CN111736157A - PPI data-based prediction method and device for nowcasting - Google Patents
PPI data-based prediction method and device for nowcasting Download PDFInfo
- Publication number
- CN111736157A CN111736157A CN202010871064.4A CN202010871064A CN111736157A CN 111736157 A CN111736157 A CN 111736157A CN 202010871064 A CN202010871064 A CN 202010871064A CN 111736157 A CN111736157 A CN 111736157A
- Authority
- CN
- China
- Prior art keywords
- data
- layer
- ppi
- elevation angle
- mapping
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000013507 mapping Methods 0.000 claims abstract description 86
- 238000012549 training Methods 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 14
- 230000015654 memory Effects 0.000 claims description 17
- 238000007781 pre-processing Methods 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 9
- 238000007405 data analysis Methods 0.000 abstract 1
- 239000010410 layer Substances 0.000 description 182
- 238000010586 diagram Methods 0.000 description 7
- 230000033001 locomotion Effects 0.000 description 7
- 238000001556 precipitation Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000002310 reflectometry Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 239000002356 single layer Substances 0.000 description 4
- 230000005291 magnetic effect Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 230000003936 working memory Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/95—Radar or analogous systems specially adapted for specific applications for meteorological use
- G01S13/958—Theoretical aspects
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/415—Identification of targets based on measurements of movement associated with the target
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
- G01S7/417—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section involving the use of neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The application aims to provide a prediction method and equipment based on the prediction of PPI data, and the method and equipment determine the multi-layer PPI original data at each moment in a target number of moments; carrying out mapping processing of a target layer elevation angle on the multi-layer PPI original data at each moment to obtain mapping data corresponding to each moment; splicing channels of mapping data from the first layer to the last layer at all times to serve as network training data; and inputting the network training data into network convolution for sequence learning to obtain PPI images of the forecast target number of frames. The PPI is adopted during data training, so that prediction of a prediction product model is not limited to learning data, and the PPI generated by prediction can be further processed to perform diversified data analysis. In addition, the multi-layer PPI data is fused, so that the model learns multi-dimensional space-time data, and the effect of generating images by the network is more stable.
Description
Technical Field
The present application relates to the field of computers, and in particular, to a prediction method and apparatus for nowcasting based on PPI data.
Background
With the development of remote sensing technology, radar has become an important means for ground detection of weather conditions. The Doppler weather radar is a new type of radar, and it uses the Doppler effect of the change between the echo frequency and the emitting frequency of the precipitation to measure the radial movement speed of the precipitation particles, and uses the speed information chart to measure the wind speed distribution, vertical airflow speed, atmospheric turbulence, precipitation particle spectrum distribution, and wind field structure characteristics in the precipitation, especially in the strong convection precipitation. The reflected signal detected by the radar is called radar echo, and the strength and the structure of the signal detected by the radar reflect the structure of weather and precipitation. By extrapolating the evolution process of the radar level echo diagram over time, it is not difficult to obtain a forecast of the future weather evolution law, such a process being generally referred to as a nowcasting.
The general doppler radar products are displayed in a PPI (PPI) (planar position indicator) mode, a CAPPI (CAPPI), a CR (CR), etc., wherein the PPI (PPI) mode is the most common radar display mode, a radar antenna is usually displayed in the center of a screen, and a target echo wave is displayed in concentric circles. Displaying the distribution condition of target objects around the observation station and the echo intensity thereof when the radar antenna scans for one week at a certain elevation angle in a polar coordinate mode in radar meteorological observation; that is, PPI data is data obtained by scanning the radar for one turn at a certain elevation angle.
With the development of machine learning, many predictions of the nowcast are performed through deep learning of a neural network, and data used for prediction is generally performed by using a CAPPI or CR with a fixed height; since the body sweep of doppler weather radar is cone shaped, there are the following disadvantages:
firstly), when the radar is close to the radar, the combined reflectivity factor only comprises basic reflectivity factors of one elevation angle and two elevation angles, and the representativeness is not strong; secondly), when the radar station is close to the radar station, the average value of the reflectivity factors of the middle and high layers is unavailable due to interference of side-lobe false echoes; thereby affecting the accuracy of the weather forecast. Third), low-layer combined reflectance factor average products are often due to non-precipitation echo composition, and therefore validation of field data is necessary. Fourthly), the data of CAPPI, VCS or CR are generated by PPI interpolation, and a part of data quantity is lost in the data, so that great uncertainty exists.
Disclosure of Invention
An object of the present application is to provide a prediction method and apparatus for nowcasting based on PPI data, which solve the above-mentioned problems of prediction data used in nowhere to perform nowcasting prediction.
According to an aspect of the present application, there is provided a prediction method based on the nowcasting of PPI data, the method including:
determining multilayer PPI original data at each moment in the target quantity moments;
carrying out mapping processing of a target layer elevation angle on the multi-layer PPI original data at each moment to obtain mapping data corresponding to each moment;
splicing channels of mapping data from the first layer to the last layer at all times to serve as network training data;
and inputting the network training data into network convolution for sequence learning to obtain PPI images of the forecast target number of frames.
