay are avoided, and the safety of transportation personnel and materials is maximized, and the impact of secondary disasters on rescue operations is reduced.

[0036] Further, the method further comprises: determining a time series record using an NLP technology; constructing a loss assessment model based on the time series record; obtaining a disaster loss prediction result according to the loss assessment model, and updating resource demand information of each risk level.

[0037] Specifically, the time series records are dynamic information describing the disaster occurrence process, such as the temporal variation of disaster impact, the expansion of regional scope, the progress of communication recovery, etc. These dynamic information exist in unstructured texts such as disaster reports, news, social media, monitoring logs, etc. Unstructured text data related to disasters in the target area are collected, including social media updates, disaster reports, news articles, etc. Using entity recognition and event extraction techniques in NLP, key information such as time, location, event type, etc. are extracted from the text. For example, from news reports, extract the event of communication disruption in a certain area, with a duration of 3 hours. Then, according to the extracted information, generate structured data with time as the axis, and determine the time series records. At the same time, obtain satellite remote sensing images before and after the disaster, and perform change detection analysis to generate quantitative ground physical change time series information, such as building damage area change, water body range expansion, vegetation damage degree evolution, ground displacement accumulation, etc. at different time phases. Fuse the time series records extracted from the text with the time series information generated by satellite remote sensing image change detection to form a more comprehensive and objective structured time series dataset. The time series records include disaster duration, impact range (spatial dimension accurately defined by remote sensing), regional population density, remote sensing quantitative assessment of infrastructure damage (such as building collapse rate, road damage length), etc. After obtaining the structured fusion time series records, use the time series records to build a loss assessment model, and the goal of the loss assessment model is to predict the loss caused by the disaster based on the time series data. Optionally, extract key factors affecting loss from the time series records, such as disaster duration, impact range (remote sensing definition), regional population density, remote sensing assessment of infrastructure damage, etc. Select appropriate model types for loss assessment, such as using linear regression, gradient boosting tree or convolutional neural network (CNN) to process remote sensing change feature maps, combine historical disaster data and current time series records, and train the linear regression model to accurately predict the loss, obtaining the loss assessment model. Use the time series records as input data, pass through the trained loss assessment model, and obtain the loss prediction results of the disaster at each region and time point, including direct economic loss, indirect loss and recovery time prediction, etc. Direct economic loss, such as infrastructure repair cost, communication recovery cost, etc. Indirect loss, such as economic loss caused by traffic disruption, impact of communication disruption on production and life, etc. Recovery time prediction, such as estimated time for network recovery, required time for crowd rescue, etc. According to the loss prediction results, dynamically update the resource demand information of each risk level region, and the updated resource demand information will be pushed to the resource allocation system in real time through the linkage allocation mechanism, guiding subsequent resource allocation and path planning.By combining the time series records extracted by NLP technology with the physical change time series generated by satellite remote sensing change detection technology, and combining the prediction ability of the loss evaluation model, the loss caused by disasters can be more comprehensively and accurately evaluated, and the resource demand of the risk level area can be dynamically updated, providing reliable data for scientific decision-making, improving the efficiency of post-disaster recovery, and ensuring that emergency response work can quickly and accurately meet actual needs.

[0038] Further, the disaster loss prediction result is obtained, and resource demand information of each risk level is updated, and the method comprises: introducing an edge computing node in the communication network, establishing a distributed computing framework; based on the loss evaluation model, using the distributed computing framework, performing parallel calculation on the target area and a plurality of critical areas.

