CN113536657A - Ocean atmosphere refractive index prediction method based on machine learning - Google Patents
Ocean atmosphere refractive index prediction method based on machine learning Download PDFInfo
- Publication number
- CN113536657A CN113536657A CN202110490030.5A CN202110490030A CN113536657A CN 113536657 A CN113536657 A CN 113536657A CN 202110490030 A CN202110490030 A CN 202110490030A CN 113536657 A CN113536657 A CN 113536657A
- Authority
- CN
- China
- Prior art keywords
- data
- refractive index
- meteorological
- sounding
- meteorological parameters
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000010801 machine learning Methods 0.000 title claims abstract description 22
- 238000012937 correction Methods 0.000 claims abstract description 44
- 230000007246 mechanism Effects 0.000 claims abstract description 35
- 238000013528 artificial neural network Methods 0.000 claims abstract description 13
- 238000005259 measurement Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 238000013507 mapping Methods 0.000 claims abstract description 9
- 230000008859 change Effects 0.000 claims abstract description 7
- 238000003066 decision tree Methods 0.000 claims abstract description 7
- 238000004088 simulation Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 10
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Chemical compound O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 239000002352 surface water Substances 0.000 claims description 2
- 238000012795 verification Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 claims 1
- 238000012512 characterization method Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 description 5
- 230000000052 comparative effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4007—Interpolation-based scaling, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
Abstract
The invention discloses a marine atmosphere refractive index prediction method based on machine learning, which comprises the following steps: firstly, extracting data characteristics of a mesoscale numerical model forecast meteorological parameter and a sounding actual measurement meteorological parameter and acquiring distribution conditions of the two data; then, carrying out data preprocessing of a time-space layer by using alignment methods such as bilinear interpolation, neural network fitting and the like to obtain a difference rule between two meteorological parameters and a change rule of the meteorological parameters; then, based on the two provided meteorological parameters, machine learning models such as a gradient lifting decision tree are introduced, so that a mesoscale numerical model forecast meteorological parameter correction mechanism which is integrated into the characteristics of the actual measurement meteorological parameters of the sounding space is provided; and finally, based on a forecast meteorological parameter correction mechanism, further providing a novel atmospheric refractive index forecasting mechanism by utilizing a mapping relation from meteorological parameters to atmospheric refractive index. Compared with the traditional mesoscale numerical mode, the prediction method of the atmospheric refractive index is more accurate in prediction result, overcomes the limitation of actually measured data distribution, and is more practical and expandable.
Description
Technical Field
The invention relates to radio meteorology and electric wave science technologies, in particular to a marine atmosphere refractive index prediction method based on machine learning.
Background
Since the climatic factors such as the humidity, the air pressure, the wind speed and the waves of the marine environment change along with the change of the sea surface height, the refractive index of the atmosphere above the sea surface also changes correspondingly. The radius of curvature of electromagnetic wave rays is different due to different atmospheric refractive index distribution, and atmospheric refraction can be divided into negative refraction, no refraction, standard refraction, super-refraction, critical refraction and the like according to the ratio of the radius of curvature of the rays to the radius of the earth. When the change of the refraction rate meets a certain condition, atmospheric waveguide is formed, so that the over-the-horizon phenomenon and the radar blind area occur, and the influence on the performance of systems such as communication, detection, navigation and the like in the low-altitude range on the sea is obvious. Therefore, detection and prediction of the ocean atmosphere refractive index are crucial to building seamless, reliable and safe ocean information perception and communication.
