Temperature zone cyclone positioning method and system and storage medium
Through the K-mean classification and density classification method, the problem of multiple cyclone classification and range determination in temperate cyclone positioning is solved, automatic and highly aged cyclone positioning is realized, and accurate forecast support for cyclone-like rainstorms is provided, and the weather process is reduced The impact of economy and life and property.
Patent Information
- Application Number
- CN202411108242.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-26
AI Technical Summary
The existing temperate cyclone positioning methods are difficult to achieve the classification of multiple cyclones, have low positioning efficiency, cannot determine the range of cyclones, and have defects in manual interpretation and objective positioning methods, which cannot provide high timeliness and accurate cyclone center position.
The K-mean classification method and density classification method are used to pre-process the sea level air pressure data, and the low-value areas of sea level air pressure are extracted, and the density is classified to determine the specific number of cyclones, the cyclone area and the central point position.
It realizes automatic and high-time positioning of temperate cyclones, can accurately determine the cyclone area and central location, provides reliable cyclone-like rainstorm forecast support, and reduces the impact of weather processes on economy and life and property.
Smart Images

Figure CN119024462A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of meteorological monitoring technology, and more specifically, to a method, system and storage medium for locating an extratropical cyclone. Background Art
[0002] Heavy rain is seriously destructive to the economy and people's lives and property. It is one of the natural disasters with the highest probability of occurrence and the greatest losses. The forecast of heavy rain has always been a key difficulty in forecasting business. Temperate cyclone-type heavy rain is a typical type of heavy rain in East China. It causes extremely large precipitation and strong intensity, which can easily cause disasters such as floods and waterlogging. The accuracy of the cyclone center position directly affects the forecast accuracy of disastrous weather such as strong winds and heavy rains. Therefore, for the automatic positioning method of temperate cyclones, in terms of output parameters, it is necessary not only to be able to determine the position of the center point, but also to give the range of the cyclone under a unified standard; in terms of positioning process and results, the positioning method needs to be universal, highly timely, and able to accurately and automatically locate.
[0003] There are two main methods for locating temperature cyclones: 1. Manual interpretation and positioning; 2. Objective positioning of temperate cyclones, which sets multiple thresholds and screens the location of the cyclone center through step-by-step iteration. However, the first method cannot automatically identify and locate cyclones, and the second method still has defects such as the inability to classify multiple cyclones, large amount of calculation, low timeliness, and inability to determine the range of cyclones. Summary of the invention
[0004] In order to solve the problems that the current temperate cyclone methods have difficulty in completing multiple cyclone classifications, low positioning efficiency, and inability to determine the cyclone range, the present invention proposes a temperate cyclone positioning method, system and storage medium, which realize cyclone classification through K-means and density classification methods, and can automatically, efficiently and accurately locate each cyclone area and center position.
[0005] In order to achieve the above technical effects, the technical solution of the present invention is as follows:
[0006] A method for locating an extratropical cyclone, comprising:
[0007] S1. Collect sea level pressure data and preprocess it;
[0008] S2. Based on the sea level pressure data, the K-means classification method is used to extract the low sea level pressure area;
[0009] S3. Density classification is performed on the areas with low sea level pressure, and the specific number of cyclones, cyclone areas and center locations are determined based on the classification results.
[0010] This technical solution uses the K-means and density classification methods to quickly and accurately locate the scope and center position of each cyclone area, which can provide reliable support for the accurate forecast of cyclone-like rainstorms, thereby reducing the impact of such weather processes on the economy and the safety of life and property.
[0011] Preferably, in step S1, the preprocessing process includes:
[0012] S11. Collecting raw sea level pressure data, including longitude and latitude of spatial points;
[0013] S12. cutting the original sea level pressure data according to the boundary of the target area;
[0014] S13. Convert the data format of the cropped sea level pressure data and save it as the initial data set.
[0015] Here, the original data is cropped to reduce the data size for subsequent processing.
[0016] Preferably, in step S13, the format of the converted initial data set is as follows:
[0017] [P 0 ,P 1 ,P 2 ,P 3 ,P 4 ,P 5 ,…,P i ,…,P n ]
[0018] Among them, P i Represents the information of the i-th spatial point, n represents the total number of n grid points in the target area after cropping, i = 0, 1, 2, ..., n.
[0019] Preferably, the information of the i-th spatial point includes the longitude lon i , dimension lat i and sea level pressure Pre i , the specific matrix of the initial data set format is as follows:
[0020]
[0021] Among them, P i =[lon i ,lat i ,,Pre i ].
