A thunderstorm disaster area analysis mapping method

Abstract

This application provides a method for analyzing and mapping thunderstorm disaster areas, including: acquiring data on the cross-sectional width distribution of valleys, data on the radius of curvature of the centerline, and data on the lateral position of the near-surface wind convergence zone within the valley; integrating the acquired data into a unified dataset; calculating the ratio of the radius of curvature to the valley width based on the radius of curvature data and valley width data in the unified dataset, identifying sharp bends where the ratio exceeds a preset threshold, and identifying potential asymmetric convergence areas; based on the location of the convergence center where the convergence zone gathers towards the outside of the bend, shifting the lateral positioning axis of the thunderstorm landing area from the valley floor centerline to the location of the convergence center, generating a corrected landing area positioning axis; and combining the adjusted thunderstorm landing area coordinates with the valley width distribution data to generate a landing area positioning layer in the disaster mapping, smoothing the boundaries of areas with high curvature, and forming a thunderstorm disaster landing area distribution map.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for analyzing and mapping thunderstorm disaster areas. Background Technology

[0002] In the fields of meteorological disaster early warning and regional analysis, accurately predicting the location of extreme weather events such as thunderstorms is of paramount importance. This not only concerns public safety but also directly impacts the efficiency of disaster prevention and mitigation and the rationality of resource allocation. Precise location of thunderstorm disasters provides reliable information for local governments and emergency departments, reducing unnecessary economic losses and casualties. Existing methods perform reasonably well in flat or large-scale terrains, but frequently fail in localized terrains such as mountain valleys. This is because these methods often simplify terrain as a background factor, ignoring its dynamic interference with weather systems. In winding valleys or narrow mountain passages, orographic lifting and channel effects amplify airflow disturbances. This deviation stems not only from differences in terrain height but also from the multiple response mechanisms of valley floor airflow, further amplifying terrain sensitivity issues. The width of a valley determines the degree of airflow convergence at its bottom: narrow valleys promote airflow acceleration and convergence, forming strong updraft zones; wide valleys allow airflow diffusion, reducing local instability. The degree of curvature alters the direction and distribution of airflow. In straight valleys, airflow generally flows along the axis, while bends in winding sections induce secondary circulation, distorting the airflow trajectory and causing airflow separation in certain areas, forming irregular convergence zones. For example, in U-shaped or V-shaped valleys with sharp bends, airflow struggles to maintain symmetrical convergence at the valley floor center, instead bulging outwards from the bend, forming asymmetrical convergence zones. This deviation contradicts the conventional assumption of central convergence, causing significant discrepancies between predicted and actual conditions. Therefore, accurately identifying airflow deviations caused by curvature in valley topography and reasonably correcting the lateral location of thunderstorm areas has become a crucial issue urgently needing resolution in meteorological disaster area analysis. Summary of the Invention

[0003] This invention provides a method for analyzing and mapping thunderstorm disaster areas, mainly including: Data on the cross-sectional width distribution along the valley, the centerline curvature radius, and the lateral location of the near-surface wind convergence zone within the valley were acquired and integrated into a unified dataset. The ratio of radius of curvature to valley width is calculated based on the radius of curvature data and valley width data in the unified dataset. Sharp bends with a ratio exceeding a preset threshold are identified, and potential asymmetric convergence areas are identified. Based on the lateral location data of the near-surface wind convergence zone in the valley, the lateral location distribution of the near-surface convergence zone in the sharp bend section is analyzed. This lateral location distribution is matched with the potential asymmetric convergence area marker to identify the location of the convergence center where airflow separation causes the convergence zone to gather towards the outside of the bend. Based on the location of the convergence center where the convergence zone gathers towards the outside of the bend, the lateral positioning axis of the thunderstorm landing area is moved from the valley centerline to the location of the convergence center, thus generating the corrected landing area positioning axis. Based on the revised landing area positioning axis, the distribution pattern of the convergence zone in the section where the large-scale weather system forecast location is located is evaluated. In the asymmetric convergence section, the forecast location is moved along the outer direction of the curve to the location of the landing area positioning axis, and the adjusted coordinates of the thunderstorm landing area location are obtained. By combining the adjusted coordinates of the thunderstorm landing area with the valley width distribution data, a landing area positioning layer is generated in the disaster mapping, and the boundary of the high curvature area is smoothed to form a thunderstorm disaster landing area distribution map. Based on the distribution map of thunderstorm disaster areas, the asymmetric convergence characteristics of the high curvature segment are analyzed, the degree of fit between this characteristic and the lateral position data of the convergence zone is evaluated, and an optimized version of the distribution map is generated.

[0004] Furthermore, data on the valley's cross-sectional width distribution along its length, the centerline radius of curvature, and the lateral location of the near-surface wind convergence zone within the valley are obtained, including: Cross-sectional profile data were acquired segment by segment along the valley axis, and the horizontal distance between the left and right boundaries of the valley floor was extracted as the valley width data to form a distribution sequence along the valley. Obtain the coordinates of the valley floor center point, fit an arc through adjacent center points, and obtain the curvature radius data of the center line of the middle segment; Wind speed and direction observation nodes are set up to collect wind direction vectors and calculate the angle between wind direction vectors of adjacent nodes. If the angle exceeds a preset threshold, it is determined to be an airflow convergence area and marked as the location of the convergence zone.

[0005] Furthermore, based on the radius of curvature data and valley width data in the unified dataset, the ratio of radius of curvature to valley width is calculated. Sharp bends with this ratio exceeding a preset threshold are identified, and potential asymmetric convergence areas are identified, including: Read the radius of curvature data and the corresponding valley width data from the unified dataset, calculate the ratio between the radius of curvature data and the corresponding valley width data, and if the ratio is lower than the preset threshold, it is determined to be a sharp bend section, and the sampling point number and axial coordinate are marked as the sharp bend section identifier. The outer direction of the bend is determined based on the change in the centerline coordinates of adjacent sampling points; Extract the lateral position data of the convergence zone within the sharp bend section, compare the lateral offset direction with the outer direction of the bend, and if they are consistent, determine that there is a potential asymmetric convergence feature, and generate the potential asymmetric convergence area identifier.