[0039] Specifically, edge computing nodes are computing devices close to data sources or terminal devices, capable of processing and storing data close to the scene, thereby reducing data transmission latency and improving computing efficiency. The introduction of edge computing nodes in the communication network establishes a distributed computing framework, and through the introduction of edge computing nodes, data such as communication quality and signal strength collected from towers, Internet of Things devices, and sensors can be directly processed on the edge nodes. The preliminary processing and feature extraction of original satellite remote sensing images (such as change detection calculation and target damage identification) are performed on the edge nodes close to the data acquisition points (such as ground receiving stations or unmanned aerial vehicle relay points), and only the processed feature data, change detection results, or lightweight analysis products (rather than massive original image data) are transmitted to the central node or other edge nodes, significantly reducing network bandwidth demand and transmission delay. The distributed computing framework is a distributed system architecture that can split large-scale computing tasks into multiple sub-tasks and distribute them to different nodes for parallel processing. The distributed computing framework combines edge nodes with central computing nodes to form a multi-level computing network, and according to the computing capacity, real-time load, and data locality (such as processing locally acquired remote sensing data) of each node, the computing tasks are reasonably allocated. For example, tasks in high-risk areas or newly acquired remote sensing image coverage areas are preferentially allocated to nodes with stronger computing capacity. At the same time, the distributed computing framework has a built-in fault tolerance mechanism, and when a node fails due to disasters or other reasons, its computing tasks will be dynamically allocated to other available nodes, ensuring the continuity of the computing process. The computing tasks of multiple critical areas are allocated to the corresponding edge nodes or central nodes, and the loss assessment model is run on each node to predict the disaster loss of each area using locally or nearby acquired fusion time series data (including text event information and locally processed remote sensing change features). After the calculation is completed, the distributed computing framework aggregates the results of each node to the central node, dynamically updates the resource demand information of each risk level area, and prioritizes the allocation of communication recovery equipment, rescue teams, medical supplies, and urgently needed resources such as engineering equipment with severe damage shown by remote sensing in high-risk areas. In medium-risk areas, resources such as logistics transportation and infrastructure repair equipment are increased in response to the dynamic changes in loss prediction. Low-risk areas maintain basic support while focusing on potential risks that may escalate due to the spread of the disaster (remote sensing can monitor the spread trend). The update of resource demand information is shared in real time through an API interface to the linkage deployment mechanism, ensuring that resource allocation can respond to changes in disaster conditions in a timely manner. By introducing edge computing nodes into the communication network and establishing a distributed computing framework, combined with the loss assessment model and the near-source processing capability of remote sensing data, parallel computing of target areas and critical areas is realized, significantly improving the speed and accuracy of disaster loss prediction, especially meeting the demand for rapid analysis of massive remote sensing data to support real-time loss assessment, thereby ensuring the efficiency and accuracy of post-disaster resource allocation.

[0040] In summary, the dynamic disaster assessment method based on tower data provided by the embodiments of the application has the following technical effects:

[0041] 1. By combining tower multi-source data and advanced analysis algorithms, the disaster area analysis technology can more accurately identify the disaster loss of each region, providing a scientific basis for precise rescue;

[0042] 2. Real-time collection and analysis of tower data can quickly lock the severely affected areas, shorten the emergency response time, and improve the efficiency of rescue;

[0043] 3. According to the disaster area analysis results, the rescue forces and resources are reasonably allocated to ensure the scientificity and pertinence of the rescue work and reduce the disaster loss.

[0044] 4. Based on historical disaster data and expert knowledge, a disaster monitoring and early warning model is constructed, which can automatically identify the correlation between tower abnormal data and disasters, and predict the possibility and impact range of disasters.

[0045] Embodiment two, based on the same inventive concept as the dynamic disaster assessment method based on tower data in the preceding embodiments, as shown in Figure 2 The application provides a dynamic disaster assessment system based on tower data, which comprises:

[0046] An information collection module 11 is configured to acquire tower state data in a target area and collect multi-dimensional basic information, including population distribution information, traffic condition information, and satellite remote sensing image information; a grade classification module 12 is configured to determine a plurality of critical areas based on the target area, classify them according to disaster grades, and add disaster grade identifiers to the plurality of critical areas, wherein the disaster grade identifiers are associated with a disaster monitoring and early warning model through a mapping relationship, and the disaster monitoring and early warning model is used to fuse the satellite remote sensing image information, identify the correlation between tower abnormal data and disaster grades, and predict the possibility and impact range of disasters; a clustering atlas formulation module 13 is configured to perform clustering analysis and formulate a disaster aggregation atlas according to the disaster grade identifiers added to the plurality of critical areas, wherein the disaster aggregation atlas is used to centrally manage different areas; a risk correction module 14 is configured to correct the disaster aggregation atlas based on the tower state data and multi-dimensional basic information, determine disaster impact grades, and identify key risk areas; and a resource allocation module 15 is configured to manage resource allocation in the disaster aggregation atlas according to the disaster impact grades and key risk areas.

[0047] Further, the resource allocation module 15 is configured to perform the following steps: obtaining real-time traffic condition information; identifying a road congestion section based on the real-time traffic condition information; and pushing the road congestion section to a traffic navigation platform through an API interface, wherein the traffic navigation platform uses a message queue and a service bus to perform resource allocation path planning.

[0048] Further, the risk correction module 14 is configured to perform the following steps: locating a traffic hotspot coordinate based on the multi-dimensional basic information; analyzing a communication quality indicator in the tower state data, and evaluating network coverage quality and signal strength in combination with the traffic hotspot coordinate; determining a potential communication interruption area based on the network coverage quality and signal strength, and in combination with a secondary disaster risk area identified by the satellite remote sensing image information, and performing risk level verification.

[0049] Further, the risk correction module 14 is further configured to perform the following steps: establishing a disaster history database and setting a risk level evaluation standard; performing risk level verification on the potential communication interruption area based on the risk level evaluation standard, and establishing a linkage allocation mechanism, wherein the linkage allocation mechanism uses an API interface to share resource demand information of each risk level in real time.

[0050] Further, the risk correction module 14 is further configured to perform the following steps: establishing a communication network by using an API interface; tracking resource transfer status by using an Internet of Things device; and formulating a closed-loop feedback loop based on the resource transfer status, wherein the closed-loop feedback loop is used to optimize a resource allocation path.

[0051] Further, the risk correction module 14 is further configured to perform the following steps: determining a time series record by using an NLP technology; constructing a loss evaluation model based on the time series record and satellite remote sensing image change detection information; obtaining a disaster loss prediction result according to the loss evaluation model, and updating resource demand information of each risk level.

[0052] Further, the risk correction module 14 is further configured to perform the following steps: introducing an edge computing node into the communication network to establish a distributed computing framework; and performing parallel computing on the target area and a plurality of critical areas by using the distributed computing framework based on the loss evaluation model.

[0053] The above only describes the preferred embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0054] The specification and drawings are, of course, subject to various interpretations and should not be viewed in any limiting sense. It will be understood that various modifications and changes can be made to the application disclosed without departing from the scope thereof. Accordingly, you are to understand that there is no intention, either express or implied, that any of the described embodiments of the application is more efficient or effective than any of the other possible embodiments. It is therefore intended to cover in the appended claims all such changes and modifications that fall within the scope of the application.

Claims

1. A dynamic disaster assessment method based on tower data, characterized in that, The method includes: Acquire the status data of the towers in the target area and collect multi-dimensional basic information, including population distribution information, traffic condition information and satellite remote sensing image information; Based on the target area, multiple critical areas are identified and classified according to the disaster level. Disaster level identifiers are added to the multiple critical areas. The disaster level identifiers are associated with the disaster monitoring and early warning model through a mapping relationship. The disaster monitoring and early warning model is used to fuse the satellite remote sensing image information, identify the correlation between tower anomaly data and disaster level, and predict the possibility and impact range of disasters. Based on the multiple critical areas, disaster level identifiers are added, cluster analysis is performed, and a disaster cluster map is proposed. The disaster cluster map is used for centralized management of different areas. By using the tower status data and multi-dimensional basic information, the disaster cluster map is risk-corrected to determine the disaster impact level and identify key risk areas; In the disaster cluster map, resource allocation and management are carried out according to the disaster impact level and key risk areas.