[0022] Here, the initial data set format is converted into a matrix form to facilitate processing in subsequent steps.
[0023] Preferably, in step S2, based on the preprocessed sea level pressure data, a K-means classification method is used to classify all spatial points in the target area; wherein the type in which the upper and lower quartiles of the sea level pressure value are both minimum values is recorded as a low pressure area.
[0024] Here, by calculating the upper and lower quartiles of the data set, we can avoid the k value from having a significant impact on the classification results.
[0025] Preferably, a K-means classification method is used to classify all spatial points in the target area, including the following steps:
[0026] S21. Set the total number of categories k;
[0027] S22. Using the K-means classification method, calculate the type of each spatial point;
[0028] S23. The information of each type of spatial point [lon i ,lat i ,,Pre i ] is constructed as a data set of each type, and a total of k data sets of corresponding types are established;
[0029] S24. Calculate the sea level pressure values Pre in k data sets respectively i The lower quartile size a25 and the upper quartile size a75 are retained to 2 decimal places respectively;
[0030] S25. Find the data set j when a25 is the minimum value, and determine whether a75 of data set j is the minimum value of a75 among k data sets. If so, record the type corresponding to data set j as a low sea level pressure area. Otherwise, set k=k+1 and return to step S22.
[0031] S26. Output and save the data set j representing the area of low sea level pressure as the input of step S3.
[0032] Here, all spatial points representing all cyclones can be quickly and accurately screened out through the K-means classification method.
[0033] Preferably, in step S3, density classification is performed on the low sea level pressure area based on the spatial density characteristics, comprising the following steps:
[0034] S31. Based on the spatial density distribution characteristics, the spatial points in the low-value sea level pressure area are classified and the noise points are eliminated;
[0035] S32. All spatial points of the same type are recorded as the same cyclone, and the information of the corresponding spatial points is constructed into an independent cyclone data set, i.e., a cyclone area; the number of the cyclone data sets is the number of cyclones;
[0036] S33. Based on the information of the spatial points in each cyclone data set, respectively calculate the longitude and latitude corresponding to the minimum point of the sea level pressure value, and calculate the average longitude and latitude of the minimum point, and mark the longitude and latitude coordinates as the center point position of the corresponding cyclone;
[0037] S34. Output the number of cyclones, cyclone areas and the positions of the center points of the cyclones.
[0038] Here, when there are multiple cyclones in the study area, it is necessary to distinguish which spatial points represent the same cyclone. This method uses a density-based classification method to determine the spatial points contained in each cyclone according to the spatial distribution characteristics.
[0039] Preferably, step S31 includes the following steps:
[0040] S311. Set the scanning radius eps and the minimum number of included points mp;
[0041] S312. Select an unvisited space point and find all points within a circle with the space point as the center and a radius eps;
[0042] S313. Compare the number of points obtained in step S312 with the minimum number of included points mp. If the number of points is less than mp, the spatial point is temporarily marked as a noise point; otherwise, the spatial point and its nearby spatial points form a cluster, and the starting point is marked as visited;
[0043] S314. Determine whether all the spatial points in the cluster have been marked as visited. If so, proceed to step S315. Otherwise, select the spatial points in the cluster that are not marked as visited and repeat steps S312 to S314.
[0044] S315. Determine whether all spatial points have been marked. If so, determine all clusters and noise points, and the spatial points in each cluster are of the same type; otherwise, return to step S312.
[0045] Here, under unified standards, the size and shape of each cyclone area and the location of the cyclone center can be accurately determined, providing a forecast basis for the movement path and development of the cyclone, and helping to forecast disastrous weather such as heavy rain.
[0046] A computer storage medium is used for computer-readable storage, wherein a program for locating a temperate cyclone is stored on the computer storage medium, and when the program for locating a temperate cyclone is executed by a processor, it is used to implement the steps of the above-mentioned method for locating a temperate cyclone.
[0047] A temperate cyclone positioning system, comprising:
[0048] Data acquisition module, used to collect sea level pressure data and perform preprocessing;
[0049] The low-pressure area extraction module uses the K-means classification method to extract the low-pressure area at sea level based on the sea level pressure data;
[0050] The cyclone classification and positioning module is used to classify the density of low sea level pressure areas and determine the specific number of cyclones, cyclone areas and center point locations;
[0051] The output module is used to output the specific number of cyclones, cyclone area and center point location.
