A data processing method and system for weather radar

Abstract

The present invention discloses a data processing method and system for a weather radar, the method comprising the steps of: obtaining an atmospheric temperature profile collected by a microwave radiometer and performing quality control, extracting a 0°C height and a -20°C height from the atmospheric temperature profile after quality control; obtaining weather radar data of multiple radars and performing quality control, fusing the weather radar data after quality control of multiple radars, obtaining three-dimensional grid point data arranged in a unified height, latitude and longitude, identifying storm components and storm cells based on the three-dimensional grid point data; calculating a storm cell hail characteristic height according to the 0°C height and the centroid height of the storm components contained in the storm cell; calculating the height difference between the storm cell hail characteristic height and the 0°C height, and calculating the probability of hail occurring in the hail cell according to the height difference. The present invention can reduce the false alarm rate of hail monitoring of weather radar and improve the accuracy of hail monitoring.

Description

Technical Field

[0001] The present invention belongs to the technical field of weather radar application, and more specifically, relates to a data processing method and system for a weather radar. Background Art

[0002] As one of the main technical means of detecting precipitation systems, weather radar plays an important role in meteorological monitoring, early warning, and meteorological disaster prevention and mitigation. At present, the new generation radar weather network (CINRAD) is a powerful tool for monitoring hail cloud activities. In order to accurately identify and predict the occurrence and development of hail clouds, its hail detection algorithm mainly inherits the hail algorithm HDA (Hail Detection Algorithm) proposed in 1998 based on the WSR-88D weather radar. Years of business applications have shown that although the CINRAD hail monitoring algorithm has a good hit rate for hail weather processes, its high false alarm rate has caused it to not play its due role in business. Therefore, there is an urgent need for a method to reduce the false alarm rate of CINRAD hail monitoring. Summary of the invention

[0003] In view of the above defects or improvement needs of the prior art, the present invention provides a data processing method and system for a weather radar, which can reduce the false alarm rate of hail monitoring of the weather radar.

[0004] To achieve the above object, according to one aspect of the present invention, a weather radar data processing method is provided, comprising the steps of:

[0005] Obtain the atmospheric temperature profile collected by the microwave radiometer and perform quality control, and extract the 0°C height and -20°C height from the atmospheric temperature profile after quality control;

[0006] Acquire weather radar data from multiple radars and perform quality control, fuse the weather radar data after quality control from multiple radars to obtain three-dimensional grid data with uniform height, latitude and longitude, and identify storm components and storm cells based on the three-dimensional grid data;

[0007] Calculating the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm component contained in the storm cell;

[0008] The height difference between the characteristic height of hail in a storm cell and the 0°C height is calculated, and the probability of hail occurring in the hail cell is calculated based on the height difference.

[0009] Furthermore, the method further comprises the steps of: calculating the severe hail index and severe hail probability index of the storm cell;

[0010] The calculation of the severe hail index and severe hail probability index of a storm cell comprises the following sub-steps:

[0011] Based on the reflectivity factor of the storm component in the storm cell, a first weight coefficient for converting each storm component in the storm cell into hail energy is defined;

[0012] Based on the 0°C height and the -20°C height, define a second weight coefficient of the influence of the centroid height of each storm component in the storm cell on the growth of hail;

[0013] Calculating the hail kinetic energy flux of each storm component in the storm cell according to the first weight coefficient and the reflectivity factor of the storm component in the storm cell;

[0014] The severe hail index of each storm component in the storm cell is calculated according to the hail kinetic energy flux corresponding to the maximum reflectivity factor of the storm component in the storm cell, the second weight coefficient and the vertical thickness of the storm component;

[0015] The severe hail index of a storm cell is calculated based on the severe hail index of each storm component in the storm cell;

[0016] The probability of severe hail in a storm cell is calculated based on the severe hail index of the storm cell.

[0017] Furthermore, the step of obtaining the atmospheric temperature profile collected by the microwave radiometer and performing quality control comprises the following steps:

[0018] Get the atmospheric temperature profile collected by the microwave radiometer, denoted as , , express The temperature of the i-th sampling point in , 1≤i≤N, N represents the total number of sampling points;

[0019] Get the L-band sounding temperature data of the corresponding time, recorded as , , express The temperature of the i-th sampling point in ;

[0020] Calculate the temperature observation deviation of the microwave radiometer, denoted as , , express The temperature deviation of the i-th sampling point in X i Normalized to , , express The mean of express If the standard deviation of , then the sampling point Value removed.

[0021] Furthermore, the quality control of the weather radar data of each radar includes the steps of:

[0022] Correction of weather radar data;

[0023] The reflectivity factor, speed, differential reflectivity factor, co-correlation coefficient, reflectivity factor texture features, differential reflectivity texture features and co-correlation coefficient texture features are extracted from the corrected weather radar data and input into the deep learning model to obtain the data quality judgment result of each sampling point. According to the data quality judgment result, the sampling point data with low quality are eliminated from the corrected weather radar data, and the retained sampling point data are used as the weather radar data after quality control.