Further, determining a plurality of layers of PPI raw data at each of the target number of time instants includes:
reading radar scanning data of N elevation angles at each moment in target quantity moments, wherein the radar scanning data comprise images with sizes determined by radar azimuth angle scanning frequency and equidistant sampling points on rays, and N is a positive integer;
and determining N layers of PPI original data at each moment in the target quantity moments according to the radar scanning data of N elevation angles at each moment in the target moments.
Further, performing mapping processing of a target layer elevation angle on the multi-layer PPI raw data at each time to obtain mapping data corresponding to each time, including:
carrying out noise pretreatment on the N layers of PPI original data to obtain N layers of PPI de-noising data;
mapping any layer elevation angle in the N layers of PPI de-noising data to an i layer elevation angle to obtain mapping data which takes the i layer as a first layer and takes M layers as a last layer, wherein i is more than or equal to 1 and less than or equal to N, and M = N-i;
splicing the channels of the mapping data from the first layer to the last layer at all times, wherein the splicing comprises the following steps:
and splicing the mapping data of the ith layer to the mapping data channel of the Mth layer at all the moments.
Further, noise preprocessing is performed on the N layers of PPI raw data to obtain N layers of PPI de-noising data, including:
preprocessing the N layers of PPI original data by adopting a horse-type distance to remove fixed position noise to obtain a preprocessing result;
and screening out a value which is more than or equal to 10dBz from the preprocessing result to obtain N layers of PPI denoising data.
Further, mapping any layer elevation angle in the N layers of PPI de-noised data to an ith layer elevation angle, including:
determining a first candidate point in the j layer elevation angle in the same horizontal direction with a fixed point of the i layer elevation angle and a second candidate point in the same vertical direction, wherein the j layer is any non-i layer in the N layers;
calculating a candidate point distance between the fixed point and a first candidate point, and determining an influence point related to the fixed point in the j-th layer elevation according to the candidate point distance and a distance threshold;
and determining the influence points of all fixed points in the ith layer elevation angle, and completing the mapping of PPI de-noising data in the jth layer elevation angle to the ith layer elevation angle.
Further, determining the influence points of all fixed points in the i-th layer elevation angle, and completing the mapping of the PPI de-noising data in the j-th layer elevation angle to the i-th layer elevation angle, including:
inquiring PPI de-noising data corresponding to the influence point of the fixed point in the j layer elevation angle in an inquiring step, and taking the inquired data as the PPI data of the fixed point in the i layer elevation angle;
determining the influence points of all fixed points in the ith layer elevation angle, and completing the mapping of the PPI de-noising data in the jth layer elevation angle to the ith layer elevation angle when the mapping of the PPI de-noising data in the jth layer elevation angle to the ith layer elevation angle is completed.
Further, determining an impact point in the elevation of the j-th layer with respect to the fixed point according to the candidate point distance and a distance threshold, comprising:
when the candidate point distance is smaller than a distance threshold, taking the first candidate point as an influence point in the j layer elevation angle relative to the fixed point;
and when the candidate point distance is greater than or equal to a distance threshold value, taking the second candidate point as an influence point in the j-th layer elevation angle relative to the fixed point.
According to another aspect of the present application, there is also provided a device for prediction based on the nowcasting of PPI data, the device comprising:
determining means for determining a plurality of layers of PPI raw data at each of a target number of times;
the mapping device is used for mapping the target layer elevation angle of the multilayer PPI original data at each moment to obtain mapping data corresponding to each moment;
the splicing device is used for splicing the channels of the mapping data from the first layer to the last layer at all times to serve as network training data;
and the forecasting device is used for inputting the network training data into network convolution for sequence learning to obtain PPI images of forecasting target number of frames.
According to yet another aspect of the present application, there is also provided a device for prediction based on the nowcast of PPI data, the device comprising:
one or more processors; and
a memory storing computer readable instructions that, when executed, cause the processor to perform the operations of the method as previously described.
According to yet another aspect of the present application, there is also provided a computer readable medium having computer readable instructions stored thereon, the computer readable instructions being executable by a processor to implement the method as described above.