[0052] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:
[0053] The present invention proposes a method, system and storage medium for locating a temperate cyclone. First, sea level pressure data is collected and preprocessed. Then, based on the sea level pressure data, a K-means classification method is used to extract low-value sea level pressure areas. Finally, density classification is performed on the low-value sea level pressure areas to determine the specific number of cyclones, cyclone areas and center point locations. This method can not only realize automatic positioning of temperate cyclones and avoid interference from human factors, but also has high timeliness and can complete the classification of multiple cyclones when cyclone data is incomplete. In addition, under a unified standard, the size and shape of each cyclone area and the position of the cyclone center point can be determined, providing a forecast basis for the movement path and development of the cyclone, and facilitating the forecast of disastrous weather such as heavy rain. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 A schematic diagram showing a flow chart of a method for locating a temperate cyclone proposed in Embodiment 1 of the present invention;
[0055] Figure 2 represents the original data of sea level pressure data in Example 1 of the present invention;
[0056] Figure 3 A schematic diagram showing spatial points after K-means classification in Example 1 of the present invention;
[0057] Figure 4 A schematic diagram showing the extraction result of the low-value sea level pressure area in Example 1 of the present invention;
[0058] Figure 5 A schematic diagram showing the output results of the extratropical cyclone positioning method proposed in Example 1 of the present invention;
[0059] Figure 6 A schematic diagram showing a computer device proposed in Embodiment 2 of the present invention;
[0060] Figure 7 A schematic diagram showing the structure of the temperate cyclone positioning system proposed in Example 3 of the present invention. DETAILED DESCRIPTION
[0061] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;
[0062] In order to better illustrate the present embodiment, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the actual size;
[0063] It is understandable to those skilled in the art that descriptions of certain well-known contents in the drawings may be omitted.
[0064] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.
[0065] The positional relationships described in the drawings are only for illustrative purposes and should not be construed as limiting the present patent;
[0066] Example 1
[0067] like Figure 1 As shown, this embodiment proposes a method for locating an extratropical cyclone. Taking the sea level pressure data of ERA5 at 18:00 (UTC) on April 29, 2021 as an example, the method proposed in this embodiment is used to locate an extratropical cyclone. The process is as follows:
[0068] S1. Obtain sea level pressure data and perform preprocessing. In this embodiment, the sea level pressure data of ERA5 at 18:00 (UTC) on April 29, 2021 is as follows: Figure 2 As shown;
[0069] The pretreatment process includes:
[0070] S11. Collecting raw sea level pressure data, including longitude and latitude of spatial points;
[0071] S12. According to the boundary of the target area, the original sea level pressure data is cropped to reduce the data size for subsequent processing;
[0072] S13. Convert the data format of the cropped sea level pressure data and save it as an initial data set; the format of the converted initial data set is as follows:
[0073] [P 0 ,P 1 ,P 2 ,P 3 ,P 4 ,P 5 ,…,P i ,…,P n ]
[0074] Among them, P iIndicates the information of the i-th spatial point, n indicates that there are a total of n grid points in the target area after clipping, i = 0, 1, 2, ..., n; the information of the i-th spatial point includes the longitude lon i , dimension lat i and sea level pressure Pre i , the specific matrix of the initial data set format is as follows:
[0075]
[0076] Among them, P i =[lon i ,lat i ,Pre i ]; Here, the initial data set format is converted into a matrix form to facilitate subsequent processing steps.
[0077] S2. Based on the preprocessed sea level pressure data, use the K-means classification method to classify all spatial points in the target area and extract the low sea level pressure area; wherein the type where the upper and lower quartiles of the sea level pressure value are both minimum values is recorded as the low pressure area, which specifically includes the following steps:
[0078] S21. Set the total number of categories k;
[0079] S22. Use the K-means classification method to calculate the type of each spatial point. The specific process is as follows:
[0080] Randomly give k initial cluster centers, calculate the distance from each spatial point to the cluster center, and assign it to the nearest cluster center to form k clusters; calculate the centroid of each cluster and regard it as the new k cluster centers; repeat the above steps until the k cluster centers are no longer changing, and record the spatial points in each cluster as the same type, forming a total of k types; in this embodiment, Figure 3 As shown, spatial points are divided into 21 types.
[0081] S23. The information of each type of spatial point [lon i ,lat i ,Pre i ] is constructed as a data set of each type, and a total of k data sets of corresponding types are established;
[0082] S24. Calculate the sea level pressure values Pre in k data sets respectively i The lower quartile size a25 and the upper quartile size a75 are retained to 2 decimal places respectively;
[0083] S25. Find the data set j when a25 is the minimum value, and determine whether a75 of data set j is the minimum value of a75 among k data sets. If so, record the type corresponding to data set j as a low sea level pressure area. Otherwise, set k=k+1 and return to step S22.