The traditional prediction method for the atmospheric refractive index mainly comprises a mode based on sounding actual measurement and mesoscale numerical prediction. The former can provide accurate atmospheric refractive index information but is highly dependent on the acquisition of a data set, thereby severely limiting the adaptability thereof, and the latter is more practical but still insufficient in quantifying the accuracy of the atmospheric refractive index. Therefore, a new mechanism for predicting the refractive index of the atmosphere, which can effectively improve the accuracy of the mesoscale numerical prediction mode, is urgently needed to be searched. In recent years, big data and machine learning techniques are being developed, and have been widely applied to various fields such as image recognition and natural language processing, and excellent prediction effects are obtained, and a related work aiming at marine atmospheric refractive index prediction is urgently needed. The accuracy of acquiring an atmosphere refractive index curve by sounding data and the practicability of forecasting the atmosphere refractive index by a numerical mode are considered, based on the two data, the machine learning theoretical method is utilized, the internal rules and the relation among the data are expected to be deeply mined, a medium-scale numerical mode atmosphere refractive index forecasting model driven by the sounding data is constructed, and the accuracy, the practicability and the expansibility of forecasting by a traditional method are improved.
Disclosure of Invention
In view of this, the invention provides a method for predicting the refractive index of the ocean atmosphere based on machine learning.
The method may specifically include: acquiring data characteristics and distribution conditions of mesoscale numerical mode forecast meteorological parameters and sounding real meteorological parameters; performing data preprocessing on a time-space level by using alignment methods such as bilinear interpolation, neural network fitting and the like to obtain a difference rule between two meteorological parameters and a change rule of the meteorological parameters; based on the two provided meteorological parameters, machine learning models such as a gradient lifting decision tree are introduced to obtain a mesoscale numerical model forecast meteorological parameter correction mechanism which is integrated into the actual measurement meteorological parameter characteristics of the sounding space; based on the forecast weather parameter correction mechanism, the novel atmospheric refractive index forecasting mechanism is obtained by utilizing the mapping relation from the weather parameters to the atmospheric refractive index.
Wherein, the data characteristics of the mesoscale numerical model forecast meteorological data and the sounding actual measurement meteorological data comprise: taking netCDF4 and numpy modules of Python as examples to extract data characteristics, forecasting meteorological parameters in a mesoscale numerical mode provides data of 24 time periods, including parameters such as sea surface water vapor pressure, air temperature, air pressure and water vapor mixing ratio, and the like, and the method is higher in practicability but is insufficient in precision; the sounding actual measurement meteorological parameters provide data of 2 time periods, including parameters such as air pressure, water vapor mixing ratio, height, temperature, dew temperature, wind direction and wind speed, and the precision is high but the expansibility is low.
The distribution of the mesoscale numerical model forecast meteorological data and the sounding actual measurement meteorological data comprises the following steps: taking the data used by the invention as an example, the mesoscale numerical model forecast data are intensively distributed in the south sea area, the spacing resolution is about 25km, the data volume can reach 711450 at a certain moment, and the characteristics of dense data and large volume are realized; the measured sounding data are distributed all over the world, the distribution is extremely sparse, the distance is long, and the total number is 536.
The data preprocessing means for performing the data preprocessing on the spatio-temporal level by using alignment methods such as bilinear interpolation, neural network fitting and the like comprises the following steps: combining the time characteristics of the two data and looking up the corresponding relation between Universal Time (UTC) and Beijing time to align the time levels of the two data; taking a Basemap module of Python as an example, drawing a distribution map of the medium-scale numerical mode grid points and the sounding observation stations, and screening to obtain 29 sounding observation stations in the south sea area; processing numerical mode data by using a bilinear interpolation method to carry out longitude and latitude alignment so as to obtain two kinds of data of the longitude and latitude alignment; and processing sounding observation data by using a neural network fitting method to perform vertical height alignment to obtain two kinds of data with aligned vertical height layers.
The correction mechanism for obtaining the forecast weather parameters by using the machine learning model comprises the following steps: designing three characteristic engineering schemes consisting of two meteorological parameters; taking two machine learning models, namely a gradient lifting decision tree (GBDT) and an extreme gradient lifting tree (XGboost) as an example, based on feature mapping provided by three feature projects, training the GBDT and the XGboost models by using different training set data, using the fitting condition of a verification set data inspection model, and finally using the correction capability of a test set data inspection model to obtain six meteorological parameter correction schemes; and obtaining an optimal meteorological parameter correction mechanism through model evaluation index comparison.