[0006] Furthermore, based on the lateral location data of the near-surface wind convergence zone within the valley, the lateral distribution of the near-surface convergence zone in the sharp bend section was analyzed. This lateral location distribution was matched with potential asymmetric convergence area markers to identify the location of the convergence center where airflow separation causes the convergence zone to gather towards the outside of the bend, including: Based on the identification of the potential asymmetric convergence area, the lateral position data of the convergence zone in the sharp bend section are screened, and a position distribution sequence is formed by combining the lateral offset distance. Compare the offset direction with the outer direction of the bend point by point. If the offset direction is consistent with the outer direction of the bend and the offset distance exceeds the preset threshold, it is marked as an outer aggregation point. The number of aggregation points on the outer side is counted using a fixed window along the axial direction. The area covered by the window with the largest number of points is determined as the convergence zone aggregation area. The average value of the lateral coordinates and the median value of the axial coordinates within this area are calculated to obtain the location of the aggregation center.

[0007] Furthermore, based on the location of the convergence center where the convergence zone gathers towards the outside of the bend, the lateral positioning axis of the thunderstorm landing area is moved from the valley centerline to the location of this convergence center, generating a corrected landing area positioning axis, including: Based on the axial and lateral coordinates of the aggregation center location, a lateral offset vector is established from the valley centerline to the aggregation center location, including the offset direction and offset distance; For the sharp bend section covered by the lateral offset vector, the lateral positioning centerline of the thunderstorm landing area is shifted from the original centerline along the offset direction to the lateral position where the gathering center is located. Record the axial and lateral coordinates of the positioning centerline at each sampling point after migration to obtain the corrected positioning centerline of the landing area.

[0008] Furthermore, based on the revised location centerline, the convergence zone distribution pattern of the large-scale weather system forecast location segment is assessed. In asymmetric convergence segments, the forecast location is moved along the outer direction of the curve to the location of the location centerline, resulting in the adjusted coordinates of the thunderstorm location, including: Obtain the original location coordinates of the thunderstorm area in the large-scale weather forecast, locate the valley section number according to the axial component, and extract the lateral coordinate value of the corresponding section's location center axis. Calculate the deviation between the original horizontal coordinate and the central horizontal coordinate. If the deviation is greater than a preset threshold, it is determined that there is an asymmetric convergence feature. Using the deviation as the migration amount, the lateral coordinates of the predicted location are translated along the outer side of the curve to the location of the central axis of the landing area, while keeping the axial coordinates unchanged. The coordinates after translation are recorded to obtain the adjusted coordinates of the thunderstorm landing area.

[0009] Furthermore, by combining the adjusted coordinates of the thunderstorm impact area with the valley width distribution data, an impact area positioning layer is generated in the disaster mapping, smoothing the boundaries of areas with high curvature, thus forming a thunderstorm disaster impact area distribution map, including: The width value of the corresponding axial position is extracted from the valley width distribution data. The adjusted thunderstorm landing area coordinates are used as the center and extended to both sides according to a preset ratio to determine the lateral coverage range. The connection is then used to form the initial boundary outline of the landing area positioning layer. For sharp bends where the ratio of radius of curvature to valley width is below a preset threshold, a moving average filter is used to process the boundary point coordinates to generate a smoothed boundary. The smoothed boundary connections form closed polygonal regions, which are then superimposed onto the geographic base map as the location layer of the affected area. Asymmetric convergence area markers are added to form the distribution map of the thunderstorm disaster affected area.

[0010] Furthermore, based on the distribution map of thunderstorm disaster areas, the asymmetric convergence characteristics of the high curvature segment were analyzed. The fit between these characteristics and the lateral position data of the convergence zone was evaluated, and an optimized version of the distribution map was generated, including: Extract the high-curvature boundary coordinate sequence from the thunderstorm disaster area distribution map, calculate the lateral offset distribution relative to the positioning axis of the area, and identify asymmetric convergence feature regions offset to the outside of the curve; Extract the measured position sequence of the corresponding segment from the lateral position data of the convergence zone, and calculate the average value of the lateral distance difference as the degree of fit by comparing point by point; If the fit value exceeds the preset threshold, the landing area positioning centerline is corrected by shifting the difference direction and average value, the boundary range is redefined and smoothing is performed, an updated landing area positioning layer is generated, and it is superimposed on the geographic base map to obtain an optimized distribution map version.

[0011] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a method for analyzing and mapping thunderstorm disaster areas. Addressing the unique operational scenario where thunderstorm location in valley terrain is affected by asymmetric convergence characteristics, this method integrates valley width, radius of curvature, and lateral position data of wind convergence zones to construct a unified dataset. It calculates the ratio of radius of curvature to valley width, identifies potential asymmetric convergence areas in sharp bends, and, combined with the distribution of near-surface convergence zones, accurately locates the convergence center caused by airflow separation. This invention corrects the forecast location by shifting the thunderstorm location axis to the convergence center, generating adjusted location coordinates. Combined with smoothed boundaries in disaster mapping, it forms a distribution map of thunderstorm disaster locations. Furthermore, it optimizes the map version by evaluating the fit between asymmetric convergence characteristics and data. The core innovation of this invention lies in achieving precise location of thunderstorm locations based on coupled analysis of terrain and wind fields, improving the accuracy of meteorological disaster forecasts in complex terrain and providing a scientific basis for disaster early warning. Attached Figure Description

[0012] Figure 1 This is a flowchart of a thunderstorm disaster area analysis and mapping method according to the present invention.

[0013] Figure 2 This is a schematic diagram of a thunderstorm disaster area analysis and mapping method according to the present invention.