[0024] Furthermore, the identifying of storm components and storm cells based on three-dimensional grid data comprises the steps of:

[0025] Predict radar echo movement based on the three-dimensional grid reflectivity field to obtain extrapolated three-dimensional grid data;

[0026] Identify storm components and storm cells based on extrapolated three-dimensional grid data.

[0027] Further, the step of calculating the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm components contained in the storm cell comprises the following steps:

[0028] S301, using the centroid heights of adjacent storm components, calculate the top height and bottom height of the storm component contained in the storm cell. The top height of the mth storm component in the storm cell is recorded as , the bottom height of the mth storm component in the storm cell is recorded as ;

[0029] S302, record the 0℃ height as , the vertical thickness of the mth storm component in the storm cell is recorded as ,like > , = - ,otherwise, = - ;

[0030] S303, sequentially calculating the hail characteristic development height HHAIL of each storm cell. The calculation of the hail characteristic development height HHAIL of a single storm cell specifically includes the following sub-steps:

[0031] S3031, taking the reflectivity of 45dBZ as the judgment threshold, recorded as REF_THD, if the maximum reflectivity factor of the highest-level storm component is greater than REF_THD, the top height of the highest-level storm component is taken as the hail characteristic development height HHAIL of the storm cell, and the calculation ends; if the maximum reflectivity factor of the highest-level storm component is not greater than REF_THD, according to the weather radar echo height from high to low, the next layer of storm components of the highest-level storm component is taken as the current layer of storm components;

[0032] S3032: Compare whether the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ. If the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ, use the maximum reflectivity factor REF_MAX of the storm component of the current layer, the centroid height C_HT of the storm component of the current layer, the reflectivity factor REF_MAX_PRV of the next layer of the current layer, and the centroid height C_HT_PRV of the next layer of the storm component of the current layer to calculate the height corresponding to 45dBZ by extrapolation as the hail characteristic development height HHAIL of the storm cell. The calculation formula is:

[0033] ;

[0034] The calculation is finished;

[0035] If the maximum reflectivity factor of the current layer storm component is not greater than 40dBZ, the next layer of the current layer storm component is taken as the new current layer storm component according to the weather radar echo height from high to low, and step S3032 is repeated.

[0036] Furthermore, the calculation formula for the hail kinetic energy flux of each storm component in the storm cell is:

[0037] ;

[0038] represents the hail kinetic energy flux of the mth storm component in the storm cell, represents the first weight coefficient of the mth storm component, represents the reflectivity factor of the mth storm component, and is a constant coefficient;

[0039] The calculation formula for the severe hail index of each storm component in a storm cell is:

[0040] ;

[0041] It represents the severe hail index of the mth storm component in a storm cell. represents the vertical thickness of the mth storm component, Represents the second weight coefficient of the mth storm component.

[0042] Furthermore, the calculation formula for the severe hail index of a storm cell is:

[0043] ;

[0044] M represents the total number of storm components in a storm cell, Indicates the severe hail index of a storm cell.

[0045] According to another aspect of the present invention, there is provided a data processing system for a weather radar, comprising:

[0046] The atmospheric temperature profile acquisition and extraction module is used to acquire the atmospheric temperature profile collected by the microwave radiometer and perform quality control, and extract the 0℃ height and -20℃ height from the atmospheric temperature profile after quality control;

[0047] A storm cell identification module is used to obtain weather radar data from multiple radars and perform quality control, fuse the weather radar data after quality control from multiple radars, obtain three-dimensional grid point data with uniform height, latitude and longitude, and identify storm components and storm cells based on the three-dimensional grid point data;

[0048] A storm cell hail characteristic height calculation module, used to calculate the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm component contained in the storm cell;

[0049] The hail probability index calculation module is used to calculate the height difference between the characteristic height of the hail in the storm cell and the 0°C height, and calculate the probability of hail occurring in the hail cell according to the height difference.

[0050] Furthermore, it also includes a severe hail prediction module;

[0051] The severe hail prediction module comprises:

[0052] A first weight coefficient conversion module, for defining a first weight coefficient for converting each storm component in the storm cell into hail energy based on a reflectivity factor of the storm component in the storm cell;

[0053] A second weight coefficient conversion module is used to define a second weight coefficient of the influence of the centroid height of each storm component in the storm cell on the growth of hail based on the 0°C height and the -20°C height;

[0054] A hail kinetic energy flux calculation module, used to calculate the hail kinetic energy flux of each storm component in the storm cell according to the first weight coefficient and the reflectivity factor of the storm component in the storm cell;

[0055] A storm component severe hail index calculation module is used to calculate the severe hail index of each storm component in the storm cell according to the hail kinetic energy flux corresponding to the maximum reflectivity factor of the storm component in the storm cell, the second weight coefficient, and the vertical thickness of the storm component;

[0056] The storm cell strong hail index calculation module is used to calculate the strong hail index and strong hail probability of the storm cell according to the strong hail index of each storm component in the storm cell.