Compared with the prior art, the method has the advantages that the multi-layer PPI original data of each moment in the target quantity moments are determined; carrying out mapping processing of a target layer elevation angle on the multi-layer PPI original data at each moment to obtain mapping data corresponding to each moment; splicing channels of mapping data from the first layer to the last layer at all times to serve as network training data; and inputting the network training data into network convolution for sequence learning to obtain PPI images of the forecast target number of frames. The data form in the deep learning network is improved from CAPPI to PPI data after mapping processing, so that the effect of converting the information quantity of network learning and the generated PPI data into CAPPI is better than that of training by directly using CAPPI; the PPI is adopted during training data, so that prediction of a prediction product model is not limited to learning data, data processing can be further performed on the predicted PPI, the PPI can be processed into data formats of various display modes such as CAPPI, CR, RHI and VCS, the data format of the training data is not limited to one, and the training data can be displayed through various display modes, so that the problem of analysis limitation caused by the single data format of the training data is solved. In addition, the multi-layer PPI data are fused into the training of the single-layer PPI data, so that the model learns the multi-dimensional space-time data, and the effect of generating the image by the network is more stable.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
fig. 1 illustrates a flow diagram of a prediction method based on the nowcasting of PPI data provided in accordance with an aspect of the present application;
fig. 2 is a schematic diagram illustrating a process of performing noise preprocessing on the N layers of PPI raw data according to an embodiment of the present application;
FIG. 3 shows a schematic view of different elevation angles in an embodiment of the present application;
FIG. 4 is a flow chart illustrating a data usage method for performing a nowcast using PPI data according to an embodiment of the present disclosure;
fig. 5 illustrates a schematic diagram of an apparatus for prediction based on the nowcasting of PPI data according to another aspect of the present application.
The same or similar reference numbers in the drawings identify the same or similar elements.
Detailed Description
The present application is described in further detail below with reference to the attached figures.
In a typical configuration of the present application, the terminal, the device serving the network, and the trusted party each include one or more processors (e.g., Central Processing Units (CPUs)), input/output interfaces, network interfaces, and memory.
The Memory may include volatile Memory in a computer readable medium, Random Access Memory (RAM), and/or nonvolatile Memory such as Read Only Memory (ROM) or flash Memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, Phase-Change RAM (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash Memory or other Memory technology, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, magnetic cassette tape, tape-Disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (transient media), such as modulated data signals and carrier waves.
Fig. 1 illustrates a flow diagram of a prediction method based on the nowcasting of PPI data according to an aspect of the present application, the method including: step S11 to step S14,
in step S11, determining a plurality of layers of PPI raw data at each of a target number of times; here, PPI (plan position indicator) is a plane position display, and PPI data is data obtained by scanning a circle at a certain elevation angle by a radar. The target number is the number of the selected multiple moments, if the future 5 frames of images are forecasted when the approach forecast is carried out, the target number moments are selected to be five historical moments of T, T +1, T +2, T +3 and T +4, and the future 5 frames of radar echo images are forecasted. In the embodiment of the application, the prediction data used selects PPI data of a plurality of elevation angles at each moment in time during acquisition, so that a plurality of layers of PPI raw data are generated to serve as prediction data sources of the nowcasting. Because the PPI original data has no uncertainty caused by data loss and frame interpolation, the subsequent information quantity of network learning and the effect of converting the generated PPI data into CAPPI are better than the effect of directly training by using the CAPPI.
In step S12, performing mapping processing of a target layer elevation angle on the multi-layer PPI raw data at each time to obtain mapping data corresponding to each time; here, the multi-layer PPI raw data at each time is processed, taking the multi-layer PPI raw data at T time as an example, the multi-layer PPI raw data at the time is mapped, where the mapping process may be directly performed, and in order to obtain more accurate data, the multi-layer PPI raw data may be denoised first, and then the denoised data is mapped at a target layer elevation angle, where the target layer elevation angle is one of the obtained PPI layer numbers, and the denoised data at other elevation angles are mapped to the elevation angle, so as to obtain the mapping data at T time. And performing the mapping processing on the multi-layer PPI original data at other moments to obtain mapping data corresponding to each moment.
In step S13, channels of the mapping data of the first layer to the last layer at all times are spliced to serve as network training data; here, the first layer at all the time points to a target layer selected during multi-layer elevation mapping at the time point, for example, the ith layer, the mapping data of the first layer is the denoised PPI data of the ith layer, the second layer is the (i + 1) th layer, the mapping data of the second layer is the mapping data mapped to the ith layer by the (i + 1) th layer, and so on for the subsequent layers and the corresponding mapping data. And splicing the multi-layer mapping data, wherein the spliced data is used as network training data.
In step S14, the network training data is input into a network convolution for sequence learning, so as to obtain PPI images with a forecast target number of frames. Here, network training data obtained from multiple layers of PPI data are input into a network convolution for convolution training, and a PPI image of a future target frame number is obtained through sequence learning, that is, when a target time is selected from 5 times, a PPI image of a future 5 frames can be predicted. The PPI data of multiple layers are fused into the training of PPI data of a single layer, so that a network model learns multidimensional space-time data, the multidimensional space-time data are time sequences, and have equal heights and vertical heights in space, and the effect of generating images by the network is more stable. Preferably, since the radar data is scanned for one week from one azimuth angle, the first line of data of the obtained PPI image is actually close to the last line of data, but is farthest away from the image, and if the PPI data is trained into CAPPI or other data, the problem of edge jump is generated, therefore, when the network training data is input into the network convolution, the normal convolution in the network can be replaced by a reel convolution, the normal convolution cannot capture the information of the other end of the image at the edge part in the convolution process, the reel convolution is designed, and the two-dimensional image padding is rolled into a reel in the convolution padding (edge padding learning), so that the information of the actual periphery can be captured in the process of learning the edge information.