2. The dynamic disaster assessment method based on tower data as described in claim 1, characterized in that, Based on the disaster impact level and key risk areas, resource allocation and management are carried out, the method including: Get real-time traffic information; Based on the real-time traffic information, identify congested road sections; Based on the road congestion section, the data is pushed to the traffic navigation platform via API interface. The traffic navigation platform uses message queues and service buses for resource allocation and route planning.

3. The dynamic disaster assessment method based on tower data as described in claim 2, characterized in that, The method further includes performing risk correction on the disaster cluster map using the tower status data and multi-dimensional basic information. Based on the aforementioned multidimensional basic information, the coordinates of traffic hotspots are located; Analyze the communication quality indicators in the tower status data, and combine them with the traffic hotspot coordinates to evaluate network coverage quality and signal strength; By analyzing the network coverage quality and signal strength, and combining this with the secondary disaster risk areas identified by the satellite remote sensing imagery, potential communication disruption areas are determined, and risk levels are verified.

4. The dynamic disaster assessment method based on tower data as described in claim 3, characterized in that, By analyzing network coverage quality and signal strength, and combining this with satellite remote sensing imagery to identify secondary disaster risk areas, potential communication outage areas are determined, and risk levels are verified. The method includes: Establish a historical disaster database and set risk level assessment standards; Based on the aforementioned risk level assessment criteria, the risk level of potential communication disruption areas is verified, and a coordinated allocation mechanism is established. This coordinated allocation mechanism uses API interfaces to share resource demand information for each risk level in real time.

5. The dynamic disaster assessment method based on tower data as described in claim 4, characterized in that, The coordinated allocation mechanism uses an API interface to share resource demand information corresponding to each risk level in real time, and the method includes: Establish a communication network using API interfaces; Use IoT devices to track the status of resource transfer; Based on the resource transfer status, a closed-loop feedback loop is proposed, which is used to optimize the resource allocation path.

6. The dynamic disaster assessment method based on tower data as described in claim 5, characterized in that, The method includes: Use NLP techniques to identify time series records; Based on the aforementioned time-series records and satellite remote sensing image change detection information, a loss assessment model is constructed. Based on the loss assessment model, disaster loss prediction results are obtained, and resource demand information for each risk level is updated.

7. The dynamic disaster assessment method based on tower data as described in claim 6, characterized in that, Based on the loss assessment model, disaster loss prediction results are obtained, and resource demand information for each risk level is updated. The method includes: Edge computing nodes are introduced into the communication network to establish a distributed computing framework; Based on the loss assessment model, a distributed computing framework is used to perform parallel computation on the target region and multiple critical regions.

8. A dynamic disaster assessment system based on tower data, characterized in that, The dynamic disaster assessment system based on tower data is used to implement the steps of the method according to any one of claims 1 to 7, including: The information acquisition module is used to acquire the status data of the towers in the target area and to collect multi-dimensional basic information, including population distribution information, traffic condition information and satellite remote sensing image information. The classification module is used to determine multiple critical areas based on the target area, classify them according to the disaster level, add disaster level identifiers to the multiple critical areas, and associate the disaster level identifiers with the disaster monitoring and early warning model through a mapping relationship. The disaster monitoring and early warning model is used to fuse the satellite remote sensing image information, identify the correlation between tower anomaly data and disaster level, and predict the possibility and impact range of disasters. The clustering map formulation module is used to add disaster level identifiers to the multiple critical areas, perform cluster analysis, and formulate a disaster clustering map. The disaster clustering map is used for centralized management of different areas. The risk correction module is used to correct the disaster cluster map based on the tower status data and multi-dimensional basic information, determine the disaster impact level, and identify key risk areas. The resource allocation module is used to manage resource allocation based on the disaster impact level and key risk areas in the disaster cluster map.

Images