Meteorological situation estimation method, device, equipment and medium based on cloud-edge computing

Through the meteorological situation estimation method based on cloud edge calculation, the dynamic K-mean clustering algorithm is used to process meteorological data, which solves the problem that traditional technology is difficult to meet the needs of meteorological situation estimation, realizes the classification of meteorological disasters and the effective evaluation of situations, and improves the prediction Accuracy and timeliness.

CN118779684BActive Publication Date: 2024-12-03四川省农村经济综合信息中心
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411237733.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-05
Publication Date
2024-12-03
Estimated Expiration
2044-09-05

AI Technical Summary

Technical Problem

Traditional meteorological data storage and calculation modes are difficult to meet the accurate, timely and effective needs of meteorological situation estimation, especially due to the complex format, large amount of information, frequent changes and wide coverage areas of meteorological data, and it is necessary to combine geography, hydrology, etc. Relevant data are analyzed.

Method used

The meteorological situation estimation method based on cloud edge calculation is adopted. By obtaining the current meteorological data of each region, the situation elements are extracted, and the dynamic K-mean clustering algorithm is used for clustering, defining the probability level of meteorological disasters, and the situation is carried out based on the clustering results. Understand and visualize presentation.

Benefits of technology

It has achieved effective estimation of meteorological disaster rating and situation, improved the accuracy and timeliness of meteorological disaster prediction, and reduced the losses caused by natural disasters.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118779684B_ABST
    Figure CN118779684B_ABST
Patent Text Reader

Abstract

The present application discloses a method, device, equipment and medium for estimating meteorological situation based on cloud-edge computing, and relates to the field of meteorological data processing technology. The method includes: obtaining the current meteorological data of each region, and determining the information set of meteorological disasters according to the preset situation elements corresponding to the meteorological disasters and the current meteorological data of each region; extracting the situation elements of each unit information in the information set to obtain the situation type characteristics corresponding to each region; inputting the situation type characteristics corresponding to each region into the dynamic K-means clustering algorithm to obtain a preset number of clusters; understanding the situation of each cluster based on the preset relationship, defining the probability level of occurrence of meteorological disasters corresponding to each cluster, and obtaining the current meteorological situation of each region based on the probability level of occurrence of meteorological disasters corresponding to each cluster, and the preset relationship is the relationship between the situation type characteristics and the situation identification. In this way, the classification of meteorological disasters and the effective estimation of meteorological situation can be achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of meteorological data processing, and in particular to a meteorological situation estimation method, device, equipment and medium based on cloud-edge computing. Background Art

[0002] With the frequent occurrence of super typhoons, severe droughts, heavy rainfall, mudslides and other events, in order to better avoid natural disasters, higher and higher requirements are placed on meteorological services. In order to predict the impending natural disasters and minimize and avoid the various losses caused by meteorological disasters, it is of great significance to study the situation estimation algorithm and visualization model of meteorological data.

[0003] However, due to the complex format, large amount of information, frequent changes and wide coverage of meteorological data (such as precipitation, temperature, wind speed, etc.), the storage and processing of meteorological data face severe challenges. In addition, the estimation of meteorological situation often needs to be combined with other related data such as geography and hydrology for reference analysis, and finally obtain the correct situation estimation results. Therefore, the traditional data storage and calculation mode can no longer meet the needs of accurate, timely and effective meteorological situation estimation. Summary of the invention

[0004] The main purpose of this application is to provide a meteorological situation estimation method, device, equipment and medium based on cloud-edge computing, so as to realize the classification of meteorological disasters and effective estimation of meteorological situation.

[0005] To achieve the above objectives, the present application provides a weather situation estimation method based on cloud-edge computing, comprising:

[0006] Acquire current meteorological data of each region, and determine an information set of the meteorological disaster according to preset situation elements corresponding to the meteorological disaster and the current meteorological data of each region, wherein each unit information in the information set corresponds to a region;

[0007] Extracting situation elements from each unit of information in the information set to obtain situation type features corresponding to each region;

[0008] Inputting the situation type characteristics corresponding to each region into a dynamic K-means clustering algorithm to obtain a preset number of clusters;

[0009] Based on the preset relationship, the situation of each cluster is understood, the probability level of the meteorological disaster corresponding to each cluster is defined, and based on the probability level of the meteorological disaster corresponding to each cluster, the current meteorological situation in each region is obtained. The preset relationship is the relationship between situation type characteristics and situation identification.

[0010] Optionally, the situation elements of each unit information in the information set are extracted to obtain the situation type characteristics corresponding to each region, including: based on the understanding of the situation characteristics of the meteorological disaster, the situation elements of each unit information are extracted in turn to obtain the situation type characteristics corresponding to each region.