[0057] In general, the above technical solutions conceived by the present invention have beneficial effects compared with the prior art:

[0058] One of the innovations of the present invention is that hail monitoring uses multi-radar data for joint monitoring, and its advantage is that compared with the traditional hail monitoring method, this method can solve the impact of radar blind spots and elevation gap coefficients on hail monitoring, so compared with the traditional hail monitoring algorithm, this method has higher accuracy. The second innovation of the present invention is that a microwave radiometer is used to detect the atmospheric temperature profile, and its advantage is that compared with the traditional L-band sounding, this method can provide real-time high-temporal resolution vertical profiles of atmospheric temperature, humidity and water vapor density. The third innovation of the present invention is that hail monitoring is performed based on extrapolated three-dimensional grid data types, which can further improve the accuracy of hail monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of a data processing method for a weather radar according to an embodiment of the present invention;

[0060] Figure 2 It is a schematic diagram of the working principle of the data processing method of the weather radar according to an embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0062] In the description of the embodiments of the present application, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. "Multiple" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0063] The terms "including" and "having" and any variations thereof in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product or equipment comprising a series of steps or modules is not necessarily limited to those steps or modules clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or equipment.

[0064] The naming or numbering of the steps in the embodiments of the present invention does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The execution order of the named or numbered process steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0065] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0066] The present invention provides a data processing method and system for a weather radar, which are respectively described below.

[0067] like Figure 1 and Figure 2 As shown, a data processing method for a weather radar according to an embodiment of the present invention comprises the following steps:

[0068] S1, obtain the atmospheric temperature profile collected by the microwave radiometer and perform quality control, and extract the 0℃ height and -20℃ height from the atmospheric temperature profile after quality control.

[0069] 0℃ altitude refers to the altitude where the 0℃ isotherm is located.

[0070] -20℃ altitude refers to the altitude where the -20℃ isotherm is located.

[0071] The formation and development of hail weather is closely related to atmospheric environmental parameters. Usually, hail growth only occurs below 0°C, and the growth of large hail mainly occurs at temperatures of -20°C or lower. At present, the CINRAD hail monitoring algorithm does not introduce the real-time changing atmospheric environmental temperature profile, which is an important reason affecting the false alarm rate of CINRAD hail monitoring products.

[0072] Microwave radiometer is a passive remote sensing detection equipment, which can measure the atmospheric microwave radiation brightness temperature from the surface to 10km in real time, continuously and unattended, and invert the atmospheric temperature and humidity profile data, which makes up for the lack of discontinuous time of existing high-altitude observation, and has the ability to accurately monitor small and medium-scale weather systems and severe convective weather. In recent years, vertical observation systems composed of microwave radiometers, wind profiler radars, cloud meters, laser aerosol radars and GNSS water vapor meters have been widely deployed. Among them, the monitoring of microwave radiometers can provide real-time and continuous vertical profiles of atmospheric temperature, humidity and water vapor density.

[0073] Therefore, in the embodiment of the present invention, by acquiring the atmospheric environment parameters collected in real time by the microwave radiometer, especially the atmospheric temperature profile information, the formation environment of hail weather can be timely and effectively determined, and the false alarm rate of hail monitoring of the weather radar can be reduced.

[0074] Furthermore, L-band sounding data are introduced to perform quality control on the atmospheric temperature profile of the microwave radiometer, which specifically includes the following steps:

[0075] Get the atmospheric temperature profile collected by the microwave radiometer, denoted as , , express The temperature of the i-th sampling point in , 1≤i≤N, N represents the total number of sampling points;

[0076] Get the L-band sounding temperature data of the corresponding time, recorded as , , express The temperature of the i-th sampling point in ;

[0077] Calculate the temperature observation deviation of the microwave radiometer, denoted as , , express The temperature deviation of the i-th sampling point in X i Normalized to , , express The mean of express If the standard deviation of , then the sampling point Value removed.

[0078] S2, obtains weather radar data from multiple radars and performs quality control, fuses the weather radar data after quality control from multiple radars, obtains three-dimensional grid data with uniform altitude, latitude and longitude arrangement, and identifies storm components and storm cells based on the three-dimensional grid data.

[0079] Furthermore, the identifying of storm components and storm cells based on three-dimensional grid data comprises the steps of:

[0080] Predict radar echo movement based on the three-dimensional grid reflectivity field to obtain extrapolated three-dimensional grid data;

[0081] Identify storm components and storm cells based on extrapolated three-dimensional grid data.

[0082] The following describes the preferred implementation of weather radar data quality control, data fusion, radar echo motion prediction, storm component and storm cell identification.

[0083] (1) Weather radar data quality control

[0084] The quality control of the weather radar data of each radar at each time is performed separately, and the same method can be used for the quality control of the weather radar data of each radar at each time.

[0085] The following specifically describes the quality control of a radar at one time as an example, including the following steps:

[0086] ① Correct the weather radar data, including:

[0087] (a) Correlation coefficient polarization correction. The signal-to-noise ratio (SNR) parameter of the radar is obtained, and the correlation coefficient polarization of the weather radar is corrected according to the SNR parameter. The specific correction algorithm can adopt any existing algorithm.

[0088] (b) Reflectivity factor attenuation correction. The self-consistency between the polarization quantities of weather radar is used to correct the reflectivity factor. The specific correction algorithm can adopt any existing algorithm.

[0089] ② Use the deep learning model to perform quality control on the corrected weather radar data. First, extract the reflectivity factor Z, velocity V, differential reflectivity factor ZDR, and correlation coefficient from the corrected weather radar data. , reflectivity factor texture features , differential reflectance texture feature SD(Phidp) and co-correlation coefficient texture feature , input into the deep learning model to obtain the data quality judgment result of each sampling point, and according to the data quality judgment result, the low-quality sampling point data are eliminated from the corrected weather radar data, and the retained sampling point data are used as the weather radar data after quality control.

[0090] The deep learning model can adopt a multi-layer perceptron (MLP) model.

[0091] A training sample set can be pre-built to train the deep learning model.

[0092] In one embodiment, the above-mentioned quality control steps ① and ② are performed sequentially. However, it is understandable that one of step ① or step ② may be omitted and a single method may be used for quality control.

[0093] (2) Data fusion of multiple weather radars

[0094] The weather radar data after quality control of multiple radars are fused according to the time, that is, the weather radar data after quality control of multiple radars at the same time are fused to obtain three-dimensional grid data with unified altitude, latitude and longitude arrangement at that time.

[0095] The specific steps may include:

[0096] ①Interpolate the radar reflectivity factor value in a single radar coordinate system to a unified Cartesian coordinate system to generate the reflectivity factor on the contour surface, which can also be called CAPPI echo data. Specifically:

[0097] For the spatial coordinates (x, y, z), where x, y represent horizontal coordinates and z represents height. First, calculate the azimuth AZ and ground distance GR of the spatial coordinate relative to the radar based on the x, y coordinates, and then use the ground distance GR and height z to calculate the elevation EL and detection distance SL of the spatial coordinate relative to the radar. Based on EL and SL, calculate the height difference between the upper and lower elevation angles of the spatial coordinate actually detected by the radar, and calculate the interpolation weights w1 and w2. Calculate the reflectivity R of the spatial coordinate (x, y, z), R = w1*R1 + w2*R2, where R1 and R2 represent the reflectivity of the upper and lower elevation angles, respectively.

[0098] ② Combined with the longitude and latitude of the radar station, the Cartesian data is spatially aligned through geographic coordinate transformation to generate three-dimensional grid data with equal longitude arrangement.

[0099] ③ The three-dimensional grid point data of multiple radars are stitched together to generate three-dimensional grid point data with uniform altitude, latitude and longitude arrangement, which can also be called three-dimensional CAPPI echo data.

[0100] Specifically, the distance between the spatial coordinates and the radar is used as a weight factor, and the data of multiple radars are overlapped and fused to generate three-dimensional grid data with uniform altitude, latitude, and longitude.

[0101] (3) Radar echo movement prediction

[0102] The three-dimensional CAPPI echo data can be used to predict the movement of radar echoes at different heights.

[0103] Get continuous 3D grid data of two consecutive times, normalize the 3D grid data of two consecutive times and convert them into grayscale image data, recorded as gray1 and gray2 respectively. Take gray1 and gray2 as input, use the calcOpticalFlowFarneback function in Opencv to calculate the optical flow field and predict the echo moving speeds Vx and Vy. According to Vx, Vy, and the time interval t of the radar data of two consecutive times, calculate the extrapolated coordinates (x_new, y_new) of the coordinates (x, y) in gray2. By remapping the data points in gray2 to the new position and combining the interpolation algorithm, the extrapolated 3D grid data is obtained. Extrapolate the CAPPI data of all layers to obtain the extrapolated 3D grid data. Opencv is a cross-platform computer vision and machine learning software library released under the Apache2.0 license (open source).

[0104] (4) Storm component and storm cell identification

[0105] In one embodiment, identifying storm components and storm cells based on extrapolated three-dimensional grid data includes the steps of:

[0106] ① Based on the extrapolated three-dimensional grid data, identify the strong echo area.

[0107] Specifically, obtain the extrapolated three-dimensional grid data, recorded as R (lat, lon, level), where lat, lon, and level represent latitude, longitude, and number of layers (height), respectively. Each (lat, lon) is a grid. For each grid (lat, lon), there are P reflectivity factors in the direction perpendicular to the surface. P represents the number of jigsaw height layers. Take the maximum value of the P reflectivity factor values, recorded as VMAX (lat, lon). For all grids, calculate the maximum value of the reflectivity factor value in the direction perpendicular to the surface to form a vertical maximum reflectivity grid field, also called a CREF grid field.

[0108] All areas with reflectivity factors greater than 45 dBZ in the CREF grid field are regarded as strong echo areas. The connected domain algorithm is used to identify all connected strong echo areas, which are recorded as ST_CREF.

