CN113111936B - Satellite data fusion air temperature estimation method - Google Patents
Satellite data fusion air temperature estimation method Download PDFInfo
- Publication number
- CN113111936B CN113111936B CN202110373975.9A CN202110373975A CN113111936B CN 113111936 B CN113111936 B CN 113111936B CN 202110373975 A CN202110373975 A CN 202110373975A CN 113111936 B CN113111936 B CN 113111936B
- Authority
- CN
- China
- Prior art keywords
- air temperature
- data
- satellite
- spatial resolution
- temperature data
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- 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)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides a satellite data fusion air temperature estimation method, which comprises the following steps: the method comprises the following steps: preprocessing the acquired temperature data of the geostationary satellite and the temperature data of the polar orbit satellite; step two: establishing an air temperature difference dictionary by utilizing the preprocessed static satellite air temperature data and polar orbit satellite air temperature data; step three: performing super-resolution reconstruction on difference data of the static satellite air temperature data at the reference moment and the static satellite air temperature data at the prediction moment by using an air temperature difference dictionary to obtain reconstructed air temperature difference value data with high spatial resolution; step four: adding polar orbit satellite air temperature data at the reference moment and the obtained air temperature difference value data with high spatial resolution to obtain a satellite air temperature product with high spatial resolution at the predicted moment; step five: and continuously generating high-spatial-resolution satellite air temperature data with corresponding high temporal resolution according to the temporal resolution of the static satellite air temperature data. The data obtained by the method of the invention has higher accuracy and stability.
Description
Technical Field
The invention relates to the technical field of satellite weather, in particular to a high-space-time resolution satellite data fusion air temperature estimation method for stationary and polar orbit weather satellite data fusion.
Background
At present, temperature data is mainly obtained through a station, an interpolation method is mostly adopted for estimating the large-range temperature data, remote sensing has the characteristics of macroscopic view, dynamic property, convenience, economy and periodicity, timely information can be provided in a large range, the station has incomparable advantages, and a way is provided for obtaining the large-range temperature data.
However, the optical satellite sensor of a single platform is difficult to acquire temperature data with high time and high spatial resolution, so that popularization and application of remote sensing temperature data are limited. The air temperature obtained by the station does not have the characteristic of large-range continuity, and the time-space contradiction between the geostationary satellite and the polar orbit satellite brings great difficulty in obtaining high-time high-spatial resolution air temperature data.
Therefore, the combination of the high spatial characteristics of polar orbit satellites and the high temporal characteristics of geostationary satellites becomes an important means for obtaining satellite remote sensing products with high temporal and spatial resolutions. How to better fuse static and polar orbit meteorological satellite temperature data and improve the accuracy and the stability of the data is a problem to be researched urgently.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides a temperature estimation method for satellite data fusion with high space-time resolution, which is used for fusing stationary and polar orbit meteorological satellite data.
A satellite data fusion air temperature estimation method comprises the following steps:
the method comprises the following steps: preprocessing the acquired static satellite air temperature data and polar orbit satellite air temperature data;
step two: establishing an air temperature difference dictionary by utilizing the preprocessed stationary satellite air temperature data and polar orbit satellite air temperature data;
step three: performing super-resolution reconstruction on difference data of the static satellite air temperature data at the reference moment and the static satellite air temperature data at the prediction moment by using an air temperature difference dictionary to obtain reconstructed air temperature difference value data with high spatial resolution;
step four: adding polar orbit satellite air temperature data at the reference moment and the obtained air temperature difference value data with high spatial resolution to obtain a satellite air temperature product with high spatial resolution at the predicted moment;
step five: and continuously generating high-spatial-resolution satellite air temperature data with corresponding high temporal resolution according to the temporal resolution of the static satellite air temperature data.
Further, in the method as described above, the preprocessing in the first step includes:
step 11: spatio-temporal matching
Performing space-time matching on the acquired temperature data of the geostationary meteorological satellite and the temperature data of the polar orbiting meteorological satellite to obtain temperature data observed in the same region at the same moment, namely the polar orbiting satellite temperature data and the geostationary satellite temperature data after the space-time matching;
step 12: data correction
And correcting the polar orbit satellite air temperature data and the static satellite air temperature data after the time-space matching by using the station data to obtain the corrected polar orbit satellite air temperature data and static satellite air temperature data.
Further, in the method as described above, the second step includes:
step 21: establishment of gas temperature difference value data set
Respectively calculating difference values of the corrected polar orbit satellite air temperature data and the corrected geostationary satellite air temperature data at the same moment in different periods, and dividing and sorting to finally obtain a polar orbit satellite air temperature difference value data set and a geostationary satellite air temperature difference value data set;
step 22: generation of high and low spatial resolution dictionaries
And training a high spatial resolution dictionary by using the polar orbit satellite gas temperature difference value data set, training a low spatial resolution dictionary by using the stationary satellite gas temperature difference value data set, and finally obtaining the high spatial resolution dictionary and the low spatial resolution dictionary.
Further, as in the method described above, the step 22 includes:
step 221: according to the obtained high spatial resolution dictionary and the low spatial resolution dictionary, carrying out blocking operation on a data set sample according to set parameters to respectively obtain a high spatial resolution data block and a low spatial resolution data block, then respectively removing the data blocks with smaller threshold values, and finally obtaining a required high spatial resolution training sample and a required low spatial resolution training sample;
step 222: and respectively carrying out combined dictionary construction on the high spatial resolution training sample and the low spatial resolution training sample by utilizing a K-SVD redundant dictionary construction algorithm according to the obtained high spatial resolution training sample and the low spatial resolution training sample to respectively obtain a high spatial resolution dictionary and a low spatial resolution dictionary.
Further, in the method described above, the third step specifically includes the following steps:
step 1: according to the obtained low spatial resolution dictionary D l And a data Y of the temperature difference between the gas of the stationary satellite fy4a at a reference time and a predicted time l Using the calculation formula X l ＝D l + ·Y l Obtaining corresponding sparse coefficient X l ；
Step 2: according to the obtained high spatial resolution dictionary D h And sparse coefficient Xl, using calculation formula Y h ＝D h ·X l Obtaining high-spatial-resolution air temperature difference value data Y of the reference time and the prediction time h 。
Further, as in the method described above, the fourth step includes:
according to the high spatial resolution air temperature difference value data Y of the obtained reference time and the prediction time h And polar satellite temperature data y at a reference time b Using the calculation formula y p ＝y b +Y h Obtaining high spatial resolution air temperature data y at the predicted moment p 。
Has the advantages that:
on one hand, the method corrects a historical data set, on the other hand, the method uses the polar orbit satellite and the geostationary satellite gas temperature difference value data as training samples, can better learn the difference between the polar orbit satellite and the geostationary satellite, and does not use the traditional resampling data as the training samples, thereby improving the accuracy and the stability of the temperature estimation.
The method provided by the invention can be used for fusing the air temperature data of the stationary satellite and the polar orbit satellite to obtain a satellite remote sensing air temperature product with high time and high spatial resolution, and the product has higher accuracy and stability, thereby improving the application of the satellite remote sensing data.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
fig. 2 is an example reference time FY4A temperature map;
fig. 3 is an example reference time FY3D temperature map;
fig. 4 is an example predicted time FY4A temperature map;
FIG. 5 is a temperature map after merging of predicted times according to an embodiment;
FIG. 6 (a) is a scatter diagram comparing the temperature data at the predicted time FY4A with the measured data at the station in the embodiment;
FIG. 6 (b) is a scatter diagram comparing air temperature data fused with station actual measurement data at the predicted time in the embodiment;
FIG. 7 is a graph of the error trend before and after fusion within 24 hours using the method of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the present invention clearer and more complete, the technical solutions of the present invention are described below clearly, and it is obvious that the described embodiments are some, not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
As shown in FIG. 1, the method for estimating the air temperature of the high space-time resolution satellite data fusion of the static and polar orbiting meteorological satellite air temperature data fusion provided by the invention comprises the following steps:
the method comprises the following steps: pre-processing of air temperature data
Step 11: spatio-temporal matching
And performing space-time matching on the acquired air temperature data of the geostationary meteorological satellite and the air temperature data of the polar orbiting meteorological satellite to obtain the air temperature data observed in the same region at the same moment, namely the polar orbiting satellite air temperature data and the geostationary satellite air temperature data after the space-time matching.
Step 12: data correction
Because the temperature data obtained by the polar orbit satellite and the stationary satellite are different greatly, in order to improve the fusion precision, the historical data set needs to be corrected by station data to obtain the corrected temperature data of the polar orbit satellite and the stationary satellite.
Step two: establishment of polar orbit satellite and stationary satellite temperature difference value dictionary
Step 21: establishment of gas temperature difference value data set
And establishing a data set by using the matched and corrected air temperature data, wherein the air temperature difference value data set needs to be established according to the used fusion method. For better fusion effect, the data set is divided into high and low resolution ratio difference data at different periods and at the same time, difference value calculation is carried out, and finally a polar orbit satellite gas temperature difference value data set (namely, a high spatial resolution gas temperature difference value data set) and a stationary satellite gas temperature difference value data set (namely, a low spatial resolution gas temperature difference value data set) are obtained.
Specifically, the method comprises the following steps:
step 21: and grouping according to the matched and corrected polar orbit satellite temperature data and the static satellite temperature data according to the spatial resolution to obtain a high spatial resolution temperature data group and a low spatial resolution temperature data group.
Step 22: and according to the obtained high-spatial-resolution air temperature data group, performing difference calculation according to the same time in different periods (for example, the high-spatial-resolution air temperature data at the time of 06: 7/1/00 and 06: 7/2/00 is subjected to difference calculation, and the high-spatial-resolution air temperature data at the time of 18: 7/1/00 and 18: 7/2/00 is subjected to difference calculation), so as to obtain a high-spatial-resolution air temperature difference value data set.
Step 23: and according to the obtained low spatial resolution air temperature data group, calculating difference values according to the same time in different periods (for example, difference values are calculated for low spatial resolution air temperature data at times 06.
Step 24: according to the obtained gas temperature difference value data set with high and low spatial resolution, carrying out blocking operation on a data set sample according to set parameters to respectively obtain data blocks with high and low spatial resolution, then eliminating the data blocks with smaller threshold values, and finally obtaining the required training sample with high and low spatial resolution.
Step 25: and performing joint dictionary construction on the high and low spatial resolution training samples by utilizing a K-SVD redundant dictionary construction algorithm according to the obtained high and low spatial resolution training samples to obtain a high spatial resolution dictionary and a low spatial resolution dictionary.
Step three: fusing the air temperature data through a calculation formula; and performing super-resolution reconstruction on difference data of the static satellite air temperature data at the reference moment and the static satellite air temperature data at the prediction moment by using the obtained high-low spatial resolution dictionary to obtain reconstructed air temperature difference value data with high spatial resolution.
Specifically, the method comprises the following steps:
step 31: according to the obtained low spatial resolution dictionary Dl and the data Y of the gas temperature difference value of the stationary satellite fy4a at a reference moment and a predicted moment l Using the calculation formula X l ＝D l + ·Y l Obtaining corresponding sparse coefficient X l 。
Step 32: according to the obtained high spatial resolution dictionary D h And a sparse coefficient X l Using the calculation formula Y h ＝D h ·X l Obtaining high-spatial-resolution air temperature difference value data Y of the reference time and the prediction time h 。
Step 33: according to the high spatial resolution air temperature difference value data Y of the obtained reference time and the prediction time h ，
And polar orbit satellite temperature data y of reference time b Using the calculation formula y p ＝y b +Y h Obtaining high spatial resolution air temperature data y at the predicted moment p 。
Step four: creation of high temporal and spatial resolution air temperature data
By using the above steps, high spatial resolution satellite air temperature data with a corresponding high temporal resolution can be continuously generated based on the temporal resolution of the geostationary satellite air temperature data.
Example (b):
air temperature data of the Hunan area are obtained by using a high spatial resolution wind cloud No. three D polar orbit meteorological satellite FY3D (250 m, twice a day) product and a low spatial resolution wind cloud No. four A static meteorological satellite FY4A (4000 m, once per hour) product. The method is adopted to carry out multi-source satellite data fusion and generate a new satellite air temperature product with high time (once in 1 hour) and high spatial resolution (250 m).
The algorithm has the following effects:
the reference time is 8 months and 5 days in 2019 (06): FY3D, FY a air temperature data and predicted time of 00 are 8 months, 5 days 07 in 2019: and fusing the 00FY4A air temperature data to obtain a satellite air temperature product with high spatial resolution at the predicted moment.
Fig. 2 and 3 are respectively a geostationary satellite temperature map and a polar satellite temperature map at a reference time, and comparing the two maps shows that even at the same time, temperatures obtained by different satellites still have great difference, which brings certain difficulty to fusion. Fig. 4 and 5 are respectively a geostationary satellite temperature map at the predicted time and a fused high spatial resolution temperature map, and comparing the two maps shows that the contours of the temperatures are basically similar, but the fused temperatures are not smooth enough in detail. Comparing the air temperature data of the two graphs with the actual measurement air temperatures of 97 stations in the Hunan area, calculating the errors of the two graphs to obtain a graph 6 (a) and a graph 6 (b), wherein the comparison of the two error scatter graphs shows that the correlation is improved after fusion, the correlation is improved from 0.405 to 0.553, and the deviation degree and the deviation are basically unchanged; on the other hand, the fusion improves the space-time resolution, and certain errors are inevitably brought, but the root mean square error is only increased to 2.31 from 2.29. It can be seen that the fusion method has a good effect on the estimation of the air temperature with high space-time resolution.
FIG. 7 is a graph of the error trend before and after fusion in 24 hours in 8 months in 2019; estimating air temperature hour by using the method of the invention to obtain air temperature data within 24 hours, and calculating the Root Mean Square Error (RMSE) between each moment and the air temperature data of the station in the Hunan region to obtain a fused error curve in the figure 7; and (3) calculating the root mean square error by using the air temperature data of the geostationary satellite fy4a at the corresponding moment and the air temperature data of the station in the Hunan region to obtain an error curve before fusion in the graph 7. Comparing the two curves, the variation trends of the air temperatures are basically consistent, and the root mean square error data of the estimated air temperatures are more uniformly distributed near the root mean square error data of the air temperatures of the geostationary satellite, which shows that the estimation result has better stability; within 24 hours, the mean values of the root mean square error of the estimated air temperature and the mean value of the root mean square error of the air temperature of the geostationary satellite are respectively 2.39 and 2.16, which shows that the estimation result has better accuracy.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (3)
1. A satellite data fusion air temperature estimation method is characterized by comprising the following steps:
the method comprises the following steps: preprocessing the acquired static satellite air temperature data and polar orbit satellite air temperature data;
step two: establishing an air temperature difference dictionary by utilizing the preprocessed static satellite air temperature data and polar orbit satellite air temperature data;
step three: performing super-resolution reconstruction on difference data of the static satellite air temperature data at the reference moment and the static satellite air temperature data at the prediction moment by using an air temperature difference dictionary to obtain reconstructed air temperature difference value data with high spatial resolution;
step four: adding polar orbit satellite air temperature data at the reference moment and the obtained air temperature difference value data with high spatial resolution to obtain a satellite air temperature product with high spatial resolution at the predicted moment;
step five: continuously generating high-spatial-resolution satellite air temperature data with corresponding high temporal resolution according to the time resolution of the geostationary satellite air temperature data;
the pretreatment in the first step comprises the following steps:
step 11: spatio-temporal matching
Performing space-time matching on the acquired temperature data of the geostationary meteorological satellite and the temperature data of the polar orbiting meteorological satellite to obtain temperature data observed in the same region at the same moment, namely the polar orbiting satellite temperature data and the geostationary satellite temperature data after the space-time matching;
step 12: data correction
Correcting the polar orbit satellite air temperature data and the static satellite air temperature data after the time-space matching by using station data to obtain corrected polar orbit satellite air temperature data and static satellite air temperature data;
the third step specifically comprises the following steps:
step 1: according to the obtained low spatial resolution dictionary D l And a data Y of the temperature difference between the gas of the stationary satellite fy4a at a reference time and a predicted time l Using the calculation formula X l ＝D l + ·Y l Obtaining corresponding sparse coefficient X l ；D l + Representative matrix D l The generalized inverse matrix of (2);
and 2, step: according to the obtained high spatial resolution dictionary D h And a sparse coefficient X l Using the calculation formula Y h ＝D h ·X l Obtaining high-spatial-resolution air temperature difference value data Y of the reference time and the prediction time h ；
The fourth step comprises:
according to the high spatial resolution air temperature difference value data Y of the obtained reference time and the prediction time h And polar orbit satellite temperature data y at reference time b Using the formula y p ＝y b +Y h Obtaining high spatial resolution air temperature data y at the predicted moment p 。
2. The method according to claim 1, wherein the second step comprises:
step 21: establishment of gas temperature difference value data set
Respectively calculating difference values of the corrected polar orbit satellite air temperature data and the corrected geostationary satellite air temperature data at the same moment in different periods, and dividing and sorting to finally obtain a polar orbit satellite air temperature difference value data set and a geostationary satellite air temperature difference value data set;
step 22: generation of high and low spatial resolution dictionaries
And training a high spatial resolution dictionary by using the polar orbit satellite gas temperature difference value data set, training a low spatial resolution dictionary by using the geostationary satellite gas temperature difference value data set, and finally obtaining the high spatial resolution dictionary and the low spatial resolution dictionary.
3. The method of claim 2, wherein the step 22 comprises:
step 221: according to the obtained high spatial resolution dictionary and low spatial resolution dictionary, according to set parameters, carrying out blocking operation on a data set sample to respectively obtain a high spatial resolution data block and a low spatial resolution data block, then respectively removing the data blocks with smaller threshold values, and finally obtaining a required high spatial resolution training sample and a required low spatial resolution training sample;
step 222: and respectively carrying out combined dictionary construction on the high spatial resolution training sample and the low spatial resolution training sample by utilizing a K-SVD redundant dictionary construction algorithm according to the obtained high spatial resolution training sample and the low spatial resolution training sample to respectively obtain a high spatial resolution dictionary and a low spatial resolution dictionary.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110373975.9A CN113111936B (en) | 2021-04-07 | 2021-04-07 | Satellite data fusion air temperature estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110373975.9A CN113111936B (en) | 2021-04-07 | 2021-04-07 | Satellite data fusion air temperature estimation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113111936A CN113111936A (en) | 2021-07-13 |
| CN113111936B true CN113111936B (en) | 2022-10-18 |
Family
ID=76714551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110373975.9A Active CN113111936B (en) | 2021-04-07 | 2021-04-07 | Satellite data fusion air temperature estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113111936B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113792091A (en) * | 2021-11-15 | 2021-12-14 | 国家卫星气象中心(国家空间天气监测预警中心) | Sea surface temperature data normalization quality inspection method |
| CN116611587B (en) * | 2023-07-19 | 2023-11-07 | 中国气象局公共气象服务中心(国家预警信息发布中心) | Solar resource prediction method based on polar orbit-stationary satellite fusion technology |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109191408A (en) * | 2018-04-19 | 2019-01-11 | 中国气象局公共气象服务中心 | Rapid Circulation Ground Meteorological fusion method, device and server |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101191864B1 (en) * | 2011-03-07 | 2012-10-16 | 부경대학교 산학협력단 | System and method for verifying of radiance temperature using radiative transfer model |
| US9436784B2 (en) * | 2013-02-08 | 2016-09-06 | University Of Alaska Fairbanks | Validating and calibrating a forecast model |
| CN104636608A (en) * | 2015-01-30 | 2015-05-20 | 国家电网公司 | Direct assimilation method of MODIS satellite data |
| CN109375290B (en) * | 2018-10-16 | 2020-09-25 | 象谱信息产业有限公司 | Cross-sea bridge fog monitoring system based on machine learning and application method thereof |
| CN109635242A (en) * | 2018-11-29 | 2019-04-16 | 南京大学 | A kind of Remote Sensing day samming calculation method based on multi-time scale model |
| CN110954482B (en) * | 2019-12-02 | 2020-12-15 | 生态环境部卫星环境应用中心 | Atmospheric pollution gridding monitoring method based on static satellite and polar orbit satellite |
-
2021
- 2021-04-07 CN CN202110373975.9A patent/CN113111936B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109191408A (en) * | 2018-04-19 | 2019-01-11 | 中国气象局公共气象服务中心 | Rapid Circulation Ground Meteorological fusion method, device and server |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113111936A (en) | 2021-07-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113111936B (en) | Satellite data fusion air temperature estimation method | |
| CN112905560A (en) | Air pollution prediction method based on multi-source time-space big data deep fusion | |
| CN113936142A (en) | Rainfall approach forecasting method and device based on deep learning | |
| CN104636608A (en) | Direct assimilation method of MODIS satellite data | |
| CN113344149B (en) | PM2.5 hourly prediction method based on neural network | |
| CN115113301B (en) | Emergency short-term forecasting method and system based on multi-source data fusion | |
| CN116468730B (en) | Aerial Insulator Image Defect Detection Method Based on YOLOv5 Algorithm | |
| CN114331842B (en) | DEM super-resolution reconstruction method combining topographic features | |
| CN114019579A (en) | High-space-time resolution near-surface air temperature reconstruction method, system and equipment | |
| Ma et al. | FY-3A/MERSI precipitable water vapor reconstruction and calibration using multi-source observation data based on a generalized regression neural network | |
| Ruiz-Arias et al. | Direct normal irradiance modeling: Evaluating the impact on accuracy of worldwide gridded aerosol databases | |
| El Alani et al. | Short term solar irradiance forecasting using artificial neural network for a semi-arid climate in Morocco | |
| CN113984198B (en) | Shortwave radiation prediction method and system based on convolutional neural network | |
| CN108896456B (en) | Aerosol extinction coefficient inversion method based on feedback type RBF neural network | |
| Cristaldi et al. | A hybrid approach for solar radiation and photovoltaic power short-term forecast | |
| CN117113828A (en) | Numerical forecast correction method based on ship-based navigation observation | |
| CN110212591B (en) | Distributed photovoltaic irradiance measurement stationing method based on compressive sensing technology | |
| CN115952743A (en) | Multi-source precipitation data collaborative downscaling method and system coupled with random forest and HASM | |
| Xia et al. | A Deep Learning Method Integrating Multi-source Data for ECMWF Forcasting Products Correction | |
| CN111814855B (en) | Global ionospheric total electron content prediction method based on residual seq2seq neural network | |
| CN114169215B (en) | Surface temperature inversion method coupling remote sensing and regional meteorological model | |
| Koubli et al. | Inference of missing PV monitoring data using neural networks | |
| CN115993668B (en) | Polynomial correction and neural network-based PWV reconstruction method and system | |
| Ma et al. | All-weather precipitable water vapor map reconstruction using data fusion and machine learning-based spatial downscaling | |
| CN112132228B (en) | Irradiance data interpolation method and system based on decision tree classification |
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 |