[0011] Optionally, the situation type features corresponding to each region are input into a dynamic K-means clustering algorithm to obtain a preset number of clusters, including: inputting the situation type features corresponding to each region into the dynamic K-means clustering algorithm in turn, creating a cluster with the first situation type feature as the center; determining the Euclidean distance between the remaining situation type features and each current cluster, wherein the Euclidean distance is determined based on the cluster center of the cluster and the situation type feature; for any situation type feature, filtering out the minimum Euclidean distance from all the Euclidean distances, and comparing the minimum Euclidean distance with the preset threshold of the corresponding cluster to assign a cluster to the situation type feature.

[0012] Optionally, the minimum Euclidean distance is compared with a preset threshold of a corresponding cluster to assign a cluster to the situation type feature, including: if the minimum Euclidean distance is less than or equal to the preset threshold of the cluster corresponding to the minimum Euclidean distance, the situation type feature is assigned to the cluster corresponding to the minimum Euclidean distance, and the cluster center of the cluster is updated; if the minimum Euclidean distance is greater than the preset threshold of the cluster corresponding to the minimum Euclidean distance, a new cluster is created with the situation type feature as the cluster center.

[0013] Optionally, the situation understanding of each cluster is performed based on a preset relationship, and the probability level of occurrence of the meteorological disaster corresponding to each cluster is defined, including: setting a situation space set, the situation space set including situation classifications corresponding to the meteorological disaster; based on the preset relationship, the situation space set and the situation type characteristics in each cluster, determining the probability level of occurrence of the meteorological disaster corresponding to each cluster.

[0014] Optionally, after obtaining the current meteorological situation of each region, the method further includes: determining the future meteorological situation of each region based on the current meteorological situation of each region and a preset case template.

[0015] Optionally, after obtaining the current meteorological situation in each region, the method further includes: visually displaying the current meteorological situation and the future meteorological situation in each region.

[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a meteorological situation estimation device based on cloud-edge computing, including: an acquisition module, used to acquire the current meteorological data of each region, and determine the information set of the meteorological disaster according to the preset situation elements corresponding to the meteorological disaster and the current meteorological data of each region, and each unit information in the information set corresponds to a region; a feature extraction module, used to extract the situation elements of each unit information in the information set to obtain the situation type characteristics corresponding to each region; a clustering module, used to input the situation type characteristics corresponding to each region into the dynamic K-means clustering algorithm to obtain a preset number of clusters; a situation estimation module, used to understand the situation of each cluster based on a preset relationship, define the probability level of the meteorological disaster corresponding to each cluster, and obtain the current meteorological situation of each region based on the probability level of the meteorological disaster corresponding to each cluster, and the preset relationship is the relationship between situation type characteristics and situation identification.

[0017] The present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for estimating a meteorological situation based on cloud-edge computing as described above is implemented.

[0018] The present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for estimating a meteorological situation based on cloud-edge computing as described above is implemented.

[0019] The meteorological situation estimation method based on cloud-edge computing of the present application obtains the current meteorological data of each region, and determines the information set of meteorological disasters according to the preset situation elements corresponding to the meteorological disasters and the current meteorological data of each region; then extracts the situation elements of each unit information in the information set to obtain the situation type characteristics corresponding to each region; further inputs the situation type characteristics corresponding to each region into the dynamic K-means clustering algorithm to obtain a preset number of clusters; finally, understands the situation of each cluster based on the preset relationship, defines the probability level of occurrence of meteorological disasters corresponding to each cluster, and obtains the current meteorological situation of each region based on the probability level of occurrence of meteorological disasters corresponding to each cluster. Therefore, by adopting the theoretical method of situation estimation and using the dynamic K-means clustering algorithm, the classification of meteorological disasters and the effective estimation of meteorological situation are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a topological diagram of a meteorological situation estimation system based on cloud-edge computing in an embodiment of the present application;

[0021] Figure 2 It is one of the flow charts of the weather situation estimation method based on cloud-edge computing in an embodiment of the present application;

[0022] Figure 3 This is the second flowchart of the weather situation estimation method based on cloud-edge computing in an embodiment of the present application;

[0023] Figure 4 is a schematic diagram of a weather situation estimation device based on cloud-edge computing according to an embodiment of the present application;

[0024] Figure 5 An example physical structure diagram of an electronic device is provided.

[0025] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] With the frequent occurrence of super typhoons, severe droughts, heavy rainfall, mudslides and other events, it has become increasingly important to estimate the weather situation. Estimating the weather situation not only helps people better understand the current weather conditions, but also predicts future weather trends and warns of meteorological disasters, providing scientific support for decision-making in various fields, thereby effectively reducing risks, improving efficiency and improving the quality of life.

[0028] However, meteorological data is diverse, with complex formats, large amounts of information, frequent changes, and wide coverage areas, which makes the storage and processing of meteorological data face severe challenges. In addition, the existing meteorological situation estimation algorithms can no longer meet the needs of accurate, timely, and effective meteorological situation estimation.