In an embodiment of the present application, in step S11, radar scan data of N elevation angles at each of a target number of times is read, where the radar data includes an image of a size determined by a radar azimuth scanning frequency and equidistant sampling points on a ray, where N is a positive integer; and determining the multilayer PPI original data at each moment in the target quantity moments according to the radar scanning data of the N elevation angles at each moment in the target moments. Taking the time T as an example, reading radar scanning data of N elevation angles at the time T, performing cone scanning on each elevation angle of the doppler radar, wherein the number of the elevation angles can be set by itself, taking the time T as an example, setting the time N elevation angles, taking the elevation angle of the bottommost layer as a first-layer elevation angle, each frame of data represents an echo signal received after the radar scans for one circle at a certain time, the signal read at each elevation angle is a matrix, the abscissa is an equidistant sampling point on a ray and is denoted by L, the interval distance of each sampling point is L0, if sampling is performed every 300m in the ray direction, L0=300m, the ordinate is a scanning azimuth angle and is denoted by R, the range of R is 0-360 degrees and is one circle, the azimuth scanning frequency is R, and the vertical height corresponding to each sampling point is denoted by h, so that an image with the size of. It should be noted that the acquired N elevation angles include a bottom elevation angle, a middle elevation angle and a top elevation angle, where one elevation angle may include one elevation angle and may also include multiple elevation angles, for example, there are 15 scanning elevation angles for radar scanning in a certain area, and the first 3 elevation angles are taken as bottom elevation angles.
In an embodiment of the present application, in step S12, noise preprocessing is performed on the N layers of PPI raw data to obtain N layers of PPI de-noising data; and mapping any layer elevation angle in the N layers of PPI de-noising data to an i layer elevation angle to obtain mapping data which takes the i layer as a first layer and takes M layers as a last layer, wherein i is more than or equal to 1 and less than or equal to N, and M = N-i. In the method, noise preprocessing is performed on N layers of PPI raw data at the time T, so that noise caused by equipment or interference of ground object noise, solar rays and the like in the radar scanning process is removed. Mapping the denoised PPI data, namely mapping the elevation angle of any other layer (marked as a j layer) to the elevation angle of an i layer, using the denoised PPI data of the i layer as spliced first layer data, using the mapping data of the denoised PPI data of the (i + 1) th layer as second layer data, and sequentially using the mapping data of the denoised PPI data of the N layer as M layer data, wherein M = N-i; for example, if N is 10 and i is 5, the spliced mapping data of the first layer is the original layer 5 data, and the spliced mapping data of the layer 5 is the mapping data obtained by mapping the original layer 10 data to the original layer 5. Fig. 2 is a schematic diagram illustrating a process of performing noise pre-processing on the N layers of PPI raw data in an embodiment of the present application, where the PPI data at the elevation angle of layer 1, the PPI data at the elevation angle of layer 2, … …, the PPI data at the elevation angle of layer i, … …, and the PPI data at the elevation angle of layer i are image-denoised, and the denoised image is mapped to the influence range of the elevation angle of layer i. That is, in the embodiment described in the present application, single-layer PPI data is fused, the mapping data of the i-th layer at all times is spliced to the channel of the mapping data of the M-th layer, and model training is performed by using the spliced data as an input of a prediction generation network, because several frames near the moment before the task learning of prediction predict several frames in the future, the input of the network in the embodiment of the present application is input together at several times, that is, after the fusion of the multiple layers of PPI data corresponding to each time in the target number of times, the fusion results at all times are input together into the network model for training. Therefore, the problem of insufficient data information of PPI at a single elevation angle is solved, the quality of network learning is improved, and the prediction accuracy of the nowcasting is improved.
Specifically, preprocessing for removing fixed position noise can be performed on the N layers of PPI raw data by adopting a mazeflo distance, so as to obtain a preprocessing result; and screening out a value which is more than or equal to 10dBz from the preprocessing result to obtain N layers of PPI denoising data. In this case, dBz is a unit of radar echo intensity, which is a unit describing a ratio of radar reflectivity with a specific parameter Z (radar reflectivity) value. Fixed position noise is removed by adopting a horse-type distance, a value smaller than 10dBz is removed to remove non-fixed position noise, and a remained de-noised radar echo map is cleaner and less-interference data, so that subsequent network learning is facilitated, and the accuracy of prediction is improved.