[0014] Figure 3 This is another schematic diagram of a thunderstorm disaster area analysis and mapping method according to the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0016] like Figures 1-3 This embodiment of a method for analyzing and mapping thunderstorm disaster areas may specifically include: Step S101: Obtain data on the distribution of valley cross-sectional width along the valley, data on the radius of curvature of the centerline, and data on the lateral position of the near-surface wind convergence zone within the valley, and integrate the obtained data into a unified dataset.

[0017] Cross-sectional profile data were acquired segment by segment along the valley axis at variable intervals. For each cross-section, the horizontal distance between the left and right boundaries of the valley floor was extracted as the valley's cross-sectional width. These widths were recorded sequentially according to the sampling sequence to form a distribution sequence along the valley. During sampling, the coordinates of the valley floor center point corresponding to each cross-section were simultaneously extracted. An arc was fitted based on the coordinates of three adjacent center points, and the radius of this arc was taken as the curvature radius of the centerline of the middle segment. The least squares method was used for fitting, and the radius value was assigned to the middle center point. Multiple wind speed and direction observation nodes were deployed at near-surface altitude within the valley. After collecting the wind direction vectors at each node, the angle between the wind direction vectors of adjacent nodes was calculated using the formula... Where u and v are the wind direction vectors of adjacent nodes, if the included angle exceeds a preset convergence angle threshold of 120 degrees, it is determined that the area has airflow convergence characteristics, because exceeding 120 degrees indicates that the wind direction converges towards the center. This convergence area is marked as the convergence zone location. The average coordinates of the nodes that meet the conditions are calculated as the center coordinates of the convergence zone, and then projected onto the valley centerline to calculate the lateral offset distance to obtain the lateral position data. Multiple lateral position data within the detection sampling period are arranged in chronological order to form a lateral position sequence of the convergence zone. Based on the axial coordinates of each sampling point in the along-path distribution sequence, the curvature radius data and the lateral position sequence of the convergence zone are spatially interpolated and aligned. The aligned cross-sectional width value, curvature radius value, and lateral position value are written into a unified data table structure in the same axial coordinate order to form a unified dataset containing geometric features and wind field features.

[0018] In one implementation, cross-sectional profile data is collected segment by segment along the axial direction of the valley at fixed intervals. The fixed intervals are preset based on the overall length of the valley and the degree of terrain variation. Larger sampling intervals can be used in sections with relatively gentle terrain undulations, while smaller sampling intervals are used in sections with drastic terrain variations to capture more detailed features.

[0019] Specifically, for each cross section, the horizontal distance between the left and right boundaries of the valley floor is extracted as the valley floor width. The valley floor boundary refers to the inflection point where the slope of the cross section profile curve changes significantly. This inflection point marks the boundary between the flat area of ​​the valley floor and the slopes on both sides. By identifying the inflection points on the left and right sides and calculating their horizontal projection distance, the valley floor width value of the cross section can be obtained. The width values ​​of each cross section are recorded sequentially according to the sampling sequence to form a distribution sequence along the route.

[0020] It should be noted that during the sampling process, the coordinates of the valley floor center point corresponding to each cross-section are extracted simultaneously. The center point is located at the midpoint of the line connecting the left and right boundaries. Its coordinate values ​​include the longitudinal position along the valley axis and the transverse position perpendicular to the axis. Based on the coordinates of three adjacent center points, a unique arc passing through these three points is fitted using a geometric method. The radius of this arc is the curvature radius data of the center line of this segment. The smaller the curvature radius, the greater the curvature of the valley segment. The larger the curvature radius, the straighter the valley segment tends to be.

[0021] In one embodiment, multiple wind speed and direction observation nodes are set up in the near-surface altitude layer within the valley. The near-surface altitude layer refers to the atmospheric layer within a certain height range above the ground surface. The airflow movement within this altitude range is directly affected by the topography. The observation nodes are distributed in a grid pattern or symmetrically along both sides of the valley floor centerline. After collecting the wind direction vector data of each node, the angle between the wind direction vectors of adjacent nodes is calculated.

[0022] For example, if the angle between the wind direction vectors of two adjacent observation nodes exceeds a preset convergence angle threshold, it is determined that the area has airflow convergence characteristics. The convergence angle threshold is predetermined based on the typical angle range when convergence occurs in the historical observation data of the valley. The convergence area that meets the judgment condition is marked as the location of the convergence zone, and the lateral offset distance of the convergence zone location relative to the valley floor centerline is measured to obtain the lateral position data. Further, the curvature radius data and the lateral position sequence of the convergence zone are spatially interpolated and aligned according to the axial coordinates of each sampling point in the distribution sequence. Since the spatial positions of the terrain profile sampling points and the wind field detection nodes are not completely coincident, the data from different sources are uniformly mapped to the same axial coordinate system through linear interpolation. The aligned cross-sectional width value, curvature radius value, and lateral position value are written into a unified data table structure in the same axial coordinate order to form a unified dataset containing geometric features and wind field features. This dataset provides basic data support for subsequent identification of sharp bends and analysis of the lateral offset characteristics of the convergence zone.

[0023] Step S102: Calculate the ratio of radius of curvature to valley width based on the radius of curvature data and valley width data in the unified dataset, identify sharp bends where the ratio exceeds a preset threshold, and determine potential asymmetric convergence area identifiers.

[0024] The curvature radius data and corresponding valley width data are read point by point from the unified dataset. For each sampling point, the ratio of the curvature radius value to the valley width value is calculated. If the ratio is lower than a preset threshold, the segment where the sampling point is located is determined to be a sharp bend. The sampling point number that meets the sharp bend determination criteria and its corresponding axial coordinate are marked as a sharp bend segment identifier. At the same time, the outer bending direction of the sharp bend segment is determined based on the direction of change of the centerline coordinates of adjacent sampling points. Based on the sharp bend segment identifier, the lateral position data of the convergence zone of the corresponding segment is extracted from the unified dataset. The lateral offset direction of the convergence zone in the segment relative to the valley bottom centerline is compared with the outer bending direction. If the lateral offset direction is consistent with the outer bending direction, it is determined that the sharp bend segment has potential asymmetric convergence characteristics, and a potential asymmetric convergence area identifier is generated.

