Strong convective rainfall short-term and imminent forecasting method based on dual-polarization radar, electronic equipment and medium

Abstract

The invention discloses a severe convective rainfall short-term and imminent forecasting method based on a dual-polarization radar, electronic equipment and a medium. The method comprises the following steps: acquiring a horizontal reflectivity factor, a differential reflectivity and a ratio differential phase shift through the dual-polarization radar; a display regression model is constructed, and the display regression model takes the horizontal reflectivity factor, the differential reflectivity and the ratio differential phase shift as inputs and takes the precipitation amount as an output; u-Net and ResNet are constructed and trained, and the U-Net and the ResNet take the horizontal reflectivity factor, the differential reflectivity and the ratio differential phase shift at the t moment as input and take corresponding data at the t + 1 moment as output; acquiring real-time horizontal reflectivity factors, differential reflectivity and ratio differential phase shift data, inputting the real-time horizontal reflectivity factors, the differential reflectivity and the ratio differential phase shift data into pre-trained U-Net and ResNet, and averaging prediction results respectively output by the U-Net and the ResNet by using a Bagging strategy to obtain prediction values of the horizontal reflectivity factors, the differential reflectivity and the ratio differential phase shift; and inputting the result into a display regression model to obtain a predicted value of the precipitation.

Description

Technical Field

[0001] The invention belongs to the field of meteorological forecasting, and in particular relates to a short-term forecasting method for severe convective precipitation based on a dual-polarization radar, electronic equipment, and a medium. Background Art

[0002] In recent years, the short-term forecast technology of severe convective precipitation has made significant progress. Traditional meteorological radar technology mainly relies on single polarization radar, which provides horizontal reflectivity factor (Z H ) is used to monitor and forecast severe convective precipitation. However, single-polarization radar can only obtain echo intensity information, resulting in limited ability to identify precipitation particle types and phases. Dual-polarization radar technology came into being. By measuring echo signals in the horizontal and vertical directions, it can provide more information on the microphysical characteristics of precipitation particles, such as the horizontal reflectivity factor (Z H ), differential reflectivity (Z DR ) and the ratio of the differential phase shift (K DP ), etc., thereby improving the accuracy of precipitation estimation and the ability to identify precipitation types.

[0003] The application of dual polarization radar in precipitation forecasting mainly depends on its ability to distinguish different types of precipitation particles. H ) and differential reflectivity (Z DR ) can be used to distinguish different precipitation types such as rain, snow, and hail; the differential phase shift (K DP ) is closely related to the shape and orientation of precipitation particles. These variables, when combined, can provide more accurate precipitation estimates and precipitation type identification. In addition, the correlation coefficient is used to measure the consistency of radar echoes, further improving the accuracy of precipitation type identification.

[0004] However, the processing and analysis of dual-polarization radar data is complex, the data volume is huge, and the computing resource requirements are high, which limits its popularity in practical applications. Existing precipitation forecast models mostly use radar observations in the past hour or less to predict future precipitation conditions. However, existing methods do not adequately extract microphysical feature information from dual-polarization radar data, resulting in low forecast accuracy. Therefore, it is necessary to develop more efficient data processing and analysis methods to fully utilize the data advantages of dual-polarization radars. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a short-term forecasting method for severe convective precipitation based on dual-polarization radar, electronic equipment, and medium.

[0006] In a first aspect, an embodiment of the present invention provides a method for short-term forecasting of severe convective precipitation based on a dual-polarization radar, the method comprising:

[0007] The horizontal reflectivity factor, differential reflectivity and differential phase shift are collected by dual polarization radar, and the collected data are preprocessed;

[0008] A display regression model is constructed, wherein the display regression model takes the horizontal reflectivity factor, the differential reflectivity and the differential phase shift as inputs and takes the precipitation as output;

[0009] Construct and train a U-Net and a ResNet, wherein the U-Net and the ResNet take the horizontal reflectivity factor, the differential reflectivity, and the differential phase shift at time t as input, and take the horizontal reflectivity factor, the differential reflectivity, and the differential phase shift at time t+1 as output;

[0010] Obtain real-time horizontal reflectivity factor, differential reflectivity, and differential phase shift data, input them into pre-trained U-Net and ResNet, and use the Bagging strategy to average the prediction results output by U-Net and ResNet respectively to obtain the predicted values ​​of horizontal reflectivity factor, differential reflectivity, and differential phase shift;

[0011] The predicted values ​​of horizontal reflectivity factor, differential reflectivity and relative differential phase shift are input into the displayed regression model to obtain the predicted value of precipitation.

[0012] In a second aspect, an embodiment of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory is coupled to the processor; wherein the memory is used to store program data, and the processor is used to execute the program data to implement the above-mentioned short-term forecasting method for severe convective precipitation based on dual-polarization radar.

[0013] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned short-term forecasting method for severe convective precipitation based on dual-polarization radar.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the above-mentioned short-term forecasting method for severe convective precipitation based on dual-polarization radar.

[0015] The beneficial effects of the present invention are:

[0016] The present invention provides a short-term forecast method for severe convective precipitation based on dual-polarization radar, which utilizes the horizontal reflectivity factor, differential reflectivity and differential phase shift based on deep learning analysis, thereby improving the accuracy of precipitation estimation and the ability to identify precipitation types. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0018] Figure 1 is a schematic diagram of a short-term forecast method for severe convective precipitation based on a dual-polarization radar provided by an embodiment of the present invention;

[0019] Figure 2 It is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention is further described below in conjunction with embodiments. The description of the following embodiments is only used to help understand the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principles of the present invention, several improvements and modifications may be made to the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

[0021] In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the fact that they can be implemented by ordinary technicians in this field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0022] like Figure 1 The present invention provides a short-term forecasting method for severe convective precipitation based on dual-polarization radar, and the method specifically includes the following sub-steps:

[0023] Step S1, collecting horizontal reflectivity factor, differential reflectivity and ratio differential phase shift through dual polarization radar, and preprocessing the collected data.

[0024] Furthermore, data is usually collected through a network of weather radar stations, which are distributed across various geographical areas and continuously observe and record these variables in real time. The temporal resolution of data collection is generally one frame every 6 minutes, and each frame of data covers a horizontal area of ​​256×256 kilometers.

[0025] Furthermore, the process of preprocessing the collected horizontal reflectivity factor, differential reflectivity and differential phase shift includes:

[0026] The collected horizontal reflectivity factors, differential reflectivity and differential phase shift are subjected to median filtering, Gaussian filtering, logarithmic restoration, data interpolation, standardization, data slicing and data enhancement.

[0027] It should be noted that radar data may be interfered by various noises during the acquisition process, such as electronic noise, ground reflection, building echo, etc. These noises will affect the accuracy of the data, so they need to be processed by denoising algorithms. Median filtering: used to remove isolated noise points and suppress noise by replacing each pixel with the median of its neighboring pixels. Gaussian filtering: applies Gaussian function to smooth the image, reduce the impact of noise, and retain important edge information.

[0028] The process of logarithmically restoring the horizontal reflectivity factor and the differential reflectivity includes:

[0029]

[0030] In the formula, Z H Represents the measured value of the horizontal reflectivity factor, Z DR Represents the measured value of differential reflectivity, Z h Represents the actual value of the horizontal reflectivity factor, Z dr Represents the actual value of the differential reflectivity.

[0031] Furthermore, the radar data is enhanced, including: rotation and flipping: rotating or flipping the spatial information of the radar data to simulate different observation angles. Noise addition: introducing slight Gaussian noise into the data to enhance the robustness of the model.

[0032] Step S2, constructing a display regression model, wherein the display regression model takes the horizontal reflectivity factor, the differential reflectivity and the ratio differential phase shift as inputs and takes the precipitation as output.

[0033] Furthermore, the process of constructing a display regression model specifically includes:

[0034] Based on the ZR relationship formula, an explicit regression model is established; the explicit regression model takes the horizontal reflectivity factor, the differential reflectivity and the differential phase shift as inputs, and takes the precipitation as output;

[0035] At the same time, the parameters of the explicit regression model are optimized by the least squares method, so that the explicit regression model can accurately reflect the empirical relationship between the horizontal reflectivity factor, differential reflectivity, ratio differential phase shift and precipitation.

[0036] Step S3, constructing and training U-Net and ResNet, wherein the U-Net and ResNet take the horizontal reflectivity factor, differential reflectivity and differential phase shift at time t as input, and take the horizontal reflectivity factor, differential reflectivity and differential phase shift at time t+1 as output.

[0037] Among them, the expression of U-Net is as follows:

[0038] ZH，t+1 =F U (Z H，t , Z DR，t , K DP，t )

[0039] In the formula, Z h，t represents the horizontal reflectivity factor at time t, Z DR,t represents the differential reflectivity at time t, K DP，t represents the differential phase shift at time t, F u (.) indicates the U-Net processing process.

[0040] The expression of ResNet is as follows:

[0041] Z H，t+1 =F R (Z H，t , Z DR，t , K DP，t )

[0042] In the formula, F R (.) indicates the ResNet processing process.

[0043] Furthermore, the process of training U-Net and ResNet also includes:

[0044] The critical success index, normalized absolute error, and correlation coefficient were calculated to evaluate the model performance of U-Net and ResNet.

[0045] Among them, the critical success index (CSI) is used to evaluate the accuracy of the model in severe convective precipitation prediction. CSI is defined as:

[0046]

[0047] Among them, hits refers to the number of hits, which refers to the number of grid pixels whose observed radar reflectivity and predicted radar reflectivity are both greater than the threshold, misses refers to the number of missed alarms, which refers to the number of grid pixels whose observed radar reflectivity is greater than the set threshold but the predicted reflectivity is lower than the threshold, and false alarms refers to the number of false alarms, which refers to the number of grid pixels whose observed reflectivity is lower than the threshold but the predicted reflectivity is higher than the threshold. The higher the CSI index, the better the prediction effect of the model.

[0048] Normalized absolute error (NE): measures the size of the model prediction error and is defined as:

[0049]

[0050] The normalized absolute error NE indicates the size of the average prediction error of the model relative to the true value. The smaller the NE, the smaller the error of the model prediction.

[0051] Correlation coefficient (CC): measures the linear correlation between the model prediction results and the actual observations, defined as:

[0052]

[0053] Among them, R' represents the cumulative rainfall predicted by the model, and R represents the cumulative rainfall actually measured. There are N pairs (R', R) in total. represents the average value of the cumulative rainfall predicted by the model, It represents the average value of the accumulated rainfall actually measured.

[0054] Normalized absolute error NE represents the size of the average prediction error of the model relative to the true value. The smaller the value of NE, the more accurate the model's prediction. CC is used to measure the strength and direction of the linear relationship between R' and R. The closer CC is to 1, the more accurate the model's prediction.

[0055] Step S4, obtain the real-time horizontal reflectivity factor, differential reflectivity and differential phase shift data, input them into the pre-trained U-Net and ResNet, and use the Bagging strategy to average the prediction results output by U-Net and ResNet respectively to obtain the predicted values ​​of the horizontal reflectivity factor, differential reflectivity and differential phase shift.

[0056] Step S5, inputting the predicted values ​​of the horizontal reflectivity factor, the differential reflectivity and the differential phase shift into the display regression model to obtain the predicted value of precipitation.

[0057] like Figure 2 As shown, an embodiment of the present application provides an electronic device, which includes a memory 101 for storing one or more programs and a processor 102. When the one or more programs are executed by the processor 102, any method in the first aspect described above is implemented.

[0058] The memory 101, the processor 102 and the communication interface 103 are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines. The memory 101 can be used to store software programs and modules, and the processor 102 executes various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 can be used to communicate signaling or data with other node devices.

[0059] Among them, the memory 101 can be, but is not limited to, a random access memory 101 (Random Access Memory, RAM), a read only memory 101 (Read Only Memory, ROM), a programmable read-only memory 101 (Programmable Read-Only Memory, PROM), an erasable read-only memory 101 (Erasable Programmable Read-Only Memory, EPROM), an electrically erasable read-only memory 101 (Electric Erasable Programmable Read-Only Memory, EEPROM), etc.

[0060] The processor 102 may be an integrated circuit chip with signal processing capability. The processor 102 may be a general-purpose processor 102, including a central processing unit 102 (CPU), a network processor 102 (NP), etc.; it may also be a digital signal processor 102 (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0061] In the embodiments provided in the present application, it should be understood that the disclosed method and system can also be implemented in other ways. The method and system embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the method and system, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0062] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0063] On the other hand, an embodiment of the present application provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by the processor 102, a method as in any one of the first aspects described above is implemented. If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory 101 (ROM, Read-Only Memory), a random access memory 101 (RAM, RandomAccess Memory), a disk or an optical disk.

[0064] Those skilled in the art will readily appreciate other embodiments of the present application after considering the description and practicing the contents disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application. The description and examples are intended to be exemplary only.

[0065] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A short-term forecast method for severe convective precipitation based on dual polarization radar, characterized in that: The method comprises: The horizontal reflectivity factor, differential reflectivity and differential phase shift are collected by dual polarization radar, and the collected data are preprocessed; A display regression model is constructed, wherein the display regression model takes the horizontal reflectivity factor, the differential reflectivity and the differential phase shift as inputs and takes the precipitation as output; Construct and train a U-Net and a ResNet, wherein the U-Net and the ResNet take the horizontal reflectivity factor, the differential reflectivity, and the differential phase shift at time t as input, and take the horizontal reflectivity factor, the differential reflectivity, and the differential phase shift at time t+1 as output; Obtain real-time horizontal reflectivity factor, differential reflectivity, and differential phase shift data, input them into pre-trained U-Net and ResNet, and use the Bagging strategy to average the prediction results output by U-Net and ResNet respectively to obtain the predicted values ​​of horizontal reflectivity factor, differential reflectivity, and differential phase shift; The predicted values ​​of horizontal reflectivity factor, differential reflectivity and relative differential phase shift are input into the displayed regression model to obtain the predicted value of precipitation.

2. The method for short-term forecasting of severe convective precipitation based on dual polarization radar according to claim 1, characterized in that: The process of preprocessing the collected horizontal reflectivity factor, differential reflectivity and differential phase shift includes: The collected horizontal reflectivity factors, differential reflectivity and differential phase shift are subjected to median filtering, Gaussian filtering, logarithmic restoration, data interpolation, standardization, data slicing and data enhancement.

3. The method for short-term forecasting of severe convective precipitation based on dual polarization radar according to claim 2, characterized in that: The process of logarithmically restoring the horizontal reflectivity factor and the differential reflectivity includes: In the formula, Z H Represents the measured value of the horizontal reflectivity factor, Z DR Represents the measured value of differential reflectivity, Z h Represents the actual value of the horizontal reflectivity factor, Z dr Indicates the actual value of the differential reflectivity.

4. The method for short-term forecasting of severe convective precipitation based on dual polarization radar according to claim 1, characterized in that: The process of building an explicit regression model specifically includes: Based on the ZR relationship formula, an explicit regression model is established; the explicit regression model takes the horizontal reflectivity factor, the differential reflectivity and the differential phase shift as inputs, and takes the precipitation as output; At the same time, the parameters of the explicit regression model are optimized by the least squares method, so that the explicit regression model can accurately reflect the empirical relationship between the horizontal reflectivity factor, differential reflectivity, ratio differential phase shift and precipitation.

5. The method for short-term forecasting of severe convective precipitation based on dual polarization radar according to claim 1, characterized in that: The process of training U-Net and ResNet also includes: The critical success index, normalized absolute error, and correlation coefficient were calculated to evaluate the model performance of U-Net and ResNet.

6. An electronic device comprising a memory and a processor, characterized in that: The memory is coupled to the processor; wherein the memory is used to store program data, and the processor is used to execute the program data to implement the short-term forecasting method for severe convective precipitation based on dual polarization radar as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for short-term forecasting of severe convective precipitation based on dual-polarization radar as described in any one of claims 1 to 5 is implemented.

8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the method for short-term forecasting of severe convective precipitation based on dual-polarization radar described in any one of claims 1 to 5 is implemented.

Images