In an embodiment of the present application, when mapping an arbitrary layer elevation angle in the N layers of PPI de-noising data to an ith layer elevation angle, a first candidate point in the jth layer elevation angle in the same horizontal direction as a fixed point of the ith layer elevation angle and a second candidate point in the same vertical direction may be determined, where the jth layer is an arbitrary non-i layer in the N layers; calculating a candidate point distance between the fixed point and a first candidate point, and determining an influence point related to the fixed point in the j-th layer elevation according to the candidate point distance and a distance threshold; and determining the influence points of all fixed points in the ith layer elevation angle, and completing the mapping of PPI de-noising data in the jth layer elevation angle to the ith layer elevation angle. Here, the selection method for mapping the elevation angle of any other layer (denoted as j layer) to the elevation angle of the i-th layer may be implemented by using a theta2ppi function, and for a meteorological environment at a certain altitude, the affected environment is preferentially the environment at the same altitude and the influence of the nearby altitude of the convective motion, so that the theta2ppi function selects different distances at the same altitude and the values of the nearby altitude corresponding to the same altitude as a set of the influence points of the elevation angle of the j layer with respect to the i-th layer, where R (L, R are horizontal and vertical coordinate sampling points), and finally, the R × L image of the elevation angle of the j layer is mapped to the R × L image with the largest corresponding influence on each point of the i-th layer. Specifically, the method comprises the following steps: as shown in fig. 3, for the fixed point a at the i-th elevation angle, candidate points are determined based on the range and spatial influence on the convective motion, that is, the candidate points at the time of determining the vertical convective motion and the candidate points for the horizontal convective motion, the first candidate point in the same horizontal direction is the point (candidate point B) having the largest influence in the horizontal convective motion, the second candidate point in the same vertical direction is the point (candidate point C) having the largest influence on the vertical convective motion, the candidate points having a large influence on the point a are determined to be B and C, whether the influence of the candidate point B is large or the influence of the point C is large is determined by determining whether the weather is mainly advection or mainly convection, and the point having the largest influence is determined to be the influence point a.
According to the above embodiment, when the distance between the candidate points is smaller than the distance threshold, the first candidate point is used as an influence point about the fixed point in the elevation angle of the j-th layer; and when the candidate point distance is greater than or equal to a distance threshold value, taking the second candidate point as an influence point in the j-th layer elevation angle relative to the fixed point. Here, with reference to fig. 3, the distance between | AB | two points is calculated to determine whether the point of influence is B or C, and if the distance exceeds a distance threshold (e.g., 6000 m), the point C is selected as the point of influence, and if the distance is less than the distance threshold, the point B is selected as the point of influence. Specifically, the logic for using the theta2ppi function is calculated as follows:
the height h of the space in which the point A is located is:
the distance from the point B to the sampling point of the radar center is as follows:
the actual distance between the points A and B is as follows:
c point is from sampling point at radar center
Wherein,
the fixed point a representing the ith elevation angle is the distance from the radar scan center,
represents the sine value of the ith elevation angle,
represents the sine value of the jth elevation angle,
represents the cosine value of the ith elevation angle,
represents the cosine of the jth elevation angle,
representing the interval distance of each sampling point;
if it is
Then select the C point, i.e. the jth elevation angle corresponding to the scan angle
Each sampling point is an influence point; if it is
Then select point B, i.e. the jth elevation angle corresponding to the scan angle
The sampling points are impact points.
In an embodiment of the present application, when mapping of the PPI de-noised data in the jth layer elevation angle to the ith layer elevation angle is completed, the corresponding PPI de-noised data of the impact point of the fixed point in the jth layer elevation angle may be queried in the querying step, and the queried data is used as the PPI data of the fixed point in the ith layer elevation angle; determining the influence points of all fixed points in the ith layer elevation angle, inquiring each influence point according to the inquiry steps, and completing the mapping of PPI de-noising data in the jth layer elevation angle to the ith layer elevation angle. After determining an influence point of a certain fixed point A1 in the i-th layer elevation angle, the influence point being a certain point B1 in the j-th layer elevation angle, querying PPI de-noising data corresponding to the influence point B1 in the j-th layer elevation angle, using the queried PPI de-noising data of B1 as PPI data of the fixed point A1 in the i-th layer elevation angle, and completing the mapping of the PPI data of the point B1 in the j-th layer elevation angle to the i-th layer elevation angle. And analogizing in turn, inquiring the PPI de-noising data corresponding to the influence points of all the fixed points in the ith layer in the elevation angle of the jth layer, and using the inquired PPI de-noising data as the PPI data of the corresponding fixed points in the elevation angle of the ith layer, thereby completing the mapping from the PPI de-noising data in the elevation angle of the jth layer to the elevation angle of the ith layer.