[0025] In one implementation, curvature radius data and valley cross-sectional width data are read point by point from a unified dataset. For each sampling point, the curvature radius value is divided by the valley width value to obtain the ratio value of that point. The ratio value reflects the relative relationship between the curvature of the valley and the width of the passage at that location.

[0026] Specifically, when the ratio value is lower than a preset threshold, it indicates that the radius of curvature of the section where the sampling point is located is relatively small compared to the valley width, meaning that the curvature of the section is relatively severe. In this case, the sampling point is identified as part of a sharp bend, and its number and axial coordinates are recorded to form a sharp bend section identifier. At the same time, the direction of the outer curve is determined based on the changing trend of the centerline coordinates of adjacent sampling points within the sharp bend section, which is the opposite direction of the centerline concave inward. Further, the lateral position data of the convergence zone of the corresponding section is extracted based on the sharp bend section identifier. The lateral offset direction of the convergence zone relative to the valley bottom centerline is compared with the direction of the outer curve. If the two directions are consistent, it indicates that the airflow convergence zone within the sharp bend section exhibits a characteristic of converging towards the outer side of the bend, thereby generating a potential asymmetric convergence area identifier.

[0027] Step S103: Based on the lateral position data of the near-surface wind field convergence zone in the valley, the lateral position distribution of the near-surface convergence zone in the sharp bend section is analyzed, and the lateral position distribution is matched with the potential asymmetric convergence area identifier to identify the location of the convergence center where the airflow separation causes the convergence zone to gather towards the outside of the bend.

[0028] Based on the potential asymmetric convergence area identifier, lateral position data of the convergence zone for the emergency bend section are filtered from a unified dataset. The lateral offset distance of the convergence zone relative to the valley centerline is extracted point-by-point along the axial coordinate of the bend section. These lateral offset distances are arranged in axial coordinate order to form a lateral position distribution sequence within the bend section. The lateral offset distance of each point in the lateral position distribution sequence is compared point-by-point with the outer direction of the bend. If the lateral offset direction of a sampling point is consistent with the outer direction of the bend and the lateral offset distance exceeds a preset offset threshold, it is determined that the sampling point exhibits outer-side aggregation characteristics caused by airflow separation. Sampling points meeting this condition are marked as outer-side aggregation points. Based on the axial coordinate distribution of the outer-side aggregation points within the bend section, the number of outer-side aggregation points within each window is counted along the axial direction with a fixed window length. The area covered by the window with the most outer-side aggregation points is taken as the convergence zone aggregation area. The average lateral coordinate and the median axial coordinate of each outer-side aggregation point within this aggregation area are taken as the aggregation center coordinates, thus obtaining the aggregation center location where airflow separation causes the convergence zone to aggregate towards the outer side of the bend.

[0029] In one implementation, lateral position data of the convergence zone for emergency bend sections are filtered from a unified dataset based on potential asymmetric convergence zone identifiers. The filtering process is based on the sampling point number and axial coordinate range recorded in the sharp bend section identifiers, and extracts all data points falling within the range from the lateral position sequence of the convergence zone.

[0030] Specifically, the lateral offset distance of the convergence zone relative to the valley floor centerline is extracted point by point along the axial coordinate of the sharp bend section. This lateral offset distance refers to the vertical distance between the actual position of the convergence zone and the valley floor centerline; positive values ​​indicate a deviation towards the outer side of the bend, and negative values ​​indicate a deviation towards the inner side. The lateral offset distances of each sampling point are arranged in axial coordinate order to form a lateral position distribution sequence. This sequence reflects the lateral position variation pattern of the convergence zone along the valley direction within the sharp bend section. Further, for each sampling point in the lateral position distribution sequence, its lateral offset direction is compared with the outer direction of the bend in that sharp bend section. If the lateral offset direction points outward and the lateral offset distance exceeds a preset offset threshold, the sampling point is determined to exhibit outer-side aggregation characteristics caused by airflow separation. The offset threshold is preset according to a certain proportion of the valley width. Sampling points meeting the above conditions are marked as outer-side aggregation points, and their axial and lateral coordinates are recorded.

[0031] It should be noted that the distribution of the outer cluster points in the sharp bend section is not uniform and continuous, but rather exhibits a localized dense distribution. The outer cluster points are statistically analyzed along the axial direction using a fixed window length, which is predetermined based on the axial span and sampling interval of the sharp bend section. The number of outer cluster points contained in each window position is counted, and the axial range covered by the window with the largest number of outer cluster points is determined as the convergence zone cluster area.

[0032] In one embodiment, the arithmetic mean of the lateral coordinates of each outer gathering point within the gathering area is taken as the lateral coordinate of the gathering center, and the median of the axial range of the gathering area is taken as the axial coordinate of the gathering center. This yields the coordinates of the gathering center where airflow separation causes the convergence zone to gather towards the outside of the curve. These coordinates indicate the spatial location where the lateral shift of the convergence zone is most concentrated within the sharp curve section.

[0033] Step S104: Based on the location of the convergence center where the convergence zone gathers towards the outside of the curve, the lateral positioning centerline of the thunderstorm landing area is moved from the valley centerline to the location of the convergence center, generating the corrected landing area positioning centerline.