[0029] To this end, the embodiments of the present application provide a method, device, equipment and medium for estimating weather situation based on cloud-edge computing. By taking the situation estimation model as theoretical guidance, the dynamic K-means clustering algorithm is used to specifically analyze the probability of natural disasters caused by meteorological conditions, and the possibility of natural disasters is graded. Finally, the weather situation estimation analysis chart is visualized through images to form a WebGIS (Web Geographic Information System, World Wide Web Geographic Information System) service product, which provides services to users in real time in the form of a website to remind users to avoid risks and achieve the purpose of reducing the loss of life and property caused by disasters.

[0030] For easy understanding, cloud computing, edge computing and weather situation estimation are explained below.

[0031] Cloud computing is a new computing method that shares resources over the network and users use them as needed. Edge computing is a technology that deploys computing resources at the edge of the network (closer to the data source). Compared with cloud computing, edge computing focuses more on real-time, low latency and data security.

[0032] Situation estimation belongs to the high-level combination of "data-information-knowledge-understanding" in the data fusion hierarchy, emphasizing the problem of obtaining knowledge and understanding from data and providing action strategies. Existing methods for implementing situation estimation systems include template matching, DS evidence theory, Bayesian networks, fuzzy reasoning, expert systems and knowledge reasoning, which are mainly used in military, network security, information fusion research and other fields.

[0033] Meteorological situation estimation is the knowledge processing based on the primary processing of meteorological observation data, and is the knowledge understanding part of meteorological data processing. Meteorological situation estimation can merge and integrate complex and diverse meteorological data, and intelligently and automatically reason about information data. By analyzing the intrinsic connection between data and meteorological events, it helps predict the meteorological disasters that may occur in the next period, such as drought, sandstorm, fire risk, frost, heavy rain, mudslide, etc. Meteorological situation estimation mainly processes, analyzes and predicts meteorological situation elements, and categorizes the possibility of meteorological disasters.

[0034] The following is an introduction to the system network environment of the weather situation estimation method based on cloud-edge computing in an embodiment of the present application.

[0035] Figure 1 It is a topological diagram of the meteorological situation estimation system based on cloud-edge computing in an embodiment of the present application.

[0036] like Figure 1 As shown in the figure, a debris flow situation estimation system is taken as an example. Under unified standard specifications, the debris flow situation estimation system uses Linux and Windows operating systems as the basic platform, stores meteorological observation data in the database, uses a multi-threaded meteorological situation estimation mechanism at the application layer through a workflow engine and timing control to infer the debris flow situation estimation results, and uses Barnes interpolation to interpolate the results into grid field data, and then uses the IDV (Interactive Data Visualization) drawing mechanism to draw the debris flow situation estimation analysis diagram, and finally presents it to the user in the form of WebGIS.

[0037] In the cloud-edge computing environment, the local meteorological bureau's big data cloud platform can be used as the central cloud, and the local economic comprehensive information center computing environment can be used as the edge cloud; further, confidential data such as meteorological data can be stored in the central cloud environment, and debris flow-related data can be stored in the edge cloud environment, thereby reducing network communication and resource usage with the core cloud environment.

[0038] The situation element data in the debris flow situation estimation system is stored in the edge cloud database server, the relevant files are stored in the file server, the graphic workstation is used to generate the debris flow situation estimation analysis diagram and store it in the file server, and the content management server uniformly manages various files and analysis diagram products; finally, the visualization model can be provided to users in the form of a website through the Web server. Users can use PC (Personal Computer), laptops, tablets and smart phones to access the visualization model through a browser. To ensure data security and fault recovery, the backup server can back up data, files, graphics, etc.

[0039] The above is the environment construction process of the meteorological situation estimation system. Based on the above system environment, the meteorological situation estimation method based on cloud-edge computing in the embodiment of the present application is described in detail below.

[0040] Figure 2 This is one of the flow charts of the weather situation estimation method based on cloud-edge computing in the embodiment of the present application. The weather situation estimation method based on cloud-edge computing can be executed by the above-mentioned weather situation estimation system, such as Figure 2 As shown, the weather situation estimation method based on cloud-edge computing may include the following steps:

[0041] Step 210: Acquire current meteorological data of each region, and determine an information set of meteorological disasters based on preset situation elements corresponding to the meteorological disasters and the current meteorological data of each region, wherein each unit of information in the information set corresponds to a region.

[0042] It should be noted that meteorological data can mainly include two categories, climate data and weather data. Specifically, it can include: air temperature, ground temperature, precipitation data, air pressure, humidity, wind speed, visibility, air temperature, air pressure, sunshine time, cloud base height, etc. Meteorological data is not specifically limited here. Meteorological data can be obtained from meteorological databases established by some meteorological agencies, or from special meteorological information service platforms. Preset situation elements refer to related elements that can affect the occurrence of meteorological disasters. Taking mudslides as a meteorological disaster as an example, without considering human factors, preset situation elements can include: terrain, altitude, current rainfall, potential rainfall, earthquakes and other elements that can affect the occurrence of mudslides.