In an embodiment of the present application, as shown in fig. 4, T =0 original PPI data, T =1 original PPI data, T =2 original PPI data, T =3 original PPI data, and T =4 original PPI data are read, the original PPI data at each time is preprocessed to obtain corresponding ith layer PPI data, the ith layer PPI data at all times are spliced and input to a deep learning image prediction network of a roll convolution, and the ith layer PPI data predicted by T =5, the ith layer PPI data predicted by T =6, the ith layer PPI data predicted by T =7, the ith layer PPI data predicted by T =8, and the ith layer PPI data predicted by T =9 are output. In the application, the data form in the deep learning network is improved from CAPPI to PPI data after mapping processing, and the PPI is the original data without uncertainty caused by data certainty and frame interpolation, so that the effect of converting network learning information quantity and generated PPI data into CAPPI is better than that of training by directly using CAPPI; the PPI is adopted during training data, so that prediction of a prediction product model is not limited to learning data, data processing can be further performed on the predicted PPI, the PPI can be processed into data formats of various display modes such as CAPPI, CR, RHI and VCS, the data format of the training data is not limited to one, and the training data can be displayed through various display modes, so that the problem of analysis limitation caused by the single data format of the training data is solved. In addition, the multi-layer PPI data are fused into the training of the single-layer PPI data, so that the model learns the multi-dimensional space-time data, and the effect of generating the image by the network is more stable.
Furthermore, a computer-readable medium is provided, on which computer-readable instructions are stored, and the computer-readable instructions are executable by a processor to implement the foregoing prediction method based on the nowcast of PPI data.
In correspondence with the method described above, the present application also provides a terminal, which includes modules or units capable of executing the steps of the method described in fig. 1 or each embodiment, and these modules or units can be implemented by hardware, software or a combination of hardware and software, and this application is not limited thereto. For example, in an embodiment of the present application, there is also provided an apparatus for prediction based on the nowcasting of PPI data, the apparatus comprising:
one or more processors; and
a memory storing computer readable instructions that, when executed, cause the processor to perform the operations of the method as previously described.
For example, the computer readable instructions, when executed, cause the one or more processors to:
determining multilayer PPI original data at each moment in the target quantity moments;
carrying out mapping processing of a target layer elevation angle on the multi-layer PPI original data at each moment to obtain mapping data corresponding to each moment;
splicing channels of mapping data from the first layer to the last layer at all times to serve as network training data;
and inputting the network training data into network convolution for sequence learning to obtain PPI images of the forecast target number of frames.
Fig. 5 illustrates a schematic structural diagram of an apparatus for prediction based on the nowcasting of PPI data according to another aspect of the present application, the apparatus comprising: the device comprises a determining device 11, a mapping device 12, a splicing device 13 and a forecasting device 14, wherein the determining device 11 is used for determining the multi-layer PPI original data at each moment in a target number of moments; the mapping device 12 is configured to perform mapping processing on the target layer elevation angle on the multi-layer PPI raw data at each time to obtain mapping data corresponding to each time; the splicing device 13 is configured to splice channels of mapping data from the first layer to the last layer at all times, and use the channels as network training data; the forecasting device 14 is configured to input the network training data into a network convolution for sequence learning, so as to obtain a PPI image with a forecast target number of frames.
It should be noted that the content executed by the determining device 11, the mapping device 12, the splicing device 13 and the forecasting device 14 is the same as or corresponding to the content in the above steps S11, S12, S13 and S14, respectively, and for the sake of brevity, the description thereof is omitted.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
It should be noted that the present application may be implemented in software and/or a combination of software and hardware, for example, implemented using Application Specific Integrated Circuits (ASICs), general purpose computers or any other similar hardware devices. In one embodiment, the software programs of the present application may be executed by a processor to implement the steps or functions described above. Likewise, the software programs (including associated data structures) of the present application may be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Additionally, some of the steps or functions of the present application may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.
In addition, some of the present application may be implemented as a computer program product, such as computer program instructions, which when executed by a computer, may invoke or provide methods and/or techniques in accordance with the present application through the operation of the computer. Program instructions which invoke the methods of the present application may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the present application comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or a solution according to the aforementioned embodiments of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. The terms first, second, etc. are used to denote names, but not any particular order.
Claims (10)
1. A method of prediction based on nowcasting of PPI data, the method comprising:
determining multilayer PPI original data at each moment in the target quantity moments;
carrying out mapping processing of a target layer elevation angle on the multi-layer PPI original data at each moment to obtain mapping data corresponding to each moment;
splicing channels of mapping data from the first layer to the last layer at all times to serve as network training data;
and inputting the network training data into network convolution for sequence learning to obtain PPI images of the forecast target number of frames.
2. The method of claim 1, wherein determining a plurality of layers of PPI raw data for each of a target number of time instants comprises:
reading radar scanning data of N elevation angles at each moment in target quantity moments, wherein the radar scanning data comprise images with sizes determined by radar azimuth angle scanning frequency and equidistant sampling points on rays, and N is a positive integer;
and determining N layers of PPI original data at each moment in the target quantity moment according to the radar scanning data of N elevation angles at each moment in the target quantity moment.