[0034] Based on the axial and lateral coordinates of the cluster center location, a lateral offset vector is established within the sharp bend section, pointing from the valley floor centerline to the cluster center location. This lateral offset vector comprises two components: offset direction and offset distance. This lateral offset vector serves as the basis for migrating the landing area positioning centerline. For the sharp bend section covered by the lateral offset vector, the lateral positioning centerline of the thunderstorm landing area is shifted from the original valley floor centerline along the offset direction to the lateral position where the cluster center is located. The axial coordinates and corresponding lateral coordinates of each sampling point within the sharp bend section are recorded, resulting in the corrected landing area positioning centerline.

[0035] In one embodiment, a lateral offset vector is established based on the axial and lateral coordinates of the aggregation center position. The starting point of the lateral offset vector is located on the valley center line at the same position as the axial coordinate of the aggregation center, and the ending point is located at the aggregation center position. The direction of the vector is the offset direction, and the length of the vector is the offset distance.

[0036] Specifically, the lateral offset vector characterizes the degree of deviation of the actual convergence zone position relative to the valley centerline within the sharp bend section. The offset direction points outward from the bend, and the offset distance reflects the magnitude of the lateral migration of the convergence zone caused by airflow separation. This vector serves as the basis for the relocation of the positioning centerline. Further, for the sharp bend section covered by the lateral offset vector, the lateral positioning centerline of the thunderstorm landing area is shifted from its original valley centerline position along the offset direction. The shift distance is equal to the offset distance of the lateral offset vector, and the endpoint of the shift is located at the lateral position of the convergence center. The axial coordinates and corresponding lateral coordinates of each sampling point within the sharp bend section are recorded after the shift, resulting in the corrected positioning centerline of the landing area. This corrected positioning centerline reflects the influence of the asymmetrical distribution of the convergence zone within the sharp bend section on the lateral position of the thunderstorm landing area.

[0037] Step S105: Based on the corrected landing area positioning axis, assess the convergence zone distribution pattern of the large-scale weather system forecast location segment. In the asymmetric convergence segment, move the forecast location along the outer direction of the curve to the location of the landing area positioning axis to obtain the adjusted thunderstorm landing area coordinates.

[0038] Obtain the original coordinates of the thunderstorm location from the large-scale weather forecast. Locate the location within the valley's axial coordinate sequence based on the axial components of the original coordinates, determining the valley segment number where the forecast location is situated. Extract the corresponding lateral coordinate value and potential asymmetric convergence area identifier from the corrected location axis. Evaluate the convergence zone distribution pattern based on the corrected location axis. The specific evaluation process is as follows: Input the original lateral coordinate X0 and the lateral coordinate Xm of the central axis. Calculate the deviation D, which is equal to the absolute value of X0 minus Xm. If D is greater than a threshold of 5, it is determined that the segment exhibits asymmetric convergence characteristics, and the convergence zone distribution pattern deviates from the valley floor centerline. Use the deviation D as the migration amount of the forecast location along the outer direction of the curve. Based on this migration amount, translate the lateral coordinates of the forecast location from the original location along the outer direction of the curve to the lateral position of the location axis, keeping the axial coordinates of the forecast location unchanged. Combine and record the translated axial and lateral coordinates to obtain the adjusted thunderstorm location coordinates.

[0039] In one implementation, the original location coordinates of the thunderstorm area output by the large-scale weather forecast are obtained. The original location coordinates include an axial component and a lateral component. The axial component represents the position of the forecast area along the valley direction, and the lateral component represents the lateral position of the forecast area relative to the center line of the valley floor.

[0040] Specifically, the valley axial coordinate sequence is searched and located based on the axial components of the original location coordinates to determine the valley segment number where the forecast location is located. This valley segment number corresponds to the sampling point number in the unified dataset. The lateral coordinate value corresponding to this segment is extracted from the corrected landing area positioning axis. Simultaneously, the potential asymmetric convergence area identifier for this segment is read. This identifier indicates whether the segment exhibits asymmetric convergence zone distribution characteristics caused by sharp bends. Further, if the potential asymmetric convergence area identifier indicates that the segment has asymmetric convergence characteristics, it is determined that the original location coordinates output by the large-scale weather forecast did not consider the influence of local topography on the lateral position of the convergence zone. In this case, the difference between the original lateral coordinates of the forecast location and the lateral coordinates of the corrected landing area positioning axis is calculated as the lateral deviation value. This lateral deviation value represents the migration of the forecast location along the outer direction of the bend, reflecting the lateral distance between the original forecast and the actual convergence zone convergence location.

[0041] It should be noted that, based on the migration amount, the lateral coordinates of the predicted location are shifted along the outside of the curve. The shift distance is equal to the absolute value of the lateral deviation. The shift direction points to the lateral position of the corrected landing area positioning centerline. During the shift, the axial coordinates of the predicted location remain unchanged. The shifted axial coordinates and lateral coordinates are combined and recorded to form the adjusted thunderstorm landing area position coordinates.

[0042] In one embodiment, if the potential asymmetric convergence area marker of the segment where the forecast location is located indicates that there is no asymmetric convergence feature in the segment, the original coordinates of the forecast location are kept unchanged, and the original location coordinates are directly output as the adjusted thunderstorm landing area location coordinates. This processing method makes the location adjustment only correct for the asymmetric convergence situation in the sharp bend segment.

[0043] Step S106: By combining the adjusted coordinates of the thunderstorm landing area with the valley width distribution data, a landing area positioning layer is generated in the disaster mapping, and the boundary of the high curvature area is smoothed to form a thunderstorm disaster landing area distribution map.