[0084] S26. Output and save the data set j representing the area of low sea level pressure as the input of step S3.
[0085] In this embodiment, the extraction result of the low value area of sea level pressure is as follows: Figure 4 As shown, the red scattered points are all the areas with low sea level pressure, which are type 11 in the K-means classification, and the base map is the sea level pressure (hPa).
[0086] S3. When there are multiple cyclones in the study area, it is necessary to distinguish which spatial points represent the same cyclone. This embodiment classifies the density of the low-value sea level pressure area to determine the specific number of cyclones, the cyclone area, and the center point location, including the following steps:
[0087] S31. Based on the spatial density distribution characteristics, the spatial points in the low-value sea level pressure area are classified and the noise points are removed. The process is as follows:
[0088] S311. Set the scanning radius eps and the minimum number of included points mp;
[0089] S312. Select an unvisited space point and find all points within a circle with the space point as the center and a radius eps;
[0090] S313. Compare the number of points obtained in step S312 with the minimum number of included points mp. If the number of points is less than mp, the spatial point is temporarily marked as a noise point; otherwise, the spatial point and its nearby spatial points form a cluster, and the starting point is marked as visited;
[0091] S314. Determine whether all the spatial points in the cluster have been marked as visited. If so, proceed to step S315. Otherwise, select the spatial points in the cluster that are not marked as visited and repeat steps S312 to S314.
[0092] S315. Determine whether all spatial points have been marked. If so, determine all clusters and noise points, and the spatial points in each cluster are of the same type; otherwise, return to step S312;
[0093] S32. All spatial points of the same type are recorded as the same cyclone, and the information of the corresponding spatial points is constructed into an independent cyclone data set, i.e., a cyclone area; the number of the cyclone data sets is the number of cyclones;
[0094] S33. Based on the information of the spatial points in each cyclone data set, respectively calculate the longitude and latitude corresponding to the minimum point of the sea level pressure value, and calculate the average longitude and latitude of the minimum point, and mark the longitude and latitude coordinates as the center point position of the corresponding cyclone;
[0095] S34. Output the number of cyclones, cyclone area and the center point of the cyclone, such as Figure 5 As shown, the purple area in the figure is the noise point, the other four areas are the cyclone centers, and the contour lines are the sea level pressure field.
[0096] Here, under a unified standard, the temperate cyclone positioning method provided in this embodiment can accurately determine the size, shape and location of each cyclone area and the center point of the cyclone, provide a forecast basis for the movement path and development of the cyclone, and provide reliable support for the accurate forecast of cyclone-type rainstorms, thereby reducing the impact of such weather processes on the economy and the safety of life and property.
[0097] Example 2
[0098] See also Figure 6 The present application also proposes a computer device, including a processor, a memory, and a computer program stored in the memory, the processor is numbered 1, the memory is numbered 2, and the processor 1 is connected to the memory 2. The processor 1 executes the computer program stored in the memory 2 to implement the extratropical cyclone positioning method described in the embodiment 1.
[0099] The memory 2 may be a disk, a flash memory or any other non-volatile storage medium, see Figure 6 The processor 1 is connected to the memory 2 and can be implemented as one or more integrated circuits, specifically a microprocessor or a microcontroller. When executing a computer program stored in the memory, the extratropical cyclone positioning method is implemented for the global model.
[0100] The present application also proposes a computer-readable storage medium, on which computer program instructions are stored. When the instructions are executed by a processor, the steps of the extratropical cyclone positioning method are implemented.
[0101] Example 3
[0102] In this embodiment, if Figure 7 As shown, a temperate cyclone positioning system is proposed, the system comprising:
[0103] The data acquisition module 101 is used to collect sea level pressure data and perform preprocessing;
[0104] The low pressure area extraction module 102 extracts the low sea level pressure area based on the sea level pressure data using the K-means classification method;
[0105] The cyclone classification and positioning module 103 is used to classify the density of the low-value sea level pressure area and determine the specific number of cyclones, the cyclone area and the center point location;
[0106] The output module 104 is used to output the specific number of cyclones, cyclone areas and center point locations.
[0107] Obviously, the above embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.
Claims
1. A method for locating an extratropical cyclone, characterized in that: include: S1. Collect sea level pressure data and preprocess it; S2. Based on the sea level pressure data, the K-means classification method is used to extract the low sea level pressure area; S3. Density classification is performed on the areas with low sea level pressure, and the specific number of cyclones, cyclone areas and center locations are determined based on the classification results.