[0109] ② For each strong echo area, identify the strongest reflectivity factor area at each altitude layer. The specific processing includes:

[0110] Each ST_CREF contains P altitude layers, and the echo area at each altitude is recorded as ECHO. The strongest reflectivity factor area is further identified for the ECHO at P altitude layers to generate N strongest reflectivity factor areas. The specific method is:

[0111] The ECHO echo area is further searched according to the connected domain algorithm with 70dbz, 65dbz, 60dbz, 55dbz, 50dbz and 45dbz as the action threshold THD. If the intensity of the identified reflectivity factor is greater than THD and the searched area exceeds 10 square kilometers, it is recorded as the strongest reflectivity factor area of ​​the ECHO and the search is terminated. The strongest reflectivity factor area of ​​the lth altitude layer is recorded as MAX_REF(l), and there are P MAX_REF(l)s in total for P altitude layers.

[0112] ③ Associate strong echo areas to identify storm cells and storm components.

[0113] Using the obtained P MAX_REF(l), calculate each centroid position and echo intensity value, and determine the altitude layer Level where the maximum reflectivity factor of MAX_REF(l) is located.

[0114] Associated strong echo area: Search the associated strongest echo area vertically upward and downward with the center of the altitude layer Level. If the horizontal distance of the center of mass is less than a certain threshold, it is considered that the two are associated and belong to the same storm cell.

[0115] Furthermore, taking the centroid position of the strong echo area of ​​each layer of the same storm cell as the center, the reflectivity factor within a range of 10 km is taken as the storm component of each layer of the storm cell.

[0116] S3, calculate the storm cell hail characteristic height based on the 0℃ height and the height of the storm components contained in the storm cell.

[0117] For each storm cell identified as above, feature information is extracted based on the storm components constituted by the storm cell.

[0118] Furthermore, S3 includes sub-steps:

[0119] S301, each layer of storm components is composed of multiple grid points, and the centroid of the storm component is used to extract hail feature information, that is, the centroid height of the adjacent storm components is used to calculate the top height and bottom height of the storm component contained in the storm cell. The top height of the mth storm component in the storm cell is recorded as , the bottom height of the mth storm component in the storm cell is recorded as , 1≤m≤M, M represents the total number of storm components in a storm cell, the bottom layer is regarded as the first storm component, and the top layer is regarded as the Mth storm component.

[0120] When the storm component is not the uppermost and lowermost layers, the top height Get the current storm component centroid height and the centroid height of the upper storm component The average of storm components is low and high Get the current storm component centroid height and the centroid height of the lower storm component The average of is expressed as:

[0121] ;

[0122] When the storm component is the uppermost layer, it is expressed as:

[0123] ;

[0124] ;

[0125] When the storm component is the lowest layer, it is expressed as:

[0126] ;

[0127] ;

[0128] S302, record the 0℃ height as , the vertical thickness of the mth storm component in the storm cell is recorded as ,like > , = - ,otherwise, = - ;

[0129] S303, sequentially calculating the hail characteristic development height HHAIL of each storm cell. The calculation of the hail characteristic development height HHAIL of a single storm cell specifically includes the following sub-steps:

[0130] S3031, taking the reflectivity of 45dBZ as the judgment threshold, recorded as REF_THD, if the maximum reflectivity factor of the highest-level storm component is greater than REF_THD, the top height of the highest-level storm component is taken as the hail characteristic development height HHAIL of the storm cell, and the calculation ends; if the maximum reflectivity factor of the highest-level storm component is not greater than REF_THD, according to the weather radar echo height from high to low, the lower-level storm component of the highest-level storm component is taken as the current-level storm component;

[0131] S3032: Compare whether the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ. If the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ, use the maximum reflectivity factor REF_MAX of the storm component of the current layer, the centroid height C_HT of the storm component of the current layer, the reflectivity factor REF_MAX_PRV of the next layer of the current layer, and the centroid height C_HT_PRV of the next layer of the storm component of the current layer to calculate the height corresponding to 45dBZ by extrapolation as the hail characteristic development height HHAIL of the storm cell. The calculation formula is:

[0132] ;

[0133] The calculation is finished;

[0134] If the maximum reflectivity factor of the current layer storm component is not greater than 40dBZ, the next layer of the current layer storm component is taken as the new current layer storm component according to the weather radar echo height from high to low, and step S3032 is repeated.

[0135] According to the method of steps S3031 and S3032, the characteristic height of hail of each storm cell is extracted in turn, and the characteristic height of hail of the nth storm cell is recorded as .

[0136] S4, calculate the height difference between the characteristic height of the hail in the storm cell and the height of 0°C, and calculate the probability index of hail occurrence in the hail cell based on the height difference.

[0137] Specifically, calculate the storm cell hail characteristic height of the nth storm cell Height of zero degree layer of atmosphere Height Difference , pre-establish the height difference The probability table of hail occurrence on the ground is used to calculate the probability index of hail occurrence in each storm cell by looking up the table. The probability index of hail occurrence in the nth storm cell is recorded as The specific table can be shown in Table 1.

[0138] Table 1 Empirical relationship between hail characteristic height and hail occurrence probability

[0139] ;

[0140] Furthermore, a weather radar data processing method according to an embodiment of the present invention may further include the steps of:

[0141] S5, calculate the strong hail index and strong hail probability index of the storm cell.

[0142] The severe hail index is a parameter used to indicate the probability of severe hail occurring in a single storm cell. The difference between the two is that one is expressed in numerical values, while the other is expressed in probability.

[0143] Severe hail probability index refers to.

[0144] Specifically, S5 includes sub-steps:

[0145] S501, based on the reflectivity factor of the storm component in the storm cell, define a first weight coefficient for converting each storm component in the storm cell into hail energy.

[0146] The reflectivity factor of the mth storm component is recorded as , calculate the first weight coefficient corresponding to the mth storm component that can be converted into hail energy .

[0147] According to forecasting experience, the high and low thresholds of the conversion interval are statistically set as follows: , , then the interval size .

[0148] When the reflectivity factor of the storm component Lower than , the first weight coefficient Set to 0; reflectivity factor of the storm component Greater than , weight coefficient is 1.

[0149] When the reflectivity factor of the storm component Between and When the first weight coefficient The linear interpolation method is used to calculate: .

[0150] S502, based on the 0°C height and the -20°C height, define a second weight coefficient of the influence of the centroid height of each storm component in the storm cell on the hail growth.

[0151] The 0° layer height obtained by the atmospheric temperature profile is recorded as , the -20° layer height is recorded as , calculate the thickness between two temperature layers - .

[0152] The second weight coefficient of the mth storm component is recorded as , when the centroid height of the mth storm component Below 0° floor height When the second weight coefficient Set to 0; Storm component centroid height Above -20° floor height When the second weight coefficient is 1.

[0153] When the storm Between and When the second weight coefficient The linear interpolation method is used to calculate: .

[0154] S503, calculating the hail kinetic energy flux of each storm component in the storm cell according to the first weight coefficient and the reflectivity factor of the storm component in the storm cell.

[0155] The reflectivity factor of the mth storm component Transformed into the hail kinetic energy flux of the mth storm component :

[0156] ;

[0157] Where m represents the mth storm component, , is the first weight coefficient and reflectivity factor corresponding to the mth storm component, and is a constant coefficient.

[0158] S504, calculating a strong hail index for each storm component in the storm cell according to the hail kinetic energy flux corresponding to the maximum reflectivity factor of the storm component in the storm cell, the second weight coefficient, and the vertical thickness of the storm component.

[0159] Severe hail index of the storm component of the mth storm component According to the hail kinetic energy flux corresponding to the maximum reflectivity factor of the mth storm component , the second weight coefficient of the mth storm component , the vertical thickness of the mth storm component The calculation formula is as follows:

[0160] ;

[0161] S505, calculating the strong hail index of the storm cell according to the strong hail index of each storm component in the storm cell.

[0162] The severe hail index of storm components at different elevation angles By adding them up, we can get the severe hail index (SHI) of the entire storm cell:

[0163] ;

[0164] S506, calculating the strong hail probability index of the storm cell according to the strong hail index of the storm cell.

[0165] Based on the severe hail index SHI and 0℃ layer height of the storm cell , the probability conversion function is used to calculate the severe hail probability index POSH of the storm cell:

[0166] ;

[0167] Calculate the hail index for all grid points using the above method.

[0168] A data processing system for a weather radar according to an embodiment of the present invention includes:

[0169] The atmospheric temperature profile acquisition and extraction module is used to acquire the atmospheric temperature profile collected by the microwave radiometer and perform quality control, and extract the 0℃ height and -20℃ height from the atmospheric temperature profile after quality control;

[0170] A storm cell identification module is used to obtain weather radar data and perform quality control, identify storm components at each elevation angle from the quality-controlled weather radar data and calculate the centroid position of the storm components, and identify storm cells based on the horizontal distance between the centroids of the storm components;

[0171] A storm cell hail characteristic height calculation module is used to calculate the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm component contained in the storm cell;

[0172] The hail probability index calculation module is used to calculate the height difference between the characteristic height of the hail in the storm cell and the 0°C height, and calculate the probability index of hail occurring in the hail cell based on the height difference.

[0173] Furthermore, the data processing system of the weather radar also includes a severe hail prediction module.

[0174] The severe hail prediction module comprises:

[0175] A first weight coefficient conversion module, used for defining a first weight coefficient for converting each storm component in the storm cell into hail energy based on a reflectivity factor of the storm component in the storm cell;

[0176] A second weight coefficient conversion module is used to define a second weight coefficient of the influence of the centroid height of each storm component in the storm cell on the growth of hail based on the 0°C height and the -20°C height;

[0177] A hail kinetic energy flux calculation module is used to calculate the hail kinetic energy flux of each storm component in the storm cell according to the first weight coefficient and the reflectivity factor of the storm component in the storm cell;

[0178] A storm component severe hail index calculation module is used to calculate the severe hail index of each storm component in the storm cell according to the hail kinetic energy flux corresponding to the maximum reflectivity factor of the storm component in the storm cell, the second weight coefficient, and the vertical thickness of the storm component;

[0179] The storm cell strong hail index calculation module is used to calculate the strong hail index and strong hail probability index of the storm cell according to the strong hail index of each storm component in the storm cell.