[0044] Based on the axial and lateral components of the adjusted thunderstorm landing area location coordinates, the valley width value corresponding to the axial position is extracted from the valley width distribution data. Extending laterally along the valley width value to both sides with the landing area location coordinates as the center, the lateral coverage range of the landing area at that location is determined. The lateral coverage ranges of the landing areas at each sampling point are connected along the axial direction to form the initial boundary contour of the landing area positioning layer. For sharp bends in the initial boundary contour where the ratio of the radius of curvature to the valley width is lower than a preset threshold, where the radius of curvature r is calculated by fitting a three-point circle at the boundary points, the valley width w is the valley width at the corresponding location, the ratio is defined as r / w, and the preset threshold value is 0.3, a simple moving average filtering method is used to process the boundary point coordinates of the sharp bend. The input is the lateral coordinate sequence of adjacent boundary points. Taking a window length of 5 as an example, the average value of the sequence is taken along the axial direction, and the output is a smoothed lateral coordinate sequence. The smoothed boundary point coordinates replace the original initial boundary points of the sharp bend, generating a smoothed landing area positioning layer boundary. Based on the smoothed boundary of the landing area positioning layer, the boundary points are connected in sequence to form a closed polygonal region. This closed polygonal region is superimposed on the geographical base map of the disaster mapping as the landing area positioning layer. Potential asymmetric convergence areas in each segment within the landing area positioning layer are marked to form a thunderstorm disaster landing area distribution map.

[0045] In one implementation, the valley width value at the corresponding location is retrieved from the valley width distribution data based on the axial component of the adjusted thunderstorm location coordinates. The lateral coverage range of the location is determined by extending the valley width value to both sides along the lateral direction with the location coordinates of the location as the center, according to a preset ratio of the valley width value. The preset ratio is pre-set based on the statistical relationship between the actual lateral coverage of the location and the valley width in historical thunderstorm events.

[0046] Specifically, the lateral coverage areas of each sampling point are sequentially connected along the axial direction. The left boundary points are connected to form the left boundary line, and the right boundary points are connected to form the right boundary line. The left and right boundary lines together constitute the initial boundary contour of the positioning layer. This initial boundary contour may exhibit an irregular sawtooth shape in sharp bend sections. Furthermore, for sharp bend sections where the ratio of the radius of curvature to the valley width in the initial boundary contour is lower than a preset threshold, a moving average filtering method is used to smooth the boundary point coordinates. The moving average filtering method sets a fixed-length sliding window along the axial direction, and the window contains several adjacent boundary points. The arithmetic mean of the lateral coordinates of each boundary point within the window is calculated as the smoothed lateral coordinate of the window's center point. The window slides point by point along the axial direction until it covers the entire sharp bend section, resulting in a smoothed sequence of boundary point coordinates. This smoothing process eliminates local abrupt changes in the boundary of the sharp bend section.

[0047] It should be noted that after the smoothed boundary point coordinates replace the initial boundary points of the original sharp bend section, all boundary points are connected in sequence to form a closed polygonal region. This closed polygonal region is superimposed on the geographic base map of the disaster mapping as the landing area positioning layer. At the same time, the potential asymmetric convergence area of ​​each section is marked in the landing area positioning layer to form a thunderstorm disaster landing area distribution map. This distribution map intuitively presents the spatial range of the thunderstorm landing area after considering the influence of valley curvature.

[0048] Step S107: Analyze the asymmetric convergence characteristics of the high curvature segment based on the distribution map of thunderstorm disaster areas, compare the characteristics with the lateral position data of the convergence zone, evaluate the degree of fit between the two, and gradually update the positioning centerline of the disaster area to obtain an optimized version of the distribution map.

[0049] Extract the boundary coordinate sequence of the landing area positioning layer in the high curvature section from the thunderstorm disaster landing area distribution map. Calculate the lateral offset distribution of this boundary coordinate sequence relative to the corrected landing area positioning axis. Associate this lateral offset distribution with potential asymmetric convergence area markers to identify asymmetric convergence feature areas where the landing area boundary in the high curvature section shifts outwards from the curve. Record the axial range and lateral offset magnitude of these feature areas. Based on the axial range of the asymmetric convergence feature areas, extract the measured lateral position sequence of the corresponding section from the lateral position data of the convergence zone. Compare this measured lateral position sequence with the lateral coordinates of the landing area positioning axis at the same axial position point by point, calculate the lateral distance difference between the two, and take the average value after taking the absolute value of the lateral distance difference to obtain the fit value of the asymmetric convergence feature area. If the fit value exceeds a preset fit threshold, it is determined that there is a deviation between the landing area positioning centerline and the measured lateral position of the convergence zone. The centerline adjustment direction is determined based on the positive or negative direction of the lateral distance difference. The lateral coordinates of the landing area positioning centerline in this segment are translated and corrected according to the average value of the lateral distance difference to obtain the adjusted landing area positioning centerline coordinates. Based on the adjusted landing area positioning centerline coordinates, the landing area boundary range determination and boundary smoothing are re-executed to generate an updated landing area positioning layer boundary. The updated landing area positioning layer is superimposed on the geographic base map to replace the original landing area positioning layer, resulting in an optimized version of the thunderstorm disaster landing area distribution map.

[0050] In one implementation, the boundary coordinate sequence of the high curvature section of the landing area is extracted from the thunderstorm disaster landing area distribution map. The high curvature section refers to the sharp bend section where the ratio of the radius of curvature to the valley width is lower than a preset threshold. For each boundary point in the boundary coordinate sequence, its lateral offset distance relative to the corrected landing area positioning center axis is calculated. The lateral offset distances of each boundary point are arranged in axial coordinate order to form a lateral offset distribution sequence.

[0051] Specifically, the lateral offset distribution sequence is correlated and matched with the potential asymmetric convergence area markers for that section. If the lateral offset distribution within a certain axial range exhibits a continuous shift towards the outside of the curve, and the potential asymmetric convergence area markers within that range indicate the presence of asymmetric convergence, then that axial range is identified as an asymmetric convergence feature area. The starting axial coordinates, ending axial coordinates, and the average lateral offset distance within the feature area are recorded as the lateral offset amplitude. Further, based on the axial range of the asymmetric convergence feature area, the measured lateral position sequence of the corresponding section is retrieved from the lateral position data of the convergence zone. This measured lateral position sequence originates from the lateral offset distance records of the convergence zone relative to the valley centerline collected within that section. This sequence reflects the actual lateral distribution position of the airflow convergence zone within that sharp curve section.

