131 results about "Radar observations" patented technology

The invention relates to an atmospheric concentration field inversion method and system based on a space-time diagram network. The method comprises the following steps: acquiring atmospheric observation data of different platforms; the method comprises the following steps: standardizing and preprocessing atmospheric observation data of different platforms to generate space-time point cloud data in a unified format; mapping the time-space point cloud data into a graph network structure for describing an atmospheric physical transmission process; constructing a hybrid neural network architecture, taking the graph network structure as the input of the hybrid neural network architecture, and training the hybrid neural network architecture to obtain a prediction model; inputting real-time radar observation data into the prediction model to obtain a three-dimensional particulate matter concentration field prediction result; and analyzing the prediction result of the three-dimensional particulate matter concentration field, and evaluating the pollution cause of the current pollution event. According to the technical scheme, a high-fidelity and time-space continuous concentration field can be generated under the condition that observation data is limited, and the accuracy and reliability of an inversion result are greatly improved.

The invention relates to a wave height parameter correction method and system for low-angle laser radar observation data, and relates to the technical field of ocean observation data processing, and the method comprises the steps: carrying out the smoothing and data cleaning processing of original wave height data, so as to construct a fitting model of polynomial and Gaussian global trend analysis; and then, through double-model dynamic fusion, polynomial short-term fluctuation capture and Gaussian long-term trend capture are deeply combined, so that the generalization ability of the model is improved. And finally, intelligent reference interval positioning is introduced to adapt to wave height distribution changes in different periods and positions, excessive correction is avoided in combination with introduced correction factors, accurate correction data are obtained, the problem that the wave height is attenuated along with the distance is solved, and the efficiency and reliability of correction processing are remarkably improved.

The invention provides a hectometer-level short temporary rainfall forecasting method based on generative adversarial network downscaling and physical constraint, which belongs to the technical field of weather forecast, and is characterized in that S-band radar jigsaw and high-resolution X-band radar observation data are fused through an adaptive rule to generate a fused radar echo combined reflectivity, and the combined reflectivity is used for forecasting the rainfall of the hectometer-level short temporary rainfall. A super-resolution downscaling model is constructed by using a generative adversarial network, a long-sequence high-resolution radar echo training data set is reconstructed, a deep learning forecasting model is constructed, a high-resolution radar echo sequence is used as input, space-time modeling training is performed by using a composite loss function, and a future radar echo forecasting field is output. The Z-R relation is converted into quantitative rainfall forecast, and finally the quantitative rainfall forecast with the spatial resolution reaching the hectometer level is generated. The interpretability of the model is enhanced through physical constraint, the bottleneck of insufficient hectometer-level resolution training data is effectively solved, and the spatial refinement degree of short temporary rainfall forecasting and the forecasting capacity for a severe convection system are remarkably improved.

The invention discloses a laser wind-finding radar and a data processing and assimilation method thereof. The method comprises the following steps: acquiring observation data of the laser wind-finding radar, atmospheric stability parameters and topographic relief feature data; a three-dimensional space weight distribution function is constructed based on the parameters, and initial influence weights of the observation points on surrounding grids are represented; dynamically adjusting the horizontal influence range and the vertical attenuation characteristic of the function according to the geomorphic stratification degree and the atmospheric vertical stratification change information, and generating dynamic weight distribution matched with the local atmospheric physical and topographic characteristics; the distribution serves as a space adjustment factor of an observation error covariance matrix and is introduced into a cost function, and the wind field analysis field and the dynamic weight distribution are synchronously optimized by minimizing the cost function; and outputting the optimized wind field analysis field for numerical weather forecast. According to the method, accurate and efficient fusion of observation information in three-dimensional variational assimilation is realized, and the authenticity and forecast reliability of wind field analysis under complex conditions are remarkably improved.

By changing the operating modes of weather observation sensors, including weather radar, according to the weather phenomenon being predicted, weather phenomena can be predicted with higher accuracy than before. [Solution] The weather forecasting device 3 is a variable-mode weather observation sensor that observes precipitation data and wind direction and wind speed data in an observation range, which is a spatial range to be observed. It includes a weather forecasting unit 5 that receives weather observation data, including precipitation data and wind direction and wind speed data, observed by a weather observation sensor including a weather radar 1 that is positioned so that the observation range encompasses a predetermined weather forecast area, and predicts the weather in the weather forecast area for a predetermined future forecast time range and outputs weather forecast data including precipitation; a sensor data storage unit 4 that stores the installation position, observation range and observation accuracy of each weather observation sensor; and a sensor control unit 6 that receives weather forecast data and controls the weather observation sensor, including changing the operating mode of the variable-mode weather observation sensor.