[0043] In this embodiment, when it is necessary to estimate the meteorological situation, the meteorological data and the preset situation elements corresponding to the meteorological disaster can be retrieved from the edge cloud database server first; further, according to the preset situation elements, the meteorological data corresponding to the preset situation elements are found from all meteorological data, and these data are combined into the information set of the meteorological disaster. This information set can be expressed as: ,in, is the relevant meteorological data corresponding to the i-th region that may cause the meteorological disaster. As an example, if the meteorological disaster is a mudslide, then ={topography, altitude, current rainfall, potential rainfall, earthquake}, that is, the terrain, altitude, current rainfall, potential rainfall, earthquake and other data corresponding to the i-th region.

[0044] Step 220: extract situation elements from each unit of information in the information set to obtain situation type features corresponding to each region.

[0045] It should be noted that the unit information is the above The relevant meteorological data corresponding to the i-th region that may cause the meteorological disaster. Situation elements refer to key information or data that can help decision makers quickly and accurately understand the current situation. Extracting situation elements refers to the process of identifying and selecting the most critical information for understanding the current situation during the situation estimation process. The situation type feature can be all situation elements represented by vectors.

[0046] Specifically, after obtaining the information set of the meteorological disaster, situation elements can be extracted from the information set to obtain situation type characteristics corresponding to each region. Still taking mudslide as an example of meteorological disaster, the extracted situation elements can be {mountainous area, altitude 1500m, current heavy rain or continuous heavy rain}, etc.

[0047] In some embodiments, step 220 extracts situation elements from each unit information in the information set to obtain situation type characteristics corresponding to each region, which may include: based on the understanding of the situation characteristics of meteorological disasters, extracting situation elements from each unit information in turn to obtain situation type characteristics corresponding to each region.

[0048] In this embodiment, the situation element extraction process can be recorded as ; Where T is the extraction process of situation elements, Describes the unit information that affects the occurrence of meteorological disasters. Expresses the understanding of the characteristics of meteorological disaster situation, Represents the situation type feature, which is represented in the form of a vector.

[0049] It should be noted here that the recognition of situational characteristics refers to the understanding and analysis of key factors in meteorological disasters, which can help decision makers quickly and accurately grasp the current situation and make reasonable decisions. Taking debris flow as an example of a meteorological disaster, the recognition of situational characteristics can be: an in-depth understanding of the overall trend and characteristics of the occurrence, development and evolution of debris flow disasters. In addition, the recognition of debris flow situational characteristics can be obtained in the following ways: Assuming that debris flow disasters have frequently occurred in a certain area in recent years, in order to deeply understand the situational characteristics of debris flow, data on debris flow events that have occurred in the area in the past few years can be collected, and the time, location and scale of its occurrence can be analyzed. It can be found that debris flow is mainly concentrated in the rainy season and mostly occurs in mountainous areas with steep terrain. Further, a detailed geological survey can be conducted in the area. When a large amount of loose deposits are found on the hillside and the soil water content is high, it is judged that debris flow is prone to occur. A meteorological monitoring station can be established in the area to monitor rainfall and rainfall intensity in real time. When the rainfall reaches a certain threshold, it is judged that debris flow is prone to occur.

[0050] Step 230: Input the situation type characteristics corresponding to each region into the dynamic K-means clustering algorithm to obtain a preset number of clusters.

[0051] It should be noted here that the preset number can be set by the staff according to the actual level classification needs. For example, in the debris flow situation estimation, the probability of debris flow can be divided into 5 levels, namely: debris flow is almost impossible to occur, the possibility of debris flow is small, the possibility of debris flow is large, the possibility of debris flow is high, and the possibility of debris flow is extremely high. Based on this, the number of clusters can be set to 5 to divide the situation type characteristics into 5 groups.

[0052] Figure 3 This is the second flowchart of the weather situation estimation method based on cloud-edge computing in an embodiment of the present application.

[0053] like Figure 3 As shown, in some embodiments, step 230 inputs the situation type features corresponding to each region into the dynamic K-means clustering algorithm to obtain a preset number of clusters, which may include the following steps:

[0054] Step 310: Input the situation type features corresponding to each region into the dynamic K-means clustering algorithm in turn, and create a cluster with the first situation type feature as the center.

[0055] Step 320: Determine the Euclidean distance between each remaining situation type feature and each current cluster. The Euclidean distance is determined based on the cluster center of the cluster and the situation type feature.

[0056] Step 330: For any situation type feature, select the minimum Euclidean distance from all Euclidean distances, and compare the minimum Euclidean distance with a preset threshold of the corresponding cluster to assign a cluster to the situation type feature.

[0057] In this embodiment, since the dynamic K-means clustering algorithm is used, it is not necessary to predefine the number of clusters K, that is, the initial number of clusters is 0. The situation type features corresponding to each region are input into the dynamic K-means clustering algorithm in sequence. When the first situation type feature is input, a cluster is created directly with the first situation type feature. At this time, the cluster center of the cluster is also the first situation type feature. Further, the second situation type feature is input. After the second situation type feature is input, the Euclidean distance between the situation type feature and the cluster center of the currently established cluster is calculated. Because there is only one cluster at this time, when calculating the Euclidean distance, it is only necessary to calculate the Euclidean distance between the second situation type feature and the first situation type feature.