[0180] The method of the above data processing system has the same principle and technical effect as the data processing method of the weather radar, and will not be repeated here.

[0181] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. 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 present invention.

Claims

1. A data processing method for a weather radar, characterized in that: Includes steps: Obtain the atmospheric temperature profile collected by the microwave radiometer and perform quality control, and extract the 0°C height and -20°C height from the atmospheric temperature profile after quality control; Acquire weather radar data from multiple radars and perform quality control, fuse the weather radar data after quality control from multiple radars to obtain three-dimensional grid data with uniform height, latitude and longitude, and identify storm components and storm cells based on the three-dimensional grid data; Calculating the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm component contained in the storm cell; Calculate the height difference between the characteristic height of hail in the storm cell and the 0°C height, and calculate the probability of hail in the hail cell according to the height difference; The step of calculating the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm component contained in the storm cell comprises the following steps: S301, using the centroid heights of adjacent storm components, calculate the top height and bottom height of the storm component contained in the storm cell. The top height of the mth storm component in the storm cell is recorded as , the bottom height of the mth storm component in the storm cell is recorded as ; S302, record the 0℃ height as , the vertical thickness of the mth storm component in the storm cell is recorded as ,like > , = - ,otherwise, = - ; S303, sequentially calculating the hail characteristic development height HHAIL of each storm cell. The calculation of the hail characteristic development height HHAIL of a single storm cell specifically includes the following sub-steps: S3031, taking the reflectivity of 45dBZ as the judgment threshold, recorded as REF_THD, if the maximum reflectivity factor of the highest-level storm component is greater than REF_THD, the top height of the highest-level storm component is taken as the hail characteristic development height HHAIL of the storm cell, and the calculation ends; if the maximum reflectivity factor of the highest-level storm component is not greater than REF_THD, according to the weather radar echo height from high to low, the next layer of storm components of the highest-level storm component is taken as the current layer of storm components; S3032: Compare whether the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ. If the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ, use the maximum reflectivity factor REF_MAX of the storm component of the current layer, the centroid height C_HT of the storm component of the current layer, the reflectivity factor REF_MAX_PRV of the next layer of the current layer, and the centroid height C_HT_PRV of the next layer of the storm component of the current layer to calculate the height corresponding to 45dBZ by extrapolation as the hail characteristic development height HHAIL of the storm cell. The calculation formula is: ； The calculation is finished; If the maximum reflectivity factor of the current layer storm component is not greater than 40dBZ, the next layer of the current layer storm component is taken as the new current layer storm component according to the weather radar echo height from high to low, and step S3032 is repeated.

2. A weather radar data processing method as claimed in claim 1, characterized in that: The steps are also included: calculating the severe hail index and severe hail probability index of the storm cell; The calculation of the severe hail index and severe hail probability index of a storm cell comprises the following sub-steps: Based on the reflectivity factor of the storm component in the storm cell, a first weight coefficient for converting each storm component in the storm cell into hail energy is defined; Based on the 0°C height and the -20°C height, define a second weight coefficient of the influence of the centroid height of each storm component in the storm cell on the growth of hail; Calculating the hail kinetic energy flux of each storm component in the storm cell according to the first weight coefficient and the reflectivity factor of the storm component in the storm cell; The severe hail index of each storm component in the storm cell is calculated according to the hail kinetic energy flux corresponding to the maximum reflectivity factor of the storm component in the storm cell, the second weight coefficient and the vertical thickness of the storm component; The severe hail index of a storm cell is calculated based on the severe hail index of each storm component in the storm cell; The probability of severe hail in a storm cell is calculated based on the severe hail index of the storm cell.

3. The data processing method of a weather radar according to claim 1, characterized in that: The method of obtaining the atmospheric temperature profile collected by the microwave radiometer and performing quality control comprises the steps of: Get the atmospheric temperature profile collected by the microwave radiometer, denoted as , , express The temperature of the i-th sampling point in , 1≤i≤N, N represents the total number of sampling points; Get the L-band sounding temperature data of the corresponding time, recorded as , , express The temperature of the i-th sampling point in ; Calculate the temperature observation deviation of the microwave radiometer, denoted as , , express The temperature deviation of the i-th sampling point in X i Normalized to , , express The mean of express If the standard deviation of , then the sampling point data Value removed.

4. The data processing method of a weather radar according to claim 1, characterized in that: The quality control of weather radar data for each radar includes the following steps: Correction of weather radar data; The reflectivity factor, speed, differential reflectivity factor, co-correlation coefficient, reflectivity factor texture features, differential reflectivity texture features and co-correlation coefficient texture features are extracted from the corrected weather radar data and input into the deep learning model to obtain the data quality judgment result of each sampling point. According to the data quality judgment result, the sampling point data with low quality are eliminated from the corrected weather radar data, and the retained sampling point data are used as the weather radar data after quality control.

5. The data processing method of a weather radar according to claim 1, characterized in that: The method of identifying storm components and storm cells based on three-dimensional grid data comprises the following steps: Predict radar echo movement based on the three-dimensional grid reflectivity field to obtain extrapolated three-dimensional grid data; Identify storm components and storm cells based on extrapolated three-dimensional grid data.

6. A weather radar data processing method as claimed in claim 2, characterized in that: The calculation formula for the hail kinetic energy flux of each storm component in a storm cell is: ； represents the hail kinetic energy flux of the mth storm component in the storm cell, represents the first weight coefficient of the mth storm component, represents the reflectivity factor of the mth storm component, and is a constant coefficient; The calculation formula for the severe hail index of each storm component in a storm cell is: ； It represents the severe hail index of the mth storm component in a storm cell. represents the vertical thickness of the mth storm component, Represents the second weight coefficient of the mth storm component.

7. A weather radar data processing method as claimed in claim 6, characterized in that: The calculation formula for the severe hail index of a storm cell is: ； M represents the total number of storm components in a storm cell, Indicates the severe hail index of a storm cell.

8. A data processing system for a weather radar, characterized in that: include: The atmospheric temperature profile acquisition and extraction module is used to acquire the atmospheric temperature profile collected by the microwave radiometer and perform quality control, and extract the 0℃ height and -20℃ height from the atmospheric temperature profile after quality control; A storm cell identification module is used to obtain weather radar data from multiple radars and perform quality control, fuse the weather radar data after quality control from multiple radars, obtain three-dimensional grid point data with uniform height, latitude and longitude, and identify storm components and storm cells based on the three-dimensional grid point data; A storm cell hail characteristic height calculation module, used to calculate the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm component contained in the storm cell; A hail probability index calculation module is used to calculate the height difference between the characteristic height of hail in a storm cell and the 0°C height, and calculate the probability of hail in the hail cell according to the height difference; The step of calculating the storm cell hail characteristic height according to the 0°C height and the centroid height of the storm component contained in the storm cell comprises the following steps: S301, using the centroid heights of adjacent storm components, calculate the top height and bottom height of the storm component contained in the storm cell. The top height of the mth storm component in the storm cell is recorded as , the bottom height of the mth storm component in the storm cell is recorded as ; S302, record the 0℃ height as , the vertical thickness of the mth storm component in the storm cell is recorded as ,like > , = - ,otherwise, = - ; S303, sequentially calculating the hail characteristic development height HHAIL of each storm cell. The calculation of the hail characteristic development height HHAIL of a single storm cell specifically includes the following sub-steps: S3031, taking the reflectivity of 45dBZ as the judgment threshold, recorded as REF_THD, if the maximum reflectivity factor of the highest-level storm component is greater than REF_THD, the top height of the highest-level storm component is taken as the hail characteristic development height HHAIL of the storm cell, and the calculation ends; if the maximum reflectivity factor of the highest-level storm component is not greater than REF_THD, according to the weather radar echo height from high to low, the next layer of storm components of the highest-level storm component is taken as the current layer of storm components; S3032: Compare whether the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ. If the maximum reflectivity factor of the storm component of the current layer is greater than 40dBZ, use the maximum reflectivity factor REF_MAX of the storm component of the current layer, the centroid height C_HT of the storm component of the current layer, the reflectivity factor REF_MAX_PRV of the next layer of the current layer, and the centroid height C_HT_PRV of the next layer of the storm component of the current layer to calculate the height corresponding to 45dBZ by extrapolation as the hail characteristic development height HHAIL of the storm cell. The calculation formula is: ； The calculation is finished; If the maximum reflectivity factor of the current layer storm component is not greater than 40dBZ, the next layer of the current layer storm component is taken as the new current layer storm component according to the weather radar echo height from high to low, and step S3032 is repeated.

9. A weather radar data processing system as claimed in claim 8, characterized in that: Also includes a severe hail forecast module; The severe hail prediction module comprises: A first weight coefficient conversion module, used for defining a first weight coefficient for converting each storm component in the storm cell into hail energy based on a reflectivity factor of the storm component in the storm cell; A second weight coefficient conversion module is used to define a second weight coefficient of the influence of the centroid height of each storm component in the storm cell on the growth of hail based on the 0°C height and the -20°C height; A hail kinetic energy flux calculation module, used to calculate the hail kinetic energy flux of each storm component in the storm cell according to the first weight coefficient and the reflectivity factor of the storm component in the storm cell; A storm component severe hail index calculation module is used to calculate the severe hail index of each storm component in the storm cell according to the hail kinetic energy flux corresponding to the maximum reflectivity factor of the storm component in the storm cell, the second weight coefficient, and the vertical thickness of the storm component; The storm cell strong hail index calculation module is used to calculate the strong hail index and strong hail probability of the storm cell according to the strong hail index of each storm component in the storm cell.

Images