The invention provides a satellite-borne conical scanning Doppler radar three-dimensional wind field inversion method and device, and the method comprises the steps: dividing a radar observation strip region into a plurality of sub-blocks, enabling the adjacent sub-blocks to have region overlapping, and carrying out the parameterization of a horizontal wind field of each height layer in each sub-block through a segmented continuous change hypothesis; obtaining an optimal horizontal wind field parameter of each height layer in each sub-block by using a variational method, determining a horizontal wind field in each sub-block according to the optimal horizontal wind field parameter, and carrying out linear weighted fusion on the horizontal wind fields at corresponding positions of each sub-block in the overlapping region to obtain a final horizontal wind field; and calculating horizontal divergence according to the optimal horizontal wind field parameter of each height layer in each sub-block, carrying out integration on the horizontal divergence in the height direction by using an atmospheric mass conservation equation to obtain a vertical wind field, and carrying out linear weighted fusion on the vertical wind fields at corresponding positions of each sub-block in the overlapping region to obtain a final vertical wind field. According to the invention, continuous and steady three-dimensional wind field inversion of the satellite-borne conical scanning Doppler radar is realized.

The invention relates to a ground-based radar super-resolution imaging method based on sparse reconstruction and local super-sampling. The ground-based radar super-resolution imaging method comprises the steps of ground-based radar signal modeling and data preprocessing; constructing a measurement model and sparse representation; sparse reconstruction of coarse grids; peak detection and local window selection; local oversampling and fine inversion are carried out; sub-pixel fitting and coordinate refining are carried out; carrying out continuous nuclear least square refining; and imaging fusion and result output are carried out, so that smooth transition among a plurality of overlapped regions is realized. According to the invention, high-precision reconstruction and detail enhancement of ground-based radar observation data are realized, the imaging quality and the target positioning precision can be effectively improved, and the method is suitable for various application scenes such as surface deformation monitoring, slope stability analysis and engineering structure safety detection.

The application discloses a LES super-high-resolution TC atmospheric flow field numerical simulation method and system, and relates to the fields of atmospheric observation, numerical assimilation and high-resolution numerical simulation. The method is characterized in that: three-dimensional dynamic-thermal information is obtained by implementing airship platform radar observation and downward sounding observation in the core area of a tropical cyclone; the radar radial wind and profile observation operators are preprocessed and constructed; a data assimilation strategy is adopted to integrate multi-source observation in a multi-nested grid numerical model to generate a three-dimensional initial / reanalysis field suitable for large eddy simulation; then, a large eddy simulation solver is configured on the innermost grid covering the boundary layer of the tropical cyclone; fine horizontal grids and boundary layer encryption vertical stratification are adopted to explicitly analyze the boundary layer vortex, small-scale convection and turbulent structure; and a super-high-resolution three-dimensional flow field product for fine evaluation of strong wind and disaster risk is output.

The embodiment of the application discloses a kind of multi-modal cloud classification methods based on visual language model, the method comprises: constructing text generation module, generates the cloud description text containing high-level semantic information;Cloud multi-modal feature extraction and fusion module is constructed, cloud image, cloud description text and millimeter wave cloud radar observation data are carried out feature coding and adaptive fusion, obtain multi-modal feature representation;Classifier module is constructed, and the cloud shape classification result of training cloud image is obtained;Total loss function is constructed, and the multi-modal cloud shape classification model consisting of cloud multi-modal feature extraction and fusion module and classifier module is optimized;Test cloud image, and the cloud description text and millimeter wave cloud radar observation data corresponding to test cloud image are input into multi-modal cloud shape classification model and carry out cloud shape classification.The application is realized by introducing high-level semantic cloud description and designing cloud multi-modal adapter, effective fusion of image, radar and text information, improve the accuracy and robustness of cloud shape classification.