[0052] It should be noted that the calculation of the fit value adopts a point-by-point comparison method. For each axial coordinate position in the asymmetric convergence feature area, the lateral coordinate value of that position in the measured lateral position sequence of the convergence zone is read and the lateral coordinate value of the landing area positioning axis at that position is calculated. The difference in lateral distance between the two is calculated. A positive difference in lateral distance indicates that the measured convergence zone position is located outside the curve of the landing area positioning axis, and a negative difference indicates that it is located inside the curve. The arithmetic mean of the absolute values ​​of the lateral distance differences of all axial positions in the area is taken to obtain the fit value of the asymmetric convergence feature area. The smaller the fit value, the smaller the deviation between the landing area positioning axis and the measured lateral position of the convergence zone.

[0053] In one embodiment, if the fit value exceeds a preset fit threshold, it is determined that there is a deviation to be corrected between the lateral position of the current landing area positioning axis and the measured convergence zone convergence position. In this case, the adjustment direction of the axis is determined based on the positive and negative distribution of the lateral distance difference within the area. If the lateral distance difference is predominantly positive, the axis is adjusted towards the outside of the curve; if it is predominantly negative, it is adjusted towards the inside of the curve. The adjustment range is taken as the average of the lateral distance differences. The lateral coordinates of each axial position of the landing area positioning axis within this section are translated and corrected according to the adjustment direction and adjustment range to obtain the adjusted landing area positioning axis coordinate sequence. If the fit value does not exceed the preset fit threshold, it is determined that there is no deviation to be corrected between the lateral position of the current landing area positioning axis and the measured convergence zone convergence position. In this case, the lateral coverage area of ​​the landing area is directly obtained.

[0054] For example, the adjusted landing area positioning center axis coordinate sequence is used as the reference input for determining the boundary range of the landing area in the new round. The operation of extending to both sides in the horizontal direction with the adjusted landing area positioning center axis as the center and according to the preset ratio of the valley width is re-executed to form the updated horizontal coverage range of the landing area. The updated horizontal coverage range of the landing area of ​​each sampling point is connected in the axial direction to form the updated initial boundary contour.

[0055] Understandably, the updated initial boundary profile still needs to be smoothed in the sharp bend section. The moving average filtering method is used to process the boundary point coordinates of the sharp bend section in the updated initial boundary profile. The average value of the lateral coordinates of adjacent boundary points is taken along the axial direction with a fixed window length to obtain the smoothed updated boundary point coordinate sequence. The smoothed updated boundary points replace the initial boundary points of the sharp bend section to form the updated landing area positioning layer boundary.

[0056] In one possible implementation, the boundary points of the updated landing area positioning layer are sequentially connected to form an updated closed polygonal region. The original landing area positioning layer is replaced with the updated closed polygonal region on the geographic base map of the disaster mapping, while retaining the potential asymmetric convergence area markers of each segment. This results in an optimized version of the thunderstorm disaster landing area distribution map. In this optimized version, the fit between the lateral position of the landing area positioning layer and the measured lateral position of the convergence zone is improved, and the landing area boundary more accurately reflects the asymmetric distribution characteristics of the convergence zone caused by airflow separation in the sharp bend section.

[0057] The above embodiments are merely one of the preferred embodiments of the present invention and should not be used to limit the scope of protection of the present invention. Any modifications or refinements made to the main design concept and spirit of the present invention that are not of substantial significance, but solve the same technical problem as the present invention, should be included within the scope of protection of the present invention.

Claims

1. A method for analyzing and mapping thunderstorm disaster areas, characterized in that, The method includes: Data on the cross-sectional width distribution along the valley, the centerline curvature radius, and the lateral location of the near-surface wind convergence zone within the valley were acquired and integrated into a unified dataset. The ratio of radius of curvature to valley width is calculated based on the radius of curvature data and valley width data in the unified dataset. Sharp bends with a ratio exceeding a preset threshold are identified, and potential asymmetric convergence areas are identified. Based on the lateral location data of the near-surface wind convergence zone in the valley, the lateral location distribution of the near-surface convergence zone in the sharp bend section is analyzed. This lateral location distribution is matched with the potential asymmetric convergence area marker to identify the location of the convergence center where airflow separation causes the convergence zone to gather towards the outside of the bend. Based on the location of the convergence center where the convergence zone gathers towards the outside of the bend, the lateral positioning axis of the thunderstorm landing area is moved from the valley centerline to the location of the convergence center, thus generating the corrected landing area positioning axis. Based on the revised landing area positioning axis, the distribution pattern of the convergence zone in the section where the large-scale weather system forecast location is located is evaluated. In the asymmetric convergence section, the forecast location is moved along the outer direction of the curve to the location of the landing area positioning axis, and the adjusted coordinates of the thunderstorm landing area location are obtained. By combining the adjusted coordinates of the thunderstorm landing area with the valley width distribution data, a landing area positioning layer is generated in the disaster mapping, and the boundary of the high curvature area is smoothed to form a thunderstorm disaster landing area distribution map. Based on the distribution map of thunderstorm disaster areas, the asymmetric convergence characteristics of the high curvature segment are analyzed, the degree of fit between this characteristic and the lateral position data of the convergence zone is evaluated, and an optimized version of the distribution map is generated.

2. The method for analyzing and mapping thunderstorm disaster areas according to claim 1, characterized in that, The acquisition of data on the cross-sectional width distribution along the valley, the centerline radius of curvature, and the lateral position of the near-surface wind convergence zone within the valley includes: Cross-sectional profile data were acquired segment by segment along the valley axis, and the horizontal distance between the left and right boundaries of the valley floor was extracted as the valley width data to form a distribution sequence along the valley. Obtain the coordinates of the valley floor center point, fit an arc through adjacent center points, and obtain the curvature radius data of the center line of the middle segment; Wind speed and direction observation nodes are set up to collect wind direction vectors and calculate the angle between wind direction vectors of adjacent nodes. If the angle exceeds a preset threshold, it is determined to be an airflow convergence area and marked as the location of the convergence zone.