3. The method of claim 2, wherein performing a mapping process of a target layer elevation angle on the multi-layer PPI raw data at each time to obtain mapping data corresponding to each time comprises:
carrying out noise pretreatment on the N layers of PPI original data to obtain N layers of PPI de-noising data;
mapping any layer elevation angle in the N layers of PPI de-noising data to an i layer elevation angle to obtain mapping data which takes the i layer as a first layer and takes M layers as a last layer, wherein i is more than or equal to 1 and less than or equal to N, and M = N-i;
splicing the channels of the mapping data from the first layer to the last layer at all times, wherein the splicing comprises the following steps:
and splicing the mapping data of the ith layer to the mapping data channel of the Mth layer at all the moments.
4. The method of claim 3, wherein noise preprocessing the N layers of PPI raw data to obtain N layers of PPI de-noised data comprises:
preprocessing the N layers of PPI original data by adopting a horse-type distance to remove fixed position noise to obtain a preprocessing result;
and screening out a value which is more than or equal to 10dBz from the preprocessing result to obtain N layers of PPI denoising data.
5. The method of claim 3, wherein mapping any layer elevation angle in the N layer PPI denoised data to an i layer elevation angle comprises:
determining a first candidate point in the j layer elevation angle in the same horizontal direction with a fixed point of the i layer elevation angle and a second candidate point in the same vertical direction, wherein the j layer is any non-i layer in the N layers;
calculating a candidate point distance between the fixed point and a first candidate point, and determining an influence point related to the fixed point in the j-th layer elevation according to the candidate point distance and a distance threshold;
and determining the influence points of all fixed points in the ith layer elevation angle, and completing the mapping of PPI de-noising data in the jth layer elevation angle to the ith layer elevation angle.
6. The method of claim 5, wherein determining the points of influence of all fixed points in the i layer elevation angle, and mapping PPI denoised data in the j layer elevation angle into the i layer elevation angle is performed, comprises:
inquiring PPI de-noising data corresponding to the influence point of the fixed point in the j layer elevation angle in an inquiring step, and taking the inquired data as the PPI data of the fixed point in the i layer elevation angle;
and determining the influence points of all fixed points in the i-th layer elevation angle, inquiring each influence point according to the inquiry steps, and completing the mapping of PPI de-noising data in the j-th layer elevation angle to the i-th layer elevation angle.
7. The method of claim 5, wherein determining the point of influence in the elevation of layer j with respect to the fixed point based on the candidate point distance and a distance threshold comprises:
when the candidate point distance is smaller than a distance threshold, taking the first candidate point as an influence point in the j layer elevation angle relative to the fixed point;
and when the candidate point distance is greater than or equal to a distance threshold value, taking the second candidate point as an influence point in the j-th layer elevation angle relative to the fixed point.
8. A device for prediction based on the nowcast of PPI data, characterized in that the device comprises:
determining means for determining a plurality of layers of PPI raw data at each of a target number of times;
the mapping device is used for mapping the target layer elevation angle of the multilayer PPI original data at each moment to obtain mapping data corresponding to each moment;
the splicing device is used for splicing the channels of the mapping data from the first layer to the last layer at all times to serve as network training data;
and the forecasting device is used for inputting the network training data into network convolution for sequence learning to obtain PPI images of forecasting target number of frames.
9. A device for prediction based on the nowcast of PPI data, characterized in that the device comprises:
one or more processors; and
a memory storing computer readable instructions that, when executed, cause the processor to perform the operations of the method of any of claims 1 to 7.