[0058] If there are multiple clusters, the cluster center of each cluster can be calculated based on the mean of all the situation type features in the cluster. At this time, the input situation type feature needs to calculate the Euclidean distance with the cluster center of each cluster. The Euclidean distance d can be calculated by the following formula:

[0059]

[0060] in, is the situation type feature of the i-th input, is the cluster center of the jth cluster, n is the total number of situation type features, is the total number of cluster centers.

[0061] After obtaining the Euclidean distance between the input situation type feature and each cluster center, the smallest Euclidean distance can be screened out from all the Euclidean distances. The cluster center corresponding to the smallest Euclidean distance is the cluster center closest to the input situation type feature. It also means that the input situation type feature may belong to this cluster.

[0062] Further, after finding the cluster to which the input situation type feature may belong, the minimum Euclidean distance is compared with the preset threshold of the cluster. It should be noted that the preset threshold of the cluster can be set according to the knowledge of meteorological experts.

[0063] In some embodiments, the minimum Euclidean distance is compared with a preset threshold of a corresponding cluster, and a cluster is assigned to the situation type feature, which may include: if the minimum Euclidean distance is less than or equal to the preset threshold of the cluster corresponding to the minimum Euclidean distance, the situation type feature is assigned to the cluster corresponding to the minimum Euclidean distance, and the cluster center of the cluster is updated; if the minimum Euclidean distance is greater than the preset threshold of the cluster corresponding to the minimum Euclidean distance, a new cluster is created with the situation type feature as the cluster center.

[0064] Specifically, when the minimum Euclidean distance is less than or equal to the preset threshold of the cluster, the input situation type feature is assigned to the cluster. After the assignment, the mean is recalculated based on all the current situation type features of the cluster to obtain a new cluster center. When the minimum Euclidean distance is greater than the preset threshold of the cluster, it indicates that the input situation type feature does not belong to the cluster. At this time, a new cluster can be established with the input situation type feature as the center.

[0065] Step 240: Understand the situation of each cluster based on the preset relationship, define the probability level of meteorological disasters corresponding to each cluster, and obtain the current meteorological situation of each region based on the probability level of meteorological disasters corresponding to each cluster. The preset relationship is the relationship between situation type characteristics and situation identification.

[0066] It is understandable that after all situation type features are classified using the dynamic K-means clustering algorithm to obtain multiple clusters, the current situation of each cluster can be understood in combination with the knowledge of meteorological experts, so as to judge the possibility and type of meteorological disasters. In the situation analysis process, meteorological expert knowledge plays a decisive role. Therefore, in this embodiment, the corresponding relationship between situation type features and situation identification (i.e., a preset relationship) can be established in advance based on meteorological expert knowledge, so that the situation type features can be directly classified and identified using the preset relationship in the future.

[0067] In some embodiments, the situation of each cluster is understood based on a preset relationship, and the probability level of occurrence of meteorological disasters corresponding to each cluster is defined, which can include: setting a situation space set, the situation space set including situation classifications corresponding to meteorological disasters; based on the preset relationship, the situation space set and the situation type characteristics in each cluster, determining the probability level of occurrence of meteorological disasters corresponding to each cluster.

[0068] Specifically, set the situation space of the meteorological disaster , the elements in the situation space are all possible situation classifications that may appear in the meteorological disaster space (i.e., the probability level of meteorological disasters); Represents the situation type characteristics of meteorological disasters. The relationship between the situation type characteristics of meteorological disasters and the identification of meteorological disaster situation is set as , classification of debris flow situation characteristics: ,in , the situation E constitutes a subset of debris flow situation type features. Thus, through the above process, the situation type features can be classified. After the situation type features are classified, the possibility and type of meteorological disasters corresponding to the situation type features are the possibility and type of meteorological disasters corresponding to the cluster.

[0069] In some embodiments, after obtaining the current meteorological situation in each region, the meteorological situation estimation method based on cloud-edge computing may also include: determining the future meteorological situation in each region based on the current meteorological situation in each region and a preset case template.

[0070] Specifically, after obtaining the current meteorological situation of each region, the future meteorological situation of each region can be predicted based on the current meteorological situation of each region, that is, the current meteorological situation of a region at time t is known. ,beg Where T is the time interval for meteorological situation estimation, which can be set according to actual needs. When predicting the probability of possible meteorological disasters, it is necessary to establish a meteorological disaster case template based on case reasoning and complete the situation prediction based on template matching.

[0071] It should be noted that case-based reasoning refers to the inference of the probability of future meteorological disasters in a region based on the probability of the current meteorological disasters in a region and the relevant data of meteorological disasters in the previous arbitrary days. The method of establishing a meteorological disaster case template can be: collecting historical data related to meteorological disasters in the region, and building a case template based on the historical data; when building a case template, the template attributes can be defined based on the common characteristics and rules in the historical data.

[0072] Taking debris flow as an example of meteorological disaster, first, in order to conduct case reasoning, data on historical debris flow events in the area can be collected and sorted, including information such as the time, location, meteorological conditions, geological conditions, and debris flow scale. This information can be used to build a debris flow case template. When building a debris flow case template, the following factors can be considered: rainfall conditions, including rainfall amount, rainfall duration, rainfall intensity, etc.; geological conditions, such as terrain slope, soil type, etc.; altitude, etc. Based on these attributes, one or more debris flow case templates can be constructed. Then, the meteorological data and geological data monitored in real time are matched with the case template to predict the possibility of debris flow in the future.

[0073] In some embodiments, after obtaining the current meteorological situation in each region, the meteorological situation estimation method based on cloud-edge computing may also include: visually displaying the current meteorological situation and future meteorological situation in each region.

[0074] Specifically, after determining the probability level of occurrence of meteorological disasters corresponding to each cluster, different representation forms can be set for different occurrence probability levels to distinguish them, preferably using different colors for distinction, and finally visually displayed in the form of an image. For example, different occurrence probability levels are represented by different colors in the map.

[0075] The following uses mudslide as a meteorological disaster to introduce the meteorological situation estimation method based on cloud-edge computing in an embodiment of the present application through a specific example.

[0076] Assume that R 山地当前雨量 : Current 24-hour rainfall (mm) at a meteorological station with an altitude greater than 500m and less than 2500m; R 山地当前雨量 : Potential rainfall (mm) at meteorological stations with an altitude greater than 500m and less than 2500m; R 海拔 Indicates altitude; R 地震 represents the earthquake level. Then the information set that affects the occurrence of debris flow in the situation element extraction is S={R 山地当前雨量 , R 山地潜势雨量 , R 海拔 , R 地震}.

[0077] As we all know, landslides and debris flows are mainly distributed in the transition zone between the first and second steps and in the vast area of ​​the second step. The altitude of the second step is concentrated between 500m and 2500m. Assuming that the altitude of the second step is 1300m and the mean square error is 400m, the Gaussian formula of altitude and debris flow probability can be expressed as:

[0078]

[0079] in, is the probability of debris flow, and h is the altitude of the station. From the above, it can be seen that the closer to the mountain with an altitude of 1300m, the higher the probability of debris flow; It is a natural constant with a value of approximately 2.718281828459045.

[0080] Potential rainfall is an important factor in the occurrence of debris flow. According to the potential forecast model of debris flow meteorological statistics: R 山地潜势雨量 , where R 潜势雨量 represents the potential rainfall for debris flow (mm); represents the previous daily rainfall (mm). Combined with the effect of altitude on debris flow, R 山地潜势雨量 =R 潜势雨量 ×2G (h), when the altitude is lower than 0m or higher than 3000m, the mountain potential rainfall is 0mm. When the altitude is 1300m, the mountain potential rainfall reaches the maximum value, R 雨山地潜势量 =R潜势雨量 .

[0081] The situation type characteristics corresponding to the information set S are taken as input and substituted into the dynamic K-means clustering algorithm for calculation. Based on the preset relationship, the situation of each cluster is understood, and five probability levels of debris flow are obtained, namely, debris flow is almost impossible, debris flow is less likely, debris flow is more likely, debris flow is likely to occur, and debris flow is extremely likely, which are represented by none, blue, yellow, orange, and red respectively.

[0082] Table 1 shows the possible situation of debris flow. It should be noted that the data in Table 1 are all experimental data and should not limit the embodiments of the present application. It can be seen from Table 1 that in mountainous areas within a fixed altitude range, the mountain potential rainfall (i.e., previous rainfall) and the mountain current rainfall are the main situation elements for the possible occurrence of debris flow; earthquakes are special situation elements for the possible occurrence of debris flow. In mountainous areas with an altitude range of 1000m to 2000m, after a perceptible earthquake occurs, the surface of the mountain becomes loose and landslides are prone to occur. In addition, heavy rain after the earthquake makes it more likely for debris flow disasters to occur.

[0083] Table 1 Debris flow potential situation table

[0084]

[0085] Therefore, through the visualization model based on cloud-edge collaboration and situation estimation proposed in the embodiment of the present application, the theoretical method of situation estimation is adopted, and the dynamic K-means clustering algorithm is used to classify the possibility of meteorological disasters. The meteorological situation estimation results of various regions are displayed through visualization service products, which has a certain warning effect on meteorological disasters, realizes the classification of meteorological disasters and effective estimation of meteorological situations.

[0086] Based on the above embodiments, the embodiments of the present application also provide a weather situation estimation device based on cloud-edge computing. Figure 4 Schematic diagram of a weather situation estimation device based on cloud-edge computing according to an embodiment of the present application. Figure 4 As shown, the meteorological situation estimation device 400 based on cloud-edge computing may include an acquisition module 410 , a feature extraction module 420 , a clustering module 430 and a situation estimation module 440 .

[0087] Among them, the acquisition module 410 is used to obtain the current meteorological data of each region, and determine the information set of meteorological disasters according to the preset situation elements corresponding to the meteorological disasters and the current meteorological data of each region. Each unit information in the information set corresponds to the region. The feature extraction module 420 is used to extract the situation elements of each unit information in the information set to obtain the situation type features corresponding to each region. The clustering module 430 is used to input the situation type features corresponding to each region into the dynamic K-means clustering algorithm to obtain a preset number of clusters. The situation estimation module 440 is used to understand the situation of each cluster based on the preset relationship, define the probability level of occurrence of meteorological disasters corresponding to each cluster, and obtain the current meteorological situation of each region based on the probability level of occurrence of meteorological disasters corresponding to each cluster. The preset relationship is the relationship between the situation type feature and the situation identification.

[0088] In some implementations, the feature extraction module 420 is specifically used to extract situation elements from each unit of information in turn based on the recognition of situation characteristics of meteorological disasters, and obtain situation type characteristics corresponding to each region.

[0089] In some embodiments, the clustering module 430 is specifically used to: input the situation type features corresponding to each region into the dynamic K-means clustering algorithm in turn, and create a cluster with the first situation type feature as the center; determine the Euclidean distances between the remaining situation type features and the current clusters, and the Euclidean distances are determined based on the cluster center and the situation type features of the cluster; for any situation type feature, filter out the minimum Euclidean distance from all Euclidean distances, and compare the minimum Euclidean distance with the preset threshold of the corresponding cluster to assign a cluster to the situation type feature.

[0090] In some embodiments, the clustering module 430 is also specifically used to: if the minimum Euclidean distance is less than or equal to the preset threshold of the cluster corresponding to the minimum Euclidean distance, the situation type feature is assigned to the cluster corresponding to the minimum Euclidean distance, and the cluster center of the cluster is updated; if the minimum Euclidean distance is greater than the preset threshold of the cluster corresponding to the minimum Euclidean distance, a new cluster is created with the situation type feature as the cluster center.

[0091] In some embodiments, the situation estimation module 440 is specifically used to: set a situation space set, the situation space set includes situation classifications corresponding to meteorological disasters; based on preset relationships, the situation space set and situation type characteristics in each cluster, determine the probability level of occurrence of meteorological disasters corresponding to each cluster.

[0092] In some implementations, the situation estimation module 440 is further specifically used to determine the future weather situation of each region based on the current weather situation of each region and a preset case template.

[0093] In some implementations, the situation estimation module 440 is also specifically used to: visualize the current weather situation and future weather situation of each region.

[0094] It should be noted that for details not disclosed in the weather situation estimation device based on cloud-edge computing in this embodiment, please refer to the details disclosed in the embodiment of the weather situation estimation method based on cloud-edge computing in the embodiments of this specification, and no further details will be given here.

[0095] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the meteorological situation estimation method based on cloud-edge computing, the method comprising: obtaining the current meteorological data of each region, and determining the information set of meteorological disasters according to the preset situation elements corresponding to the meteorological disasters and the current meteorological data of each region, wherein each unit information in the information set corresponds to the region; extracting the situation elements of each unit information in the information set to obtain the situation type features corresponding to each region; inputting the situation type features corresponding to each region into the dynamic K-means clustering algorithm to obtain a preset number of clusters; understanding the situation of each cluster based on the preset relationship, defining the probability level of occurrence of meteorological disasters corresponding to each cluster, and obtaining the current meteorological situation of each region based on the probability level of occurrence of meteorological disasters corresponding to each cluster, wherein the preset relationship is the relationship between the situation type features and the situation identification.

[0096] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0097] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the meteorological situation estimation method based on cloud-edge computing provided by the above methods, which includes: obtaining current meteorological data for each region, and determining an information set of meteorological disasters based on preset situation elements corresponding to the meteorological disasters and the current meteorological data of each region, wherein each unit information in the information set corresponds to a region; extracting situation elements from each unit information in the information set to obtain situation type features corresponding to each region; inputting the situation type features corresponding to each region into a dynamic K-means clustering algorithm to obtain a preset number of clusters; understanding the situation of each cluster based on a preset relationship, defining the probability level of occurrence of meteorological disasters corresponding to each cluster, and obtaining the current meteorological situation of each region based on the probability level of occurrence of meteorological disasters corresponding to each cluster, wherein the preset relationship is the relationship between situation type features and situation identification.

[0098] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the meteorological situation estimation method based on cloud-edge computing provided by the above-mentioned methods, the method comprising: obtaining current meteorological data of each region, and determining an information set of meteorological disasters based on preset situation elements corresponding to the meteorological disasters and the current meteorological data of each region, wherein each unit information in the information set corresponds to a region; extracting situation elements from each unit information in the information set to obtain situation type features corresponding to each region; inputting the situation type features corresponding to each region into a dynamic K-means clustering algorithm to obtain a preset number of clusters; understanding the situation of each cluster based on a preset relationship, defining the probability level of occurrence of meteorological disasters corresponding to each cluster, and obtaining the current meteorological situation of each region based on the probability level of occurrence of meteorological disasters corresponding to each cluster, wherein the preset relationship is the relationship between situation type features and situation identification.

[0099] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A meteorological situation estimation method based on cloud-edge computing, characterized in that: include: Acquire current meteorological data of each region, and determine an information set of the meteorological disaster according to preset situation elements corresponding to the meteorological disaster and the current meteorological data of each region, wherein each unit information in the information set corresponds to a region; Extracting situation elements from each unit of information in the information set to obtain situation type features corresponding to each region; Inputting the situation type characteristics corresponding to each region into a dynamic K-means clustering algorithm to obtain a preset number of clusters; Based on the preset relationship, the situation of each cluster is understood, the probability level of the meteorological disaster corresponding to each cluster is defined, and based on the probability level of the meteorological disaster corresponding to each cluster, the current meteorological situation of each region is obtained, wherein the preset relationship is the relationship between the situation type feature and the situation identification; The process of understanding the situation of each cluster based on the preset relationship and defining the probability level of the meteorological disaster corresponding to each cluster includes: Setting a situation space set, wherein the situation space set includes situation classifications corresponding to the meteorological disasters; Based on the preset relationship, the situation space set and the situation type characteristics in each of the clusters, the occurrence probability level of the meteorological disaster corresponding to each of the clusters is determined.

2. The method for estimating weather situation based on cloud-edge computing according to claim 1, characterized in that: The extracting of situation elements from each unit of information in the information set to obtain situation type features corresponding to each region includes: Based on the recognition of the situation characteristics of the meteorological disasters, situation elements are extracted from each unit information in turn to obtain the situation type characteristics corresponding to each region.

3. The method for estimating weather situation based on cloud-edge computing according to claim 1, characterized in that: The situation type characteristics corresponding to each region are input into the dynamic K-means clustering algorithm to obtain a preset number of clusters, including: Inputting the situation type features corresponding to each region into the dynamic K-means clustering algorithm in turn, and creating a cluster with the first situation type feature as the center; Determine the Euclidean distance between each of the remaining situation type features and each of the current clusters, wherein the Euclidean distance is determined based on the cluster center of the cluster and the situation type feature; For any situation type feature, the minimum Euclidean distance is selected from all the Euclidean distances, and the minimum Euclidean distance is compared with a preset threshold of the corresponding cluster to assign a cluster to the situation type feature.

4. The method for estimating weather situation based on cloud-edge computing according to claim 3, characterized in that: The step of comparing the minimum Euclidean distance with a preset threshold of a corresponding cluster and assigning a cluster to the situation type feature includes: If the minimum Euclidean distance is less than or equal to the preset threshold of the cluster corresponding to the minimum Euclidean distance, the situation type feature is assigned to the cluster corresponding to the minimum Euclidean distance, and the cluster center of the cluster is updated; If the minimum Euclidean distance is greater than a preset threshold of the cluster corresponding to the minimum Euclidean distance, a new cluster is created with the situation type feature as the cluster center.

5. The method for estimating weather situation based on cloud-edge computing according to any one of claims 1 to 4, characterized in that: After obtaining the current weather situation in each region, the method further includes: Determine the future weather situation in each region based on the current weather situation in each region and the preset case template.

6. The method for estimating weather situation based on cloud-edge computing according to claim 5, characterized in that: After obtaining the current weather situation in each region, the method further includes: The current weather situation and the future weather situation of each region are displayed visually.

7. A weather situation estimation device based on cloud-edge computing, characterized in that: include: An acquisition module, used to acquire current meteorological data of each region, and determine an information set of the meteorological disaster according to a preset situation element corresponding to the meteorological disaster and the current meteorological data of each region, wherein each unit information in the information set corresponds to a region; A feature extraction module is used to extract situation elements from each unit of information in the information set to obtain situation type features corresponding to each region; A clustering module, used for inputting the situation type characteristics corresponding to each region into a dynamic K-means clustering algorithm to obtain a preset number of clusters; A situation estimation module, for understanding the situation of each cluster based on a preset relationship, defining the probability level of occurrence of the meteorological disaster corresponding to each cluster, and obtaining the current meteorological situation of each region based on the probability level of occurrence of the meteorological disaster corresponding to each cluster, wherein the preset relationship is the relationship between situation type characteristics and situation identification; The situation estimation module is specifically used to set a situation space set, which includes situation classifications corresponding to the meteorological disasters; based on the preset relationship, the situation space set and the situation type characteristics in each cluster, determine the probability level of occurrence of the meteorological disasters corresponding to each cluster.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the weather situation estimation method based on cloud-edge computing as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the weather situation estimation method based on cloud-edge computing as described in any one of claims 1 to 6 is implemented.