3. The method for analyzing and mapping thunderstorm disaster areas according to claim 1, characterized in that, The step of calculating the ratio of radius of curvature to valley width based on the radius of curvature data and valley width data in a unified dataset, identifying sharp bends where the ratio exceeds a preset threshold, and determining potential asymmetric convergence area identifiers includes: Read the radius of curvature data and the corresponding valley width data from the unified dataset, calculate the ratio between the radius of curvature data and the corresponding valley width data, and if the ratio is lower than the preset threshold, it is determined to be a sharp bend section, and the sampling point number and axial coordinate are marked as the sharp bend section identifier. The outer direction of the bend is determined based on the change in the centerline coordinates of adjacent sampling points; Extract the lateral position data of the convergence zone within the sharp bend section, compare the lateral offset direction with the outer direction of the bend, and if they are consistent, determine that there is a potential asymmetric convergence feature, and generate the potential asymmetric convergence area identifier.

4. The method for analyzing and mapping thunderstorm disaster areas according to claim 1, characterized in that, The process involves analyzing the lateral location data of the near-surface wind convergence zone within the valley to determine its lateral distribution in the sharp bend section. This lateral distribution is then matched with potential asymmetric convergence area markers to identify the convergence center location where airflow separation causes the convergence zone to gather towards the outside of the bend. This includes: Based on the identification of the potential asymmetric convergence area, the lateral position data of the convergence zone in the sharp bend section are screened, and a position distribution sequence is formed by combining the lateral offset distance. Compare the offset direction with the outer direction of the bend point by point. If the offset direction is consistent with the outer direction of the bend and the offset distance exceeds the preset threshold, it is marked as an outer aggregation point. The number of aggregation points on the outer side is counted using a fixed window along the axial direction. The area covered by the window with the largest number of points is determined as the convergence zone aggregation area. The average value of the lateral coordinates and the median value of the axial coordinates within this area are calculated to obtain the location of the aggregation center.

5. The method for analyzing and mapping thunderstorm disaster areas according to claim 1, characterized in that, The step of shifting the lateral positioning axis of the thunderstorm landing area from the valley centerline to the location of the convergence center, based on the location of the convergence center where the convergence zone gathers towards the outside of the curve, to generate a corrected landing area positioning axis includes: Based on the axial and lateral coordinates of the aggregation center location, a lateral offset vector is established from the valley centerline to the aggregation center location, including the offset direction and offset distance; For the sharp bend section covered by the lateral offset vector, the lateral positioning centerline of the thunderstorm landing area is shifted from the original centerline along the offset direction to the lateral position where the gathering center is located. Record the axial and lateral coordinates of the positioning centerline at each sampling point after migration to obtain the corrected positioning centerline of the landing area.

6. The method for analyzing and mapping thunderstorm disaster areas according to claim 1, characterized in that, The process involves assessing the convergence zone distribution pattern of the large-scale weather system forecast location segment based on the corrected landing area positioning axis, and migrating the forecast location along the outer direction of the curve to the location of the landing area positioning axis in asymmetric convergence segments to obtain the adjusted thunderstorm landing area coordinates, including: Obtain the original location coordinates of the thunderstorm area in the large-scale weather forecast, locate the valley section number according to the axial component, and extract the lateral coordinate value of the corresponding section's location center axis. Calculate the deviation between the original horizontal coordinate and the central horizontal coordinate. If the deviation is greater than a preset threshold, it is determined that there is an asymmetric convergence feature. Using the deviation as the migration amount, the lateral coordinates of the predicted location are translated along the outer side of the curve to the location of the central axis of the landing area, while keeping the axial coordinates unchanged. The coordinates after translation are recorded to obtain the adjusted coordinates of the thunderstorm landing area.

7. The method for analyzing and mapping thunderstorm disaster areas according to claim 1, characterized in that, The process involves combining the adjusted coordinates of the thunderstorm impact area with the valley width distribution data to generate an impact area positioning layer in the disaster mapping, smoothing the boundaries of areas with high curvature, and forming a thunderstorm disaster impact area distribution map, including: The width value of the corresponding axial position is extracted from the valley width distribution data. The adjusted thunderstorm landing area coordinates are used as the center and extended to both sides according to a preset ratio to determine the lateral coverage range. The connection is then used to form the initial boundary outline of the landing area positioning layer. For sharp bends where the ratio of radius of curvature to valley width is below a preset threshold, a moving average filter is used to process the boundary point coordinates to generate a smoothed boundary. The smoothed boundary connections form closed polygonal regions, which are then superimposed onto the geographic base map as the location layer of the affected area. Asymmetric convergence area markers are added to form the distribution map of the thunderstorm disaster affected area.

8. The method for analyzing and mapping thunderstorm disaster areas according to claim 1, characterized in that, The process involves analyzing the asymmetric convergence characteristics of the high curvature segment based on the thunderstorm disaster area distribution map, evaluating the degree of fit between these characteristics and the lateral position data of the convergence zone, and generating an optimized distribution map version, including: Extract the high-curvature boundary coordinate sequence from the thunderstorm disaster area distribution map, calculate the lateral offset distribution relative to the positioning axis of the area, and identify asymmetric convergence feature regions offset to the outside of the curve; Extract the measured position sequence of the corresponding segment from the lateral position data of the convergence zone, and calculate the average value of the lateral distance difference as the degree of fit by comparing point by point; If the fit value exceeds the preset threshold, the landing area positioning centerline is corrected by shifting the difference direction and average value, the boundary range is redefined and smoothing is performed, an updated landing area positioning layer is generated, and it is superimposed on the geographic base map to obtain an optimized distribution map version.

Images