The application discloses a meteorological data processing method, a meteorological data processing device and an electronic equipment, and relates to the technical field of meteorology. The method can reduce the error of meteorological data and improve the accuracy of meteorological data. The method comprises the following steps: acquiring base data observed by each weather radar; performing gridding on the base data to obtain first gridding data of an observation area of each weather radar; determining a blur range of radial velocity; performing velocity de-blurring processing on the radial velocity in the first gridding data according to the blur range; determining a shielding range of a reflectivity factor; performing shielding removal processing on the reflectivity factor in the first gridding data according to the shielding range; performing gridding on the first gridding data after the velocity de-blurring processing and the shielding removal processing to obtain second gridding data; merging the second gridding data of each weather radar to obtain meteorological gridding data of a whole area; and converting the meteorological gridding data into target data corresponding to a weather forecasting system, so that the weather forecasting system can use the target data.

This invention discloses a high-fidelity cooperative jamming method and system for UAV swarms targeting networked radar. The method includes: constructing a UAV dynamics model; constructing a UAV energy consumption model, including UAV propulsion energy consumption and DRFM relay energy consumption; designing false flight paths based on the dynamic characteristics, RCS characteristics, and scene adaptability factors of the false target; calculating the required relay power to simulate RCS fluctuations by calculating the target attitude angle and radar observation angle, combined with three-dimensional RCS data; establishing an optimization model containing UAV dynamics constraints with the goal of minimizing the total energy consumption of the UAV swarm, using the flight heading angle as the optimization variable, and solving the optimization model using a genetic algorithm. This invention improves the credibility of false flight paths, increases the duration of flight path deception, ensures the stability of deception in complex battlefield environments, and enhances the practical adaptability of cooperative jamming.

This paper proposes an aerial target identification method based on the fusion of one-dimensional range profile and pole features. It addresses the problem that changes in the flight attitude, heading, or radar observation angle of an aerial target can significantly alter the number, location, and intensity of the target's scattering centers, leading to decreased range profile stability and consequently affecting target identification accuracy. This application achieves more robust and accurate aerial target identification by fusing resonant region pole features and optical region HRRP features, thus balancing identification stability and target discrimination capability, thereby improving the accuracy of civil airspace monitoring and aerial target detection and identification.

The application discloses a radar photoelectric signal processing method for multi-target detection and belongs to the field of radar multi-target identification. The method solves the problem of poor fusion identification effect of radar signals and photoelectric signals in the multi-target identification process. The technical scheme comprises the following steps: synchronously acquiring an original intermediate frequency signal sequence of a phased array radar and a laser reflection spectrum image sequence of a laser radar, and extracting radar observation data vectors and laser observation data vectors; aligning the laser field of view and the radar field of view, and realizing a three-dimensional common field of view; calculating mutual information values of the radar observation data and the laser observation data; calculating an adaptive adjustment factor, jointly filtering the radar signals, and outputting a radar signal frequency domain response; and calculating the correlation probability between targets, and evaluating whether the targets are the same real target through the correlation probability. The application proposes a probability data correlation model, which is used for more accurately acquiring the number of targets and target trajectories in the field of view and realizing high-precision multi-target detection.

The invention discloses a method for inverting warm cloud microphysical parameters based on a ground-based millimeter wave cloud radar and a microwave radiometer, and the method comprises the steps: 1, obtaining the observation data of the millimeter wave cloud radar, and obtaining a whole-layer liquid water path and a temperature gallery line through the microwave radiometer; 2, identifying warm cloud in millimeter wave cloud radar observation by using a temperature gallery line T; 3, constructing an initial liquid water content profile based on the reflectivity factor and the particle concentration in the warm cloud layer, and calculating an initial liquid water path of the warm cloud layer; 4, the whole-layer liquid water path obtained by the microwave radiometer is used as whole-layer constraint, the initial particle concentration and the initial liquid water content at the vertical height in the warm cloud layer are calibrated according to the proportion, the adjusted warm cloud layer liquid water content and the adjusted warm cloud layer inner particle concentration are obtained, and a physical boundary constraint interval is applied to the particle concentration; and 5, calculating the effective radius of the particles in the warm cloud layer according to the adjusted liquid water content of the warm cloud layer and the particle concentration in the warm cloud layer. According to the method, high-precision warm cloud microphysical parameter vertical distribution can be obtained.

The invention discloses a satellite position real-time resolving platform and method based on multi-source data fusion, and relates to the technical field of satellite position resolving, the platform comprises a multi-source data access and time reference module, a real-time resolving engine module, a rendering and interaction module and an event bus and auditing module, the multi-source data access and time reference module carries out timestamp labeling and quality rating on two rows of element data, radar observation data, optical observation data and global navigation satellite system broadcast data, and the real-time resolving engine module generates a resolving result containing confidence coefficient parameters and check metadata based on an incremental fusion algorithm. According to the method, unified fusion of multi-source data and verifiability of the resolving process are achieved, and the precision and reliability of satellite position resolving are improved.

The invention discloses a rainstorm disaster alarm method based on an X wave band of a phased array rain measuring radar, and belongs to the technical field of radar observation processing, and the method comprises the steps: obtaining rainfall data obtained through the monitoring of an X wave band radar, carrying out the preprocessing of the rainfall data, carrying out the matrix threshold evaluation analysis of the rainfall data of a target region, and obtaining a risk evaluation result of a whole region. The rainfall data is subjected to multi-moment alarm data continuous inspection and identification based on a risk assessment result, continuous tracking and abnormity elimination of a rainfall process are realized, geographic position and area boundary information is superposed, a rainstorm disaster alarm picture result is generated, and a rainstorm disaster early warning technology based on a phased array rain measurement radar can be formed. Therefore, flood-causing short-term rainstorm alarm information production based on refined rainfall products is realized.

The application discloses a TLE target library matching method and system for radar observation data, and belongs to the field of space target monitoring; the method comprises the following specific steps: S1, using an SGP4 model to derive and store position information of all targets in a TLE library; S2, converting radar measurement data into position coordinates of targets in an inertial system; S3, using an extended Laplace algorithm to perform initial orbit determination on a measurement arc segment; S4, screening targets in the TLE library according to the orbit determination result; S5, performing coarse matching of an observation arc segment and a space target by taking position mean square error as a criterion, and obtaining candidate targets; and S6, considering observation error and TLE orbit error, and determining a matching result according to covariance analysis; the application reduces the calculation amount of orbit prediction by using Lagrange interpolation technology, and screens TLE by using initial orbit determination results, so that unnecessary matching calculation can be avoided, coarse matching and fine matching are combined, the target matching speed can be ensured, and the matching precision can also be ensured.

The application provides a kind of space target ISAR image component intelligent segmentation and attitude estimation method, through inverse synthetic aperture radar to space target is continuously observed, and utilize distance-Doppler imaging algorithm to the echo is sequentially imaged, obtain the ISAR image sequence of space target;Utilize deep learning network Pix2pixGAN to the ISAR image sequence is segmented, and the segmentation accuracy is higher;After removing the invalid connected region in each component segmentation result, obtain the final segmentation image of each component;Linear structure extraction is carried out to the final each component segmentation image using the minimum circumscribed rectangle method, and the method is lower to the image component segmentation accuracy requirement, and the robustness is stronger;Finally, according to the imaging principle of radar observation and ISAR image, an unconstrained optimization problem for solving three-dimensional attitude is constructed, and a particle swarm optimization algorithm is used for solving, and the solving efficiency is higher, to realize the attitude inversion of space target key component.

The application relates to a classification method and device based on microwave and laser radar observation data association features. The method comprises the following steps: an association feature learning network is constructed according to a feature decomposition autoencoder structure of a P3S network and a twin neural network, and the association feature learning network is trained, so that the association feature learning network can extract association features of observation data acquired by multiple sensors for multiple targets, and target classification is performed based on the association features of the multiple observation data. The method can effectively improve the accuracy of target classification.

The application discloses a radar data blind area filling method and system based on 3D-GANs, and the method comprises the following steps: based on the hierarchical characteristics of three-dimensional radar data, a simulated blind area mask is constructed and spliced with corresponding radar reflectivity data to form an input tensor; the input tensor is input into a generator network to output the completed radar volume data; the generator network adopts an improved three-dimensional U-Net structure based on a three-dimensional attention mechanism, and in the encoding layer downsampling, the whole is compressed first, then horizontally compressed to reserve the vertical structure; in the process of the skip connection, a three-dimensional attention module is introduced to weight and control the low-order features output by the encoder; the real radar data and the completion result are compared and discriminated through a three-dimensional discriminator network, and the generator network performance is optimized in combination with the adversarial loss and the reconstruction loss. The application can fill the blind area of three-dimensional radar observation data, keep the echo structure consistency, and improve the detail restoration ability of three-dimensional radar data.

The application provides a weather radar network wind field self-adaptive correction method and system, the method comprises the following steps: according to the radar coverage condition of the multi-radar network in the target area, adopting the local least square method or the three-dimensional variation method based on multi-radar observation to carry out wind field inversion, and obtaining a first wind field; according to the second wind field obtained by the observation equipment in the target area, the first wind field is fused and corrected; based on the digital elevation model data, the terrain forced correction is carried out on the fused and corrected wind field, and the final wind field is obtained. The application can obtain more fine and more physically consistent real-time wind field under the condition of complex terrain and strong convection, and provide more reliable data support for the monitoring, prediction and mechanism research of disastrous weather.

The invention relates to the technical field of weather radar synthesis, in particular to a weather radar reflectivity synthesis method and system based on a synchronous stationary satellite. The method comprises the following steps: acquiring satellite multi-channel observation data and weather radar data; performing data preprocessing based on the acquired satellite multi-channel observation data and weather radar data; the method comprises the following steps of: constructing a weather radar synthesis model based on an improved Swin-Transform and a U-Net; performing model training on the constructed weather radar synthesis model; and performing forward reasoning by using the trained model and obtaining weather radar reflectivity synthesis. According to the method, high-quality weather observation data of the areas are provided, the problem of data vacancy of the weather radar is solved, the observation range of the weather radar is remarkably expanded, and particularly, the monitoring capability of a large-scale weather system and severe convective weather is improved.

The application provides a target joint angle estimation method and device based on projection noise modeling, radar observation data is acquired, and a corresponding signal model is constructed; a strong target set is extracted based on the observation data, and first data is obtained through projection; a non-target noise region is extracted based on the observation data, and second data is calculated; a sparse covariance model and a target optimization function are constructed according to the first data and the second data, iteration update is performed to obtain angle power parameters; the angle power parameters are screened according to a set threshold to obtain a weak target set; and a final target joint angle estimation result is obtained according to the strong target set and the weak target set. Through the combination of strong target suppression and sparse covariance fitting, and the modeling of the non-white noise generated after the strong target is stripped, the robust angle estimation in a complex scene is realized, and the overall estimation precision and robustness are improved.

The invention discloses an asteroid internal structure three-dimensional reverse time migration imaging method based on a single-station radar, and belongs to the technical field of planet detection, and the method comprises the steps: estimating wavelets from the observation data of the single-station radar; constructing an asteroid three-dimensional dielectric model based on a velocity analysis result and prior information; selecting a group of transmitting and receiving source positions, and performing forward modeling by taking wavelets as excitation sources to obtain forward transmission wave fields of electric field components at all moments; interchanging the positions of the transmitting source and the receiving source, and performing forward modeling by taking radar observation data as an excitation source to obtain a corresponding back propagation wave field; zero-delay cross-correlation is carried out on the forward and backward propagation wave fields, and an imaging result of a single source point is obtained; repeating the three steps until the imaging results of all the source points are obtained; and superposing imaging results of all source points, and suppressing low-frequency noise through Laplace filtering to obtain a high-resolution three-dimensional imaging result. By implementing the method provided by the invention, the three-dimensional internal structure of the asteroid can be quickly and accurately obtained, and then a reliable basis is provided for analyzing the origin and evolution of the asteroid.

This invention discloses a high-precision deformation and vibration measurement system for full-field structures based on microwave interferometric radar, comprising: an observation configuration module for acquiring radar observation parameters and completing signal transmission and reception; a signal processing module for preprocessing echo signals and constructing an interferometric phase matrix; a phase unwrapping module for generating a continuous phase sequence and calculating the displacement of reflection points; a spectrum extraction module for extracting the frequency components and vibration amplitude of each reflection point; a response clustering module for extracting clusters of reflection points with similar frequency responses and constructing a vibration distribution map; a modal recognition module for generating vibration mode feature vectors and marking discrepancy regions; and a trajectory generation module for generating a dynamic deformation trajectory map of the full-field structure. This invention achieves high-precision synchronous measurement of deformation and vibration modes of complex structures under full-field, non-contact conditions, effectively improving the accuracy, efficiency, and intelligence level of structural monitoring.

The application discloses a global average sea clutter near real-time estimation method, system, device and storage medium, including the following steps: obtaining multi-source satellite-borne radar historical backscattering coefficient data, and performing time and space matching and binning on the ERA5 wind field; an average sea clutter estimation model is established according to a geophysical model function fitting; the satellite-borne microwave scatterometer wind field is corrected in deviation and fused with a numerical weather prediction wind field, so that a high-coverage, continuous near real-time global sea surface wind field is obtained; the relative azimuth is calculated in combination with radar observation parameters and fused wind direction, and global average sea clutter data is obtained by using the model; the application improves the precision and application range of sea clutter estimation under the condition of medium and small incident angles, provides a high-quality input wind field through a multi-source wind field fusion technology, realizes rapid, stable and near real-time estimation of global average sea clutter, has high timeliness and business feasibility, and can meet the needs of near real-time sea clutter monitoring and ocean remote sensing application.

This invention discloses a method for reconstructing missing frames from meteorological radar by fusing spatiotemporal context information. The method includes: preprocessing raw meteorological radar observation data to obtain standardized radar echo sequence data; employing a random masking strategy to simulate missing frames in the standardized radar echo sequence data, constructing a training sample set; introducing an efficient upsampling module and a multi-scale progressive fusion mechanism to construct a meteorological radar missing frame reconstruction model that integrates spatiotemporal context information and training the model; preserving high-level semantics and low-level details through residual fusion; and aligning cross-level features spatially through upsampling with embedded coordinate information; inputting the actual missing radar echo sequence into the trained missing frame reconstruction model, and outputting a completed radar echo sequence. This method significantly improves the integrity and usability of radar data, providing reliable data support for downstream operations such as accurate monitoring and timely early warning of meteorological disasters.

This application provides a millimeter-wave cloud radar observation method and related apparatus based on multi-source data. A fused dataset is formed by real-time acquisition and preprocessing of at least one of the following: weather radar mosaic data, meteorological satellite data, ground visibility observation data, and co-located or adjacent aerosol lidar data. A preset rule set is used to determine that the current weather scene at the cloud radar station includes at least weather echo approaching, weather echo receding, local weather processes, fog, and non-precipitating clouds. Based on the identified weather scene type, the corresponding combined scanning strategy (composed of a basic scanning mode consisting of vertical top-view scanning, distance-height scanning, and three-dimensional volume scanning, and a fine scanning mode consisting of local planar horizontal scanning and layered planar horizontal scanning, combined according to preset logic) is called from a preset strategy mapping table and executed. During execution, the parameters of the fine scanning mode are dynamically adjusted based on the position and intensity information of the real-time radar echo or changes in the detection mode.

The invention discloses a severe convection weather intelligent early warning method based on phased array weather radar collaborative networking, and belongs to the field of weather early warning. Comprising the following steps: S1, adopting multiple phased array (dual-polarization) weather radars to carry out collaborative networking observation, and obtaining observation data with high temporal-spatial resolution; s2, carrying out quality control on the obtained dual-polarization radar observation base data, wherein the quality control comprises ground clutter suppression, electromagnetic wave interference suppression, radial velocity defuzzification and missing measurement data filling; and S3, projecting observation data of the plurality of phased array weather radars into three-dimensional grid point data under a Cartesian coordinate system, and performing data fusion processing. The method has the beneficial effects that the intelligent severe convection weather early warning method based on phased array weather radar collaborative networking is provided, the dual-polarization detection data and the characteristic quantity thereof are increased, and the recognition precision and the suppression rate of ground clutters are remarkably improved.

The application provides an AI-oriented radar precipitation echo spatio-temporal super-resolution image dataset construction method, which comprises the following steps: judging whether the obtained high spatio-temporal resolution radar precipitation echo original image data has precipitation; performing improved down-sampling on the image data with precipitation; constructing initial matrices according to the multi-scale spatio-temporal target resolution data obtained through down-sampling, and outputting high spatio-temporal resolution radar multi-scale spatio-temporal target resolution precipitation echo image data; screening the output precipitation echo image data through a rainfall coverage rate threshold, and cutting the precipitation echo image data into multiple radar precipitation echo images of the same size and having the same size overlapping area, so as to construct a high spatio-temporal resolution radar precipitation echo multi-scale spatio-temporal resolution image dataset. The application can provide more comprehensive precipitation echo information, improve the generalization ability and robustness of an AI model, and significantly improve the limitation of original radar observation data.