Wherein, three kinds of characteristic engineering schemes that above-mentioned two kinds of meteorological parameters are constituteed include: the first characteristic engineering: only the aligned 1 simulation data is used; and (2) characteristic engineering II: using 4 original simulation data surrounding the sounding data; and a third characteristic engineering: aligned 1 simulation data and 3 geographical variables were used.
Wherein the model evaluation index includes: model goodness of fit; correcting the average error of the sounding data compared with the analog data; and Root Mean Square Error (RMSE).
The mechanism for obtaining the novel atmospheric refractive index prediction comprises the following steps: obtaining more accurate forecast weather parameters according to a forecast weather parameter correction mechanism; obtaining a novel atmosphere refractive index forecasting mechanism with higher precision according to the mapping relation from the corrected meteorological parameters to the atmosphere refractive index; and verifying the accuracy of the novel atmospheric refractive index forecasting mechanism by comparing the atmospheric refractive index curve condition with the correction and average error.
Drawings
FIG. 1 is a flow chart of a method for machine learning based prediction of marine atmospheric refractive index according to some embodiments of the present invention;
FIG. 2 is a diagram of data preprocessing results of bilinear interpolation, neural network fitting, etc. according to some embodiments of the present invention;
FIG. 3 is a diagram of the results of meteorological parameter modification based on three feature engineering schemes and two machine learning models, in accordance with some embodiments of the present invention;
FIG. 4 is a graph comparing the goodness of fit of the model, the mean error of the modified data compared to the simulated data, and the Root Mean Square Error (RMSE) for some embodiments of the present invention;
FIG. 5 is a graph comparing refractive index profiles of three atmospheres according to some embodiments of the invention;
FIG. 6 is a graph comparing the average error of three atmospheric refractive indices according to some embodiments of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the accompanying drawings.
FIG. 1 is a flow chart of a method for predicting the refractive index of marine atmosphere based on machine learning according to some embodiments of the present invention. As shown in fig. 1, for the problem of inaccurate prediction of atmospheric refractive index by traditional mesoscale numerical simulation, first, data characteristics of mesoscale numerical model prediction meteorological parameters and sounding actual measurement meteorological parameters are extracted and distribution conditions of the two data are obtained; then, carrying out data preprocessing on a time-space level by using alignment methods such as bilinear interpolation, neural network fitting and the like to obtain a difference rule between two meteorological parameters and a change rule of the meteorological parameters; then, based on the two provided meteorological parameters, a machine learning model such as a gradient lifting decision tree is introduced, so that a mesoscale numerical model forecast meteorological parameter correction mechanism which is integrated with the characteristics of the actual measurement meteorological parameters of the sounding space is provided; and finally, based on a forecast meteorological parameter correction mechanism, a novel atmospheric refractive index forecasting mechanism is provided by utilizing the mapping relation from meteorological parameters to atmospheric refractive index.
Fig. 2 is a diagram illustrating data preprocessing results such as bilinear interpolation and neural network fitting according to some embodiments of the present invention. As can be seen from fig. 2, the relative position distribution of the mesoscale numerical model prediction data and the sounding measured data is obtained, and the bilinear interpolation method is used to process the numerical simulation data to perform longitude and latitude alignment, so as to obtain two kinds of data with aligned longitude and latitude. The bilinear interpolation method is as follows:
first, linear interpolation is performed in the x direction to obtain the following equation:
then, linear interpolation is performed in the y direction to obtain the following equation:
wherein, 4 simulation point data around the sounding data site are Q as a function f11=(x1,y1), Q12=(x1,y2),Q21=(x2,y1) And Q22=(x2,y2) The values of 4 points, the sounding data site data are the values of the function f at point P ═ x, y.
And then, processing sounding observation data by using a neural network fitting method to perform vertical height alignment, so as to obtain two kinds of data with vertical height level alignment. The neural network fit is shown as follows:
yj=f(βj-θj) (4)
wherein, betajFor the j-th output neuronInput, θjIs the threshold value of the jth node of the output layer, yjIs the output value of the jth node of the output layer. The neural network training is to use the model output value y and the actual value ykThe objective is to make the mean square error as small as possible, and the learning process is realized by updating the weight and the threshold value, so as to finally obtain the optimal solution.
As shown in FIG. 2, a fitting curve of meteorological parameters (water vapor pressure, temperature, air pressure, water vapor mixing ratio) of 0-3000m can be fitted by a neural network fitting method. And then, finding meteorological parameters with the same height as the numerical simulation data from the fitting curve to realize vertical height alignment of the two data.
Further, the data preprocessing means includes: combining the time characteristics of the two data and looking up the corresponding relation between Universal Time (UTC) and Beijing time to align the time levels of the two data; and screening to obtain 29 sounding observation sites in the south sea area by combining the distribution conditions of the mesoscale numerical mode lattice points and the sounding observation sites.
The data preprocessing means can realize the one-to-one correspondence relationship between the numerical simulation data and the sounding observation data.
FIG. 3 is a diagram of the results of meteorological parameter modification based on three feature engineering schemes and two machine learning models, according to some embodiments of the present invention. Aiming at the three characteristic engineering schemes and two machine learning models, six forecast meteorological parameter correction mechanisms are combined, the result is presented by taking a correction curve of meteorological parameter water vapor pressure as an example, a green line in a correction curve graph is a meteorological parameter curve corrected by GBDT and XGboost models, and a red line is a sounding meteorological parameter curve. As shown in fig. 3, the six correction results are GBDT and XGBoost correction results using the first feature engineering in sequence; correcting results by using GBDT and XGboost of the second characteristic engineering; and correcting results by using GBDT and XGboost of the third characteristic engineering.
The coincidence degree of the correction curve of the characteristic engineering III (containing the geographic variable) and the exploration data curve is higher, which represents that the geographic variable has a crucial influence on the correction result. Compared with two machine learning models, the XGboost is faster than the GBDT in training speed, and the result is more accurate. Therefore, the forecast weather parameter correction mechanism provided by the invention is an XGboost correction model using the aligned simulation data and geographic variables.
In order to demonstrate the service performance of each embodiment of the invention, the inventor carries out a comparative experiment of various indexes and methods. Fig. 4, 5 and 6 show experimental results in various aspects. FIG. 4 is a graph comparing goodness of fit of the model, mean error of the corrected data compared to the simulated data, and Root Mean Square Error (RMSE) according to one embodiment of the invention; FIG. 5 is a graph comparing the refractive indices of three atmospheres according to some embodiments of the invention; FIG. 6 is a graph comparing the average error of three atmospheric refractive indices according to some embodiments of the present invention.
The comparative experiment shown in fig. 4 uses model goodness of fit, mean error of the corrected data compared with the simulated data, and Root Mean Square Error (RMSE) as evaluation indexes to verify the correction performance of the prediction meteorological parameter correction mechanism provided by the invention, and an experimental result can be obtained by calculating through the following three formulas:
wherein, the training set and the testing set R of six forecast weather parameter correction mechanisms can be calculated by formula (5)2,R2The closer to 1, the better the correction effect; the average error of the measured sounding data of the six types of corrected data compared with the simulation data can be calculated through a formula (6), when the correction-sounding average error is smaller than the simulation-sounding average error, the correction effect of the forecast meteorological parameter correction mechanism can be shown, and the smaller the average error is, the better the correction effect is; tong (Chinese character of 'tong')The root mean square errors of the training set and the test set of the six forecast weather parameter correction mechanisms under the conditions of different decision trees can be calculated through the formula (7), and the smaller the error is, the better the correction effect is.
As can be seen from FIG. 4, the training set and the test set R of the forecast weather parameter correction mechanism (XGboost correction model using aligned simulation data and geographic variables) provided by the invention2And the average error between the obtained correction data and the actual measurement data of the sounding is the minimum. And with the increase of the number of the decision trees, the root mean square error of the training set and the testing set is reduced faster and is closer to 0.
The comparative experiment shown in fig. 5 gives a comparison of the results of the corrected atmospheric refractive index profile, the simulated atmospheric refractive index profile and the probed atmospheric refractive index profile at three stations. The three atmospheric refractive indexes are respectively obtained by correcting meteorological parameters, simulating meteorological parameters and exploring meteorological parameters, and the calculation formula of the atmospheric refractive index M is as follows:
wherein T is the atmospheric temperature, K; p is atmospheric pressure, hPa; e is the water vapor pressure, hPa, Z is the altitude, m, these parameters can be directly or indirectly obtained from numerical simulation data, R is the average earth radius, and the value is 6.371 multiplied by 106m。
As can be seen from fig. 5, the forecast weather parameter correction mechanism provided in the embodiment of the present invention, in combination with the mapping relationship between the weather parameter and the atmospheric refractive index, the corrected atmospheric refractive index curve obtained is closer to the sounding atmospheric refractive index curve than the simulated atmospheric refractive index curve, and can well reflect the waveguide characteristics (which are shown in an enlarged manner) displayed in the sounding atmospheric refractive index curve. Therefore, the method can obtain high-precision correction of the atmospheric refractive index, and is vital to the improvement of the accuracy of waveguide judgment and prediction and the construction of seamless, reliable and safe ocean information perception and communication.
The comparative experiment shown in fig. 6 gives a plot of the average error of the corrected atmospheric refractive index versus the simulated atmospheric refractive index versus the probe atmospheric refractive index for the three stations. Six average errors of the three stations can be calculated in turn by equation (6).
As is apparent from fig. 6, the average error of the correction-sounding data is much smaller than the average error of the simulation-sounding data, and the accuracy of the atmospheric refractive index correction at the three stations is improved by more than 80%, so that a more accurate atmospheric refractive index can be obtained by predicting a meteorological parameter correction mechanism. In summary, compared with the conventional atmospheric refractive index prediction in the mesoscale numerical mode, the novel atmospheric refractive index prediction mechanism provided in the embodiment of the present invention can obtain a more accurate atmospheric refractive index prediction result.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A marine atmosphere refractive index prediction method based on machine learning. Characterized in that the method comprises:
extracting data characteristics of the mesoscale numerical model forecast meteorological parameters and the sounding actual measurement meteorological parameters by using a data reading processing module, and acquiring the distribution conditions of the two data by using a data reading drawing module;
according to different characteristics of the two meteorological parameters, performing data preprocessing of a time-space layer by using alignment methods such as bilinear interpolation, neural network fitting and the like to obtain a difference rule between the two meteorological parameters and a change rule of the meteorological parameters;
based on the two provided meteorological parameters, machine learning models such as a gradient lifting decision tree are introduced, so that a mesoscale numerical model forecast meteorological parameter correction mechanism which is integrated into the characteristics of the actual measurement meteorological parameters of the sounding space is provided;
based on the forecast weather parameter correction mechanism, the mapping relation from the weather parameters to the atmospheric refractive index is utilized to further provide a novel atmospheric refractive index forecasting mechanism.
2. The method of claim 1, wherein said extracting data features of the mesoscale numerical model forecasted meteorological parameters and sounding measured meteorological parameters using a data reading processing module comprises:
the mesoscale numerical model forecast meteorological parameters provide data of 24 time periods, including parameters such as sea surface water vapor pressure, air temperature, air pressure, water vapor mixing ratio and the like, and the practicability is higher but the accuracy is insufficient; and
the sounding actual measurement meteorological parameters provide data of 2 time periods including air pressure, water vapor mixing ratio,
Height, temperature, dew temperature, wind direction and wind speed, high precision but low expansibility.
3. The method of claim 1, wherein the obtaining the distribution of the two data by the data reading and drawing module comprises:
the mesoscale numerical model forecast data are distributed in the south China sea area in a centralized manner, the data volume can reach hundreds of thousands, and the mesoscale numerical model forecast data have the characteristics of dense data and large volume; and
the sounding measured data are distributed all over the world, the distribution is extremely sparse, the distance is long, and the number of the sounding measured data is hundreds.
4. The method according to claim 1, wherein the data preprocessing means comprises:
combining the time characteristics of the two data and looking up the corresponding relation between Universal Time (UTC) and Beijing time to align the time levels of the two data; and
drawing a distribution map for the scale numerical value mode grid points and the sounding observation stations by using a data reading drawing module, and screening to obtain sounding observation stations distributed in the south sea area; and
processing numerical mode data by using a bilinear interpolation method to carry out longitude and latitude alignment so as to obtain two kinds of data of the longitude and latitude alignment; and
and processing sounding observation data by using a neural network fitting method to perform vertical height alignment to obtain two kinds of data with aligned vertical height layers.
5. The method of claim 1, wherein the mechanism for using the machine learning model to obtain forecasted weather parameter corrections comprises:
designing three characteristic engineering schemes consisting of two meteorological parameters; and
based on the feature mapping provided by the three feature projects, training two machine learning models by using different training set data, checking the fitting condition of the models by using verification set data, and finally checking the correction capability of the models by using test set data to obtain six meteorological parameter correction schemes; and
and obtaining an optimal meteorological parameter correction mechanism by comparing evaluation indexes such as goodness of fit of the analysis model, average error of the corrected data compared with the simulated data, Root Mean Square Error (RMSE) and the like.
6. The method of claim 5, wherein the three characterization schemes comprise:
the first characteristic engineering: only the aligned 1 simulation data is used; and
and (2) characteristic engineering II: using 4 original simulation data surrounding the sounding data; and
and (3) characteristic engineering: aligned 1 simulation data and 3 geographical variables were used.
7. The method of claim 1, wherein the novel atmospheric refractive index prediction mechanism comprises:
based on a forecast meteorological parameter correction mechanism, combining a mapping relation from meteorological parameters to atmospheric refractive index to obtain a novel atmospheric refractive index forecast mechanism; and
the accuracy of the novel atmospheric refractive index forecasting mechanism is verified by comparing, analyzing, correcting, simulating and exploring the atmospheric refractive index curve condition and correcting and simulating the average error of the atmospheric refractive index compared with the atmospheric refractive index.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110490030.5A CN113536657A (en) | 2021-05-06 | 2021-05-06 | Ocean atmosphere refractive index prediction method based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110490030.5A CN113536657A (en) | 2021-05-06 | 2021-05-06 | Ocean atmosphere refractive index prediction method based on machine learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113536657A true CN113536657A (en) | 2021-10-22 |
Family
ID=78095351
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110490030.5A Pending CN113536657A (en) | 2021-05-06 | 2021-05-06 | Ocean atmosphere refractive index prediction method based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113536657A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114626458A (en) * | 2022-03-15 | 2022-06-14 | 中科三清科技有限公司 | High-voltage rear part identification method and device, storage medium and terminal |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120185414A1 (en) * | 2010-12-15 | 2012-07-19 | Vaisala, Inc. | Systems and methods for wind forecasting and grid management |
CN110031877A (en) * | 2019-04-17 | 2019-07-19 | 山东科技大学 | A kind of region NWP tropospheric delay correction method based on GRNN model |
CN110472782A (en) * | 2019-08-01 | 2019-11-19 | 软通动力信息技术有限公司 | A kind of data determination method, device, equipment and storage medium |
CN112100921A (en) * | 2020-09-16 | 2020-12-18 | 平衡机器科技(深圳)有限公司 | Method for acquiring wind resource and wind speed based on WRF and random forest |
-
2021
- 2021-05-06 CN CN202110490030.5A patent/CN113536657A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120185414A1 (en) * | 2010-12-15 | 2012-07-19 | Vaisala, Inc. | Systems and methods for wind forecasting and grid management |
CN110031877A (en) * | 2019-04-17 | 2019-07-19 | 山东科技大学 | A kind of region NWP tropospheric delay correction method based on GRNN model |
CN110472782A (en) * | 2019-08-01 | 2019-11-19 | 软通动力信息技术有限公司 | A kind of data determination method, device, equipment and storage medium |
CN112100921A (en) * | 2020-09-16 | 2020-12-18 | 平衡机器科技(深圳)有限公司 | Method for acquiring wind resource and wind speed based on WRF and random forest |
Non-Patent Citations (2)
Title |
---|
焦林 等: "基于中尺度模式MM5下的海洋蒸发波导预报研究", 气象学报, vol. 67, no. 3, pages 382 - 387 * |
许立兵 等: "基于机器学习的海洋环境预报订正方法研究", 海洋通报, vol. 39, no. 6, pages 695 - 704 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114626458A (en) * | 2022-03-15 | 2022-06-14 | 中科三清科技有限公司 | High-voltage rear part identification method and device, storage medium and terminal |
CN114626458B (en) * | 2022-03-15 | 2022-10-21 | 中科三清科技有限公司 | High-voltage rear part identification method and device, storage medium and terminal |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107390298B (en) | A kind of analogy method and device of Complex Mountain underlying surface strong wind | |
CN112711899B (en) | Fusion prediction method for height of evaporation waveguide | |
CN114280696A (en) | Intelligent sea fog level forecasting method and system | |
CN105095589B (en) | A kind of mountain area power grid wind area is distributed drawing drawing method | |
Mortensen | Wind resource assessment using WAsP software | |
CN108981616B (en) | Method for inverting effective leaf area index of artificial forest by unmanned aerial vehicle laser radar | |
CN109543356A (en) | Consider the ocean interior temperature-salinity structure remote sensing inversion method of Space atmosphere | |
CN111598942A (en) | Method and system for automatically positioning electric power facility instrument | |
CN108319772A (en) | A kind of analysis method again of wave long term data | |
CN116449331B (en) | Dust particle number concentration estimation method based on W-band radar and meteorological satellite | |
CN114819737B (en) | Method, system and storage medium for estimating carbon reserves of highway road vegetation | |
Mortensen | Wind resource assessment using the WAsP software (DTU Wind Energy E-0135) | |
CN115294147A (en) | Method for estimating aboveground biomass of single trees and forests based on unmanned aerial vehicle laser radar | |
CN113536657A (en) | Ocean atmosphere refractive index prediction method based on machine learning | |
CN117172149A (en) | Evaporation waveguide prediction method based on data feature classification and neural network model | |
CN117035174A (en) | Method and system for estimating biomass on single-woodland of casuarina equisetifolia | |
KR102002593B1 (en) | Method and apparatus for analyzing harmful gas diffusion in a specific space | |
CN109948175B (en) | Satellite remote sensing albedo missing value inversion method based on meteorological data | |
CN116609859A (en) | Weather disaster high-resolution regional mode forecasting system and method | |
CN112632799B (en) | Method and device for evaluating design wind speed of power transmission line | |
CN115511192A (en) | Rainfall forecasting method and system based on lightning data assimilation | |
CN115480032A (en) | Point source discharge intensity prediction method based on ground remote sensing measurement | |
CN114880933A (en) | Atmospheric temperature and humidity profile inversion method and system for non-exploration-site foundation microwave radiometer based on reanalysis data | |
CN111538943B (en) | Novel high-space-time resolution global ZTD vertical section grid model construction method | |
CN115840908A (en) | Method for constructing PM2.5 three-dimensional dynamic monitoring field by microwave link based on LSTM model |
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 |