2. The method for locating a temperate cyclone according to claim 1, characterized in that: In step S1, the preprocessing process includes: S11. Collecting raw sea level pressure data, including longitude and latitude of spatial points; S12. cutting the original sea level pressure data according to the boundary of the target area; S13. Convert the data format of the cropped sea level pressure data and save it as the initial data set.
3. The method for locating a temperate cyclone according to claim 2, characterized in that: In step S13, the format of the converted initial data set is as follows: [P0,P1,P2,P3,P4,P5,…,P i ,…,P n ] Among them, P i Represents the information of the i-th spatial point, n represents the total number of n grid points in the target area after cropping, i = 0, 1, 2, ..., n.
4. The method for locating a temperate cyclone according to claim 3, characterized in that: The information of the i-th spatial point includes the longitude lon i , dimension lat i and sea level pressure Pre i , the specific matrix of the initial data set format is as follows: Among them, P i =[lon i ,lat i ,,Pre i ].
5. The method for locating a temperate cyclone according to claim 4, characterized in that: In step S2, based on the preprocessed sea level pressure data, all spatial points in the target area are classified using the K-means classification method; wherein the type where the upper and lower quartiles of the sea level pressure values are both minimum values is recorded as a low pressure area.
6. The method for locating a temperate cyclone according to claim 5, characterized in that: Using the K-means classification method, all spatial points in the target area are classified, including the following steps: S21. Set the total number of categories k; S22. Using the K-means classification method, calculate the type of each spatial point; S23. The information of each type of spatial point [lon i ,lat i ,,Pre i ] is constructed as a data set of each type, and a total of k data sets of corresponding types are established; S24. Calculate the lower quartile size a25 and the upper quartile size a75 of the sea level pressure values in the k data sets respectively; S25. Find the data set j when a25 is the minimum value, and determine whether a75 of data set j is the minimum value of a75 among k data sets. If so, record the type corresponding to data set j as a low sea level pressure area. Otherwise, set k=k+1 and return to step S22. S26. Output and save the data set j representing the area of low sea level pressure as the input of step S3.
7. The method for locating a temperate cyclone according to claim 6, characterized in that: In step S3, based on the spatial density characteristics, density classification is performed on the low-value sea level pressure area, including the following steps: S31. Based on the spatial density distribution characteristics, the spatial points in the low-value sea level pressure area are classified and the noise points are eliminated; S32. All spatial points of the same type are recorded as the same cyclone, and the information of the corresponding spatial points is constructed into an independent cyclone data set, i.e., a cyclone area; the number of the cyclone data sets is the number of cyclones; S33. Based on the information of the spatial points in each cyclone data set, respectively calculate the longitude and latitude corresponding to the minimum point of the sea level pressure value, and calculate the average longitude and latitude of the minimum point, and mark the longitude and latitude coordinates as the center point position of the corresponding cyclone; S34. Output the number of cyclones, cyclone areas and the positions of the center points of the cyclones.
8. The method for locating a temperate cyclone according to claim 7, characterized in that: Step S31 includes the following steps: S311. Set the scanning radius eps and the minimum number of included points mp; S312. Select an unvisited space point and find all points within a circle with the space point as the center and a radius eps; S313. Compare the number of points obtained in step S312 with the minimum number of included points mp. If the number of points is less than mp, the spatial point is temporarily marked as a noise point; otherwise, the spatial point and its nearby spatial points form a cluster, and the starting point is marked as visited; S314. Determine whether all the spatial points in the cluster have been marked as visited. If so, proceed to step S315. Otherwise, select the spatial points in the cluster that are not marked as visited and repeat steps S312 to S314. S315. Determine whether all spatial points have been marked. If so, determine all clusters and noise points, and the spatial points in each cluster are of the same type; otherwise, return to step S312.
9. A computer storage medium for computer-readable storage, characterized in that: The computer storage medium stores a program for locating a temperate cyclone. When the program for locating a temperate cyclone is executed by a processor, it is used to implement the steps of the method for locating a temperate cyclone according to any one of claims 1 to 8.
10. A temperate cyclone positioning system, characterized in that: include: Data acquisition module, used to collect sea level pressure data and perform preprocessing; The low-pressure area extraction module uses the K-means classification method to extract the low-pressure area at sea level based on the sea level pressure data; The cyclone classification and positioning module is used to classify the density of low sea level pressure areas and determine the specific number of cyclones, cyclone areas and center point locations; The output module is used to output the specific number of cyclones, cyclone area and center point location.
Citation Information
Cited By
Explosive cyclone identification and intelligent classification method
CN121211193A
Recognition of explosive cyclones and intelligent classification method thereof
CN121211193B