10. A computer readable medium having computer readable instructions stored thereon which are executable by a processor to implement the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010871064.4A CN111736157B (en) | 2020-08-26 | 2020-08-26 | PPI data-based prediction method and device for nowcasting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010871064.4A CN111736157B (en) | 2020-08-26 | 2020-08-26 | PPI data-based prediction method and device for nowcasting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111736157A true CN111736157A (en) | 2020-10-02 |
| CN111736157B CN111736157B (en) | 2021-01-05 |
Family
ID=72658870
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010871064.4A Expired - Fee Related CN111736157B (en) | 2020-08-26 | 2020-08-26 | PPI data-based prediction method and device for nowcasting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111736157B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117153000A (en) * | 2023-11-01 | 2023-12-01 | 天宇航空数据科技(合肥)有限责任公司 | Method and system for analyzing influence of approach and departure routes based on three-dimensional radar data |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106886023A (en) * | 2017-02-27 | 2017-06-23 | 中国人民解放军理工大学 | A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks |
| US20170357029A1 (en) * | 2016-06-14 | 2017-12-14 | The Climate Corporation | Supervised neural network to predict unlabeled rain rates |
| CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
| CN108508505A (en) * | 2018-02-05 | 2018-09-07 | 南京云思创智信息科技有限公司 | Heavy showers and thunderstorm forecasting procedure based on multiple dimensioned convolutional neural networks and system |
| CN109143182A (en) * | 2018-09-10 | 2019-01-04 | 南京莱斯电子设备有限公司 | A kind of multiple dimensioned radar block plan immediate processing method |
| CN110135654A (en) * | 2019-05-24 | 2019-08-16 | 北京百度网讯科技有限公司 | Method and apparatus for predicting strong convective weather |
| CN110991690A (en) * | 2019-10-17 | 2020-04-10 | 宁波大学 | Multi-time wind speed prediction method based on deep convolutional neural network |
-
2020
- 2020-08-26 CN CN202010871064.4A patent/CN111736157B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170357029A1 (en) * | 2016-06-14 | 2017-12-14 | The Climate Corporation | Supervised neural network to predict unlabeled rain rates |
| CN106886023A (en) * | 2017-02-27 | 2017-06-23 | 中国人民解放军理工大学 | A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks |
| CN107703564A (en) * | 2017-10-13 | 2018-02-16 | 中国科学院深圳先进技术研究院 | A kind of precipitation predicting method, system and electronic equipment |
| CN108508505A (en) * | 2018-02-05 | 2018-09-07 | 南京云思创智信息科技有限公司 | Heavy showers and thunderstorm forecasting procedure based on multiple dimensioned convolutional neural networks and system |
| CN109143182A (en) * | 2018-09-10 | 2019-01-04 | 南京莱斯电子设备有限公司 | A kind of multiple dimensioned radar block plan immediate processing method |
| CN110135654A (en) * | 2019-05-24 | 2019-08-16 | 北京百度网讯科技有限公司 | Method and apparatus for predicting strong convective weather |
| CN110991690A (en) * | 2019-10-17 | 2020-04-10 | 宁波大学 | Multi-time wind speed prediction method based on deep convolutional neural network |
Non-Patent Citations (2)
| Title |
|---|
| 庞博等: "基于改进全卷积神经网络的高分遥感影像不透水面提取制图", 《遥感信息》 * |
| 胡琦等: "基于多普勒激光雷达的风场预测", 《激光与红外》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117153000A (en) * | 2023-11-01 | 2023-12-01 | 天宇航空数据科技(合肥)有限责任公司 | Method and system for analyzing influence of approach and departure routes based on three-dimensional radar data |
| CN117153000B (en) * | 2023-11-01 | 2024-02-02 | 天宇航空数据科技(合肥)有限责任公司 | Method and system for analyzing influence of approach and departure routes based on three-dimensional radar data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111736157B (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3805794B1 (en) | Analysis of sar data using neural networks | |
| CN113705478B (en) | Mangrove single wood target detection method based on improved YOLOv5 | |
| CN112766221B (en) | Ship direction and position multitasking-based SAR image ship target detection method | |
| CN111476159B (en) | Method and device for training and detecting detection model based on double-angle regression | |
| CN103971127B (en) | Forward-looking radar imaging sea-surface target key point detection and recognition method | |
| JP6756889B1 (en) | Vortex detector, vortex detection method, program and trained model | |
| KR102440355B1 (en) | Deep Learning-based Ocean Cluster Data Measurement System Using Sea Level Wave Reflectance | |
| CN110888147B (en) | Scanning laser radar detection data gridding visualization method | |
| CN115979972B (en) | Real-time monitoring method and system for hyperspectral of crude oil film on sea surface | |
| CN114708257B (en) | SAR moving ship target detection method and device | |
| CN114565824B (en) | Single-stage rotating ship detection method based on full convolution network | |
| CN111736157B (en) | PPI data-based prediction method and device for nowcasting | |
| CN112053375A (en) | Method and equipment for predicting prediction based on improved network convolution model | |
| CN116229419B (en) | Pedestrian detection method and device | |
| CN114488070B (en) | Radar echo extrapolation method and device based on deep learning model | |
| CN116129320A (en) | Target detection method, system and equipment based on video SAR | |
| CN114594477A (en) | Echo domain-based satellite-borne SAR (synthetic Aperture Radar) marine target signal detection method and device | |
| CN114881763A (en) | Method, device, equipment and medium for post-loan supervision of aquaculture | |
| CN115015957A (en) | Laser reflection tomography quality evaluation method based on modulation transfer function | |
| CN113762271A (en) | SAR image semantic segmentation method and system based on irregular convolution kernel neural network model | |
| Narayanan et al. | Classification of SHOALS 3000 bathymetric LiDAR signals using decision tree and ensemble techniques | |
| CN110502979A (en) | A kind of laser radar waveform Modulation recognition method based on decision tree | |
| CN113009470B (en) | Processing method, system, device and medium for target situation characteristic data | |
| Yuan et al. | Water Mapping with SAR and Optical Data Cube | |
| Wang et al. | Real-time extraction of water surface boundary using shipborne radar |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210105 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |