Ground wave radar data multi-dimensional quality control and evaluation system based on electric wave and ocean characteristics

Abstract

The invention discloses a ground wave radar multi-dimensional data quality control and evaluation system based on electric wave and ocean characteristics, and the system comprises the steps: obtaining different types of ground wave radar ocean environment data obtained through preprocessing, and carrying out the inspection of the different types of data through different quality control methods and flows, the quality control method comprises format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold value inspection, geometric accuracy factor inspection, median filtering inspection, Rheinda inspection, gradient inspection, relevance inspection and visualization inspection; and selecting different evaluation methods for different types of data to carry out quality evaluation according to variable precision characteristics of ground wave radar marine environment data plane distribution, and obtaining corresponding evaluation conclusions. According to the invention, a refined quality control and evaluation scheme for the ground wave radar data is established by comprehensively considering that the ground wave radar marine environment data has the electric wave and marine dual characteristics at the same time.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics. Background Art

[0002] Ground wave radar ocean environment monitoring technology is a new discipline formed by the intersection of radio science and physical oceanography. Ocean radar technology upgrades ground-based ocean dynamic environment monitoring from traditional "point" and "line" observation methods to beyond-horizon, all-weather, large-area continuous remote sensing methods. Compared with satellite observations, it has advantages in time and space resolution, and can continuously obtain relatively complete spatiotemporal variation information of dynamic parameters in the observed sea area. Ground wave radar has become a conventional ocean observation equipment and is widely used around the world. Faced with multi-source radar observation data with many sources, different formats, increasing data volume, and increasing timeliness, accurate and effective data quality control and evaluation is one of the major challenges faced by marine workers.

[0003] In recent years, there have been many research results in the quality control of marine field measurements and remote sensing data, but there has been no refined quality control and evaluation scheme for ground wave radar data. Compared with other observation methods, ground wave radar data has both radio wave and ocean characteristics, and it is difficult to perform quality control through universal outlier detection methods. Therefore, it is necessary to establish practical and effective quality control methods and processes based on the characteristics of ground wave radar data. At the same time, it is also crucial to establish a scientific and reasonable quality evaluation method for the variable precision characteristics of the plane distribution of ground wave radar data. Summary of the invention

[0004] The purpose of this application is to propose a multi-dimensional quality control and evaluation system for ground wave radar data based on radio wave and ocean characteristics in response to the above-mentioned technical problems.

[0005] In a first aspect, the present invention provides a multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics, comprising:

[0006] A data acquisition module is configured to acquire different types of ground wave radar ocean environment data obtained through preprocessing, wherein the data types of the ground wave radar data include ground wave radar radial flow data, ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar sea wave data;

[0007] The data quality control module is configured to use different quality control methods and processes to test different types of data in the ground wave radar ocean environment data, and separate the ground wave radar ocean environment data with quality identification after quality control according to the discrimination criteria set by the quality control method. The quality control methods include format test, time range test, position test, radar effective observation range test, threshold test, geometric precision factor test, median filter test, Rheinland test, gradient test, correlation test and visualization test;

[0008] The data evaluation module is configured to select different evaluation methods to perform quality evaluation on different types of data in the ground wave radar ocean environment data after quality control, and obtain corresponding evaluation conclusions.

[0009] As a preferred method, different quality control methods and processes are used to inspect different types of data in the ground wave radar ocean environment data, and the ground wave radar ocean environment data after quality control with quality identification is separated according to the discrimination criteria set by the quality control method, specifically including:

[0010] The quality control methods used for the ground wave radar radial flow data include format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, median filter inspection, Rheinda inspection, gradient inspection and visualization inspection, and the ground wave radar radial flow data after quality control is obtained;

[0011] The quality control method used for ground wave radar vector flow data includes format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, geometric precision factor inspection, median filter inspection, Rheinland inspection, gradient inspection and visualization inspection, wherein the flow velocity and flow direction of the ground wave radar vector flow data are firstly subjected to format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection and geometric precision factor inspection respectively to obtain the ground wave radar vector flow data after the first inspection, and then the ground wave radar vector flow data after the first inspection is decomposed into the east component and the north component; the flow velocity, east component and north component in the ground wave radar vector flow data after the first inspection are subjected to the median filter inspection and Rheinland inspection of the spatial neighborhood and the Rheinland inspection and gradient inspection of the temporal neighborhood respectively to obtain the ground wave radar vector flow data after the second inspection, and finally the ground wave radar vector flow data after the second inspection is subjected to visualization inspection to obtain the ground wave radar vector flow data after quality control;

[0012] The quality control methods used for the ground wave radar sea surface wind data include format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, median filter inspection, Rheinland inspection, gradient inspection and visualization inspection; wherein the wind speed and wind direction in the ground wave radar sea surface wind data are firstly subjected to format inspection, time range inspection, position inspection, radar effective observation range inspection and threshold inspection respectively, to obtain the ground wave radar sea surface wind data after the first inspection, and then the ground wave radar sea surface wind data after the first inspection is decomposed into east component and north component, and then the wind speed, east component and north component in the ground wave radar sea surface wind data after the first inspection are subjected to median filter inspection and Rheinland inspection in spatial neighborhood and Rheinland inspection and gradient inspection in temporal neighborhood respectively, to obtain the ground wave radar sea surface wind data after the second inspection, and finally the ground wave radar sea surface wind data after the second inspection is subjected to visualization inspection to obtain the ground wave radar sea surface wind data after quality control;

[0013] The quality control methods used for ground wave radar wave data include format test, time range test, position test, radar effective observation range test, threshold test, median filter test, Rheinland test, gradient test, correlation test and visualization test; the effective wave height and wave direction in the ground wave radar wave data are firstly subjected to format test, time range test, position test, radar effective observation range test and threshold test to obtain the ground wave radar wave data after the first test, and then the ground wave radar wave data after the first test is decomposed into east component and north component, and then the first The effective wave height, east component and north component in the tested ground wave radar wave data are subjected to median filtering test and Rheinland test in spatial neighborhood and Rheinland test and gradient test in temporal neighborhood respectively to obtain the ground wave radar wave data after the second test. Based on the wind speed in the ground wave radar sea surface wind data after quality control, the effective wave height in the ground wave radar wave data after the second test is subjected to correlation test to obtain the ground wave radar wave data after the third test. Finally, the ground wave radar wave data after the third test is re-visualized to obtain the ground wave radar wave data after quality control.

[0014] Preferably, the quality identifier includes an identifier for marking credible data, suspicious data, erroneous data and missing data.

[0015] As a preferred method, the steps for checking the geometric precision factor are as follows:

[0016] The geometric dilution of precision is calculated using the following formula:

[0017]

[0018] Among them, α represents the angle between the line connecting the observation target and the middle position of the two radar stations and the east direction, 2θ represents the angle between the observation target and the line connecting the two radar stations, GDOP, GDOP nand GDOP e respectively represent the values of the geometric dilution of precision and its north and east components;

[0019] When GDOP < 4, the test result is reliable data;

[0020] When 4 < GDOP ≤ 6, the test result is suspicious data;

[0021] When GDOP > 6, the test result is incorrect data;

[0022] The steps of the median filtering test in the spatial neighborhood are as follows:

[0023] For the ground wave radar radial flow data: Search for the first observation results of the target radar element within the detection angle ±1.5 × angular resolution and the detection distance ±1.5 × range resolution, calculate the median of the first observation results. If the absolute value of the difference between the observed value of the target radar element and the median of the first observation results is greater than the threshold, the test result is judged as incorrect data. If the absolute value of the difference between the observed value of the target radar element and the median of the first observation results is less than or equal to the threshold, the test result is judged as reliable data;

[0024] For the ground wave radar vector flow, ground wave radar sea surface wind data, and ground wave radar sea wave data: Search for the second observation results of the target grid point within the detection angle ±1.5 × spatial resolution, calculate the median of the second observation results. If the absolute value of the difference between the observed value of the target grid point and the median of the second observation results is greater than the threshold, the test result is judged as incorrect data. If the absolute value of the difference between the observed value of the target grid point and the median of the second observation results is less than or equal to the threshold, the test result is judged as reliable data;

[0025] The steps of the Grubbs' test in the spatial neighborhood are as follows:

[0026] For the ground wave radar radial flow data: Calculate the mean and standard deviation of the first observation results. If the difference between the observed value of the target radar element and the mean of the first observation results is greater than or equal to three times the standard deviation of the first observation results, the test result is incorrect data. If the difference between the observed value of the target radar element and the mean of the first observation results is greater than or equal to two times the standard deviation of the first observation results and less than three times the standard deviation of the first observation results, the test result is suspicious data. If the difference between the observed value of the target radar element and the mean of the first observation results is less than two times the standard deviation of the first observation results, the test result is reliable data;

[0027] For ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar wave data: calculate the mean value and standard deviation of the second observation result; if the difference between the observation value of the target grid point and the mean value of the second observation result is greater than or equal to three times the standard deviation of the second observation result, the test result is erroneous data; if the difference between the observation value of the target grid point and the mean value of the second observation result is greater than or equal to two times the standard deviation of the second observation result and less than three times the standard deviation of the second observation result, the test result is suspicious data; if the difference between the observation value of the target grid point and the mean value of the second observation result is less than two times the standard deviation of the second observation result, the test result is credible data;

[0028] The steps of the Rheinland test for the time neighborhood are as follows:

[0029] For a certain specified target radar element or target grid point at a certain moment, search for the third observation result within a certain time window before and after the data time point of the target radar element or target grid point, calculate the average value and standard deviation of the third observation result, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is greater than or equal to three times the standard deviation of the third observation result, then the test result is erroneous data, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is greater than or equal to two times the standard deviation of the third observation result and less than three times the standard deviation of the third observation result, then the test result is suspicious data, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is less than two times the standard deviation of the third observation result, then the test result is credible data;

[0030] The steps of the association test are as follows:

[0031] Establish the fitting relationship between wind speed and significant wave height in the target sea area;

[0032] Calculate the 95% prediction interval in the fitting relationship, and combine the wind speed in the surface wave radar sea surface wind data after quality control to infer the reasonable range of the effective wave height data in the surface wave radar sea wave data after the second test;

[0033] Visual inspection involves spatial misjudgment rate and temporal misjudgment rate. The spatial misjudgment rate is for the marine environment data field at each moment, and the temporal misjudgment rate is for the marine environment data field that has misjudgment within the first preset period. The specific definitions are as follows:

[0034] Spatial misjudgment rate = number of misjudged data points in each marine environment data field / total number of valid data points in each marine environment data field × 100%;

[0035] If the spatial misjudgment rate is lower than the first misjudgment threshold, the data of the marine environment data field is regarded as a normal field; if the spatial misjudgment rate is higher than the first misjudgment threshold, the data of the marine environment data field is regarded as a misjudgment field;

[0036] Time misjudgment rate = number of misjudged fields in a certain period of time / total number of marine environment data fields in the same period of time × 100%;

[0037] If the time misjudgment rate is lower than the second misjudgment threshold, the corresponding quality control method is effective. If the time misjudgment rate is higher than the second misjudgment threshold, the corresponding quality control method needs to be optimized until the requirements are met.

[0038] As a preference, different evaluation methods are selected for different types of data in the quality-controlled ground wave radar ocean environment data to perform quality evaluation, and corresponding evaluation conclusions are obtained, specifically including:

[0039] The evaluation method used for the ground wave radar radial flow data after quality control is as follows: statistics on the distribution of spatial coverage of all grid points in the ground wave radar radial flow data after quality control within the first preset period, the spatial coverage of the grid points is the number of radar radial flow grid point samples in the first preset period / the number of radar theoretical sampling events in the first preset period, and the number of radar radial flow grid point samples is the number of credible data;

[0040] Further calculate the angle average spatial coverage rate and the distance average spatial coverage rate. The angle average spatial coverage rate is obtained by dividing the detection viewing angle according to preset intervals, averaging the spatial coverage rates of the grid points within the same detection viewing angle interval, and obtaining the angle average spatial coverage rate. The distance average spatial coverage rate is obtained by dividing the detection distance according to preset intervals, averaging the spatial coverage rates of the grid points within the same detection distance interval, and obtaining the distance average spatial coverage rate. Analyze the changing characteristics of the angle average spatial coverage rate with the detection viewing angle and the changing characteristics of the distance average spatial coverage rate with the detection distance. If the angle average spatial coverage rate gradually decreases with the increase of the detection viewing angle and the distance average spatial coverage rate gradually decreases with the increase of the detection distance, then the detection conditions of the ground wave radar radial flow data after quality control and the corresponding ground wave radar vector flow data after quality control within the first preset time period are normal, otherwise they are abnormal.

[0041] The evaluation method used for the ground wave radar vector flow data after quality control is as follows: the stable observation area of ​​the ground wave radar vector flow data after quality control within the second preset time period is subjected to ocean current reconciliation analysis, the stable observation area is defined as the area where the spatial coverage of the ground wave radar vector flow data is above 70%, the main tidal components are separated according to the tidal characteristics of the sea area, and the tidal ellipse elements of each main tidal component are calculated, the tidal ellipse elements include the maximum tidal component, the minimum tidal component, the ellipticity and the maximum tidal direction, and the numerical value of the tidal ellipse element is compared with the historical numerical value; if the difference between the numerical value of the tidal ellipse element and the historical numerical value is less than or equal to the element threshold area, the detection of the ground wave radar vector flow data after quality control within the second preset time period is normal, and if the numerical value of the tidal ellipse element and the historical numerical value are greater than the element threshold area, the detection of the ground wave radar vector flow data after quality control within the second preset time period is abnormal;

[0042] The evaluation method used for the quality-controlled ground wave radar sea surface wind data is as follows: the quality-controlled ground wave radar sea surface wind data is vector averaged with reference to the temporal and spatial resolution of the ERA5 reanalysis wind field to obtain data consistent with the temporal and spatial resolution of the ERA5 wind field, and then the time series diagrams of wind speed and wind direction of the ERA5 and quality-controlled ground wave radar sea surface wind data in the corresponding grid are compared, and the credibility of the quality-controlled ground wave radar sea surface wind data in the grid point is judged by analyzing whether the change trends between the two are consistent;

[0043] The evaluation method used for the quality-controlled ground wave radar wave data is as follows: the quality-controlled ground wave radar wave data are vector averaged with reference to the spatiotemporal resolution of the ERA5 reanalysis wind field to obtain data consistent with the spatial resolution of the ERA5 wind field. The quality-controlled ground wave radar wave data in the corresponding grid point are then compared with the time series diagram of the ERA5 wind vector. The credibility of the quality-controlled ground wave radar wave data in the grid is judged by analyzing whether the change trends between the two are consistent.

[0044] Preferably, the preprocessing process includes checking the readability, completeness and format standardization of the ground wave radar ocean environment data, decoding, sorting and deduplication processing, and format conversion according to the data file format required for quality control to obtain a data file with a unified format.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] (1) The multi-dimensional quality control and evaluation system for ground wave radar data based on radio wave and ocean characteristics proposed in the present invention proposes a refined quality control and evaluation scheme for ground wave radar data. This scheme fully considers that the ground wave radar environment has both radio wave and ocean characteristics, and comprehensively controls and evaluates data quality from multiple dimensions such as time and space, providing a guarantee for obtaining complete, accurate, and reliable ocean observation data.

[0047] (2) The multi-dimensional quality control and evaluation system for ground wave radar data based on radio wave and ocean characteristics proposed in the present invention introduces a geometric precision factor verification method into the quality control method, and performs multi-dimensional quality control on the ground wave radar environment from the spatial neighborhood and the temporal neighborhood respectively.

[0048] (3) The multi-dimensional quality control and evaluation system for ground wave radar data based on radio wave and ocean characteristics proposed in the present invention targets the variable precision characteristics of the planar distribution of ground wave radar ocean environment data, fully considers the spatiotemporal variation characteristics of ground wave radar radio wave signals and the oceanographic characteristics of each element, and establishes a set of scientific and reasonable quality evaluation methods to comprehensively evaluate the overall quality of ground wave radar data and the reliability of data in different observation areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A schematic diagram of a multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics according to an embodiment of the present application;

[0051] Figure 2 A flowchart of a multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics according to an embodiment of the present application;

[0052] Figure 3 A flow chart of a quality control method of a multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics according to an embodiment of the present application;

[0053] Figure 4 A schematic diagram of the relative positions between the observation target and two ground wave radar stations in the ground wave radar data multi-dimensional quality control and evaluation system based on radio waves and ocean characteristics according to an embodiment of the present application;

[0054] Figure 5 A schematic diagram of a quality assessment process of a multi-dimensional quality control and assessment system for ground wave radar data based on radio waves and ocean characteristics according to an embodiment of the present application;

[0055] Figure 6A comparison diagram of the ground wave radar vector flow before and after quality control of the multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics of an embodiment of the present application at 14:20 on August 6, 2021, wherein (a) represents after quality control, and (b) represents before quality control;

[0056] Figure 7 Four tidal ellipse distributions of the ground wave radar vector flow from August 6, 2021 to August 31, 2021 of the multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics of an embodiment of the present application, wherein (a) represents the M2 tidal component, (b) represents the S2 tidal component, (c) represents the K1 tidal component, and (d) represents the O1 tidal component;

[0057] Figure 8 A schematic diagram of the comparison of the ground wave radar sea surface wind data and the ERA5 wind field from August 6, 2021 to August 25, 2021 in radar grid 1 (118.5°E, 23.75°N) of the multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics of an embodiment of the present application;

[0058] Fig. 9 A schematic diagram of the comparison of the ground wave radar sea surface wind data and the ERA5 wind field from September 15, 2021 to October 4, 2021 in radar grid 1 (118.5°E, 23.75°N) of the multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics of an embodiment of the present application;

[0059] Fig.10 A schematic diagram of the comparison of the ground wave radar sea surface wind data and the ERA5 wind field from August 6, 2021 to August 25, 2021 in the radar grid 2 (117.5°E, 23°N) of the ground wave radar data multi-dimensional quality control and evaluation system based on radio waves and ocean characteristics of an embodiment of the present application;

[0060] Fig.11 This is a schematic diagram of the comparison of the ground wave radar sea surface wind data and the ERA5 wind field in radar grid 2 (117.5°E, 23°N) from September 15, 2021 to October 4, 2021 of the multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics of an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] Figure 1 The embodiment of the present application shows a multi-dimensional quality control and evaluation system for ground wave radar data based on radio waves and ocean characteristics, including the following steps:

[0063] The data acquisition module 1 is configured to acquire different types of preprocessed ground wave radar ocean environment data, and the data types of the ground wave radar data include ground wave radar radial flow data, ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar wave data.

[0064] In a specific embodiment, the preprocessing process includes checking the readability, completeness and format standardization of the ground wave radar ocean environment data, decoding, sorting and deduplication processing, and format conversion according to the data file format required for quality control to obtain a data file with a unified format.

[0065] Specifically, the embodiments of the present application establish a set of practical data standardization processing procedures to provide a guarantee for obtaining complete, accurate and reliable ground wave radar ocean observation data. Based on the reference national standard GB / T14914.6-2021 "Ocean Observation Specification Part 6: Data Processing and Quality Control" and the marine industry standard HY / T0315-2021 "Ocean Observation Delay Data Quality Control Audit Technical Specifications", this set of procedures fully considers the oceanographic characteristics of different observation elements, the variable precision characteristics of the plane distribution of ground wave radar data, etc. for comprehensive optimization and upgrading. The embodiments of the present application provide standardized processing steps for ground wave radar radial flow data, ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar wave data, including data quality control and overall quality assessment. It can batch process ground wave radar data, automatically complete data quality control, and output quality-controlled data and comparison images before and after quality control. Then, based on the quality-controlled data of a certain period of time, the overall data quality assessment is performed to draw corresponding assessment conclusions.

[0066] refer to Figure 2First, the raw data of the ground wave radar is preprocessed, the observation elements and the corresponding raw data are checked, and the format is converted according to the data file format required for quality control. The observation elements are the data type of the ground wave radar data. Secondly, the corresponding quality control method is selected according to different observation elements for quality control, and the quality identification of different types of discrimination data is marked, and the quality control data with quality identification and unified format is output, and the comparison images before and after quality control are generated; the overall quality of the data after quality control is evaluated, and different evaluation methods are selected for the ground wave radar radial flow data, ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar wave data for overall quality evaluation, and the corresponding evaluation conclusions are obtained.

[0067] The data quality control module 2 is configured to use different quality control methods and processes to inspect different types of data in the ground wave radar ocean environment data, and separate the quality-controlled ground wave radar ocean environment data with quality identification according to the judgment criteria set by the quality control method. The quality control methods include format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, geometric precision factor inspection, median filtering inspection, Rheinland inspection, gradient inspection, correlation inspection and visualization inspection.

[0068] In a specific embodiment, the quality identifier includes an identifier for marking reliable data, suspicious data, erroneous data, and missing data.

[0069] In a specific embodiment, different types of data in the ground wave radar ocean environment data are inspected using different quality control methods and processes, and the ground wave radar ocean environment data after quality control with quality identification is separated according to the discrimination criteria set by the quality control method, which specifically includes:

[0070] The quality control methods used for the ground wave radar radial flow data include format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, median filter inspection, Rheinda inspection, gradient inspection and visualization inspection, and the ground wave radar radial flow data after quality control is obtained;

[0071] The quality control method used for ground wave radar vector flow data includes format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, geometric precision factor inspection, median filter inspection, Rheinland inspection, gradient inspection and visualization inspection, wherein the flow velocity and flow direction of the ground wave radar vector flow data are firstly subjected to format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection and geometric precision factor inspection respectively to obtain the ground wave radar vector flow data after the first inspection, and then the ground wave radar vector flow data after the first inspection is decomposed into the east component and the north component; the flow velocity, east component and north component in the ground wave radar vector flow data after the first inspection are subjected to the median filter inspection and Rheinland inspection of the spatial neighborhood and the Rheinland inspection and gradient inspection of the temporal neighborhood respectively to obtain the ground wave radar vector flow data after the second inspection, and finally the ground wave radar vector flow data after the second inspection is subjected to visualization inspection to obtain the ground wave radar vector flow data after quality control;

[0072] The quality control methods used for the ground wave radar sea surface wind data include format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, median filter inspection, Rheinland inspection, gradient inspection and visualization inspection; wherein the wind speed and wind direction in the ground wave radar sea surface wind data are firstly subjected to format inspection, time range inspection, position inspection, radar effective observation range inspection and threshold inspection respectively, to obtain the ground wave radar sea surface wind data after the first inspection, and then the ground wave radar sea surface wind data after the first inspection is decomposed into east component and north component, and then the wind speed, east component and north component in the ground wave radar sea surface wind data after the first inspection are subjected to median filter inspection and Rheinland inspection in spatial neighborhood and Rheinland inspection and gradient inspection in temporal neighborhood respectively, to obtain the ground wave radar sea surface wind data after the second inspection, and finally the ground wave radar sea surface wind data after the second inspection is subjected to visualization inspection to obtain the ground wave radar sea surface wind data after quality control;

[0073] The quality control methods for ground-wave radar ocean wave data include format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, median filtering inspection, Grubbs' test, gradient inspection, correlation inspection, and visualization inspection. Among them, the significant wave height and wave direction in the ground-wave radar ocean wave data are first respectively subjected to format inspection, time range inspection, position inspection, radar effective observation range inspection, and threshold inspection to obtain the first inspected ground-wave radar ocean wave data. Then, the first inspected ground-wave radar ocean wave data is decomposed into east and north components. Subsequently, median filtering inspection and Grubbs' test in the spatial neighborhood, as well as Grubbs' test and gradient inspection in the time neighborhood, are respectively carried out on the significant wave height, east component, and north component of the first inspected ground-wave radar ocean wave data to obtain the second inspected ground-wave radar ocean wave data. Based on the wind speed in the quality-controlled ground-wave radar sea surface wind data, correlation inspection is carried out on the significant wave height in the second inspected ground-wave radar ocean wave data to obtain the third inspected ground-wave radar ocean wave data. Finally, visualization inspection is performed again on the third inspected ground-wave radar ocean wave data to obtain the quality-controlled ground-wave radar ocean wave data.

[0074] In a specific embodiment, the steps of geometric dilution of precision (GDOP) inspection are as follows:

[0075] The geometric dilution of precision is calculated using the following formula:

[0076]

[0077] where α represents the angle between the line connecting the observation target and the midpoint of the two radar stations and the due east direction, 2θ represents the angle between the line connecting the observation target and the two radar stations, GDOP, GDOP n and GDOP e respectively represent the values of the geometric dilution of precision and its north and east components;

[0078] When GDOP < 4, the inspection result is reliable data;

[0079] When 4 < GDOP ≤ 6, the inspection result is suspicious data;

[0080] When GDOP > 6, the inspection result is incorrect data;

[0081] The steps of median filtering inspection in the spatial neighborhood are as follows:

[0082] For ground wave radar radial flow data: search for the first observation result of the target radar element at the detection angle of view ±1.5×angle resolution and the detection distance ±1.5×distance resolution, calculate the median of the first observation result, if the absolute value of the difference between the observation value of the target radar element and the median of the first observation result is greater than the threshold, the test result is judged as erroneous data; if the absolute value of the difference between the observation value of the target radar element and the median of the first observation result is less than or equal to the threshold, the test result is judged as credible data;

[0083] For ground wave radar vector flow, ground wave radar sea surface wind data and ground wave radar wave data: search for the second observation result of the target grid point within the detection angle of view ±1.5×spatial resolution, calculate the median of the second observation result, if the absolute value of the difference between the observation value of the target grid point and the median of the second observation result is greater than the threshold, the test result is judged as erroneous data, if the absolute value of the difference between the observation value of the target grid point and the median of the second observation result is less than or equal to the threshold, the test result is judged as credible data;

[0084] The steps of the Rheinland test for spatial neighborhood are as follows:

[0085] For the radial flow data of ground wave radar: calculate the mean value and standard deviation of the first observation result. If the difference between the observation value of the target radar element and the mean value of the first observation result is greater than or equal to three times the standard deviation of the first observation result, the test result is wrong data. If the difference between the observation value of the target radar element and the mean value of the first observation result is greater than or equal to two times the standard deviation of the first observation result and less than three times the standard deviation of the first observation result, the test result is suspicious data. If the difference between the observation value of the target radar element and the mean value of the first observation result is less than two times the standard deviation of the first observation result, the test result is reliable data.

[0086] For ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar wave data: calculate the mean value and standard deviation of the second observation result; if the difference between the observation value of the target grid point and the mean value of the second observation result is greater than or equal to three times the standard deviation of the second observation result, the test result is erroneous data; if the difference between the observation value of the target grid point and the mean value of the second observation result is greater than or equal to two times the standard deviation of the second observation result and less than three times the standard deviation of the second observation result, the test result is suspicious data; if the difference between the observation value of the target grid point and the mean value of the second observation result is less than two times the standard deviation of the second observation result, the test result is credible data;

[0087] The steps of the Rheinland test for the time neighborhood are as follows:

[0088] For a certain specified target radar element or target grid point at a certain moment, search for the third observation result within a certain time window before and after the data time point of the target radar element or target grid point, calculate the average value and standard deviation of the third observation result, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is greater than or equal to three times the standard deviation of the third observation result, then the test result is erroneous data, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is greater than or equal to two times the standard deviation of the third observation result and less than three times the standard deviation of the third observation result, then the test result is suspicious data, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is less than two times the standard deviation of the third observation result, then the test result is credible data;

[0089] The steps of the association test are as follows:

[0090] Establish the fitting relationship between wind speed and significant wave height in the target sea area;

[0091] Calculate the 95% prediction interval in the fitting relationship, and combine the wind speed in the surface wave radar sea surface wind data after quality control to infer the reasonable range of the effective wave height data in the surface wave radar sea wave data after the second test;

[0092] Visual inspection involves spatial misjudgment rate and temporal misjudgment rate. The spatial misjudgment rate is for the marine environment data field at each moment, and the temporal misjudgment rate is for the marine environment data field that has misjudgment within the first preset period. The specific definitions are as follows:

[0093] Spatial misjudgment rate = number of misjudged data points in each marine environment data field / total number of valid data points in each marine environment data field × 100%;

[0094] If the spatial misjudgment rate is lower than the first misjudgment threshold, the data of the marine environment data field is regarded as a normal field; if the spatial misjudgment rate is higher than the first misjudgment threshold, the data of the marine environment data field is regarded as a misjudgment field;

[0095] Time misjudgment rate = number of misjudged fields in a certain period of time / total number of marine environment data fields in the same period of time × 100%;

[0096] If the time misjudgment rate is lower than the second misjudgment threshold, the corresponding quality control method is effective. If the time misjudgment rate is higher than the second misjudgment threshold, the corresponding quality control method needs to be optimized until the requirements are met.

[0097] Specifically, the corresponding quality control methods and processes are selected according to different types of data for processing, and the reliable data, suspicious data, erroneous data and missing data after quality control are separated, and the quality identification of different types of data is marked. The identification mark of reliable data is 1, the identification mark of suspicious data is 3, the identification mark of erroneous data is 4, and the identification mark of missing data is 9. The quality control objects include ground wave radar radial flow data, ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar wave data. The quality control of these elements includes the quality control of spatial neighborhood and temporal neighborhood. The quality control methods used mainly include format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, GDOP (geometric dilution of precision) inspection, median filtering inspection, gradient inspection, Rheinda inspection, visualization inspection, etc. Among them, the GDOP inspection only inspects the ground wave radar vector flow data, and the correlation inspection is based on the wind speed of the ground wave radar sea surface wind data after quality control to inspect the effective wave height. The quality control process is as follows Figure 3 shown.

[0098] The embodiments of the present application innovatively introduce the GDOP test method, and perform multi-dimensional quality control on the ground wave radar ocean environment data from the spatial neighborhood (median filter test and Rheinland test) and the temporal neighborhood (Rheinland test and gradient test).

[0099] refer to Figure 4 , GDOP test is: Geometric Dilution of Precision (GDOP) is a unitless coefficient, which characterizes the influence of the relative position between the target position and the radar station on the measurement and positioning errors, and it will affect the uncertainty of the ground wave radar ocean environment data. The lower the GDOP value, the more accurate the observation result of the target position, on the contrary, the greater the uncertainty of the observation result. For the station located on the line connecting the radar stations, since the radial flows observed by the two stations are basically parallel, the synthesized vector flow is inaccurate. In one embodiment, when GDOP>6, the test result is judged as erroneous data and marked as 4, when GDOP≤6, the test result is judged as suspicious data and marked as 3, and when GDOP<4, the test result is judged as credible data and marked as 1.

[0100] As shown in Table 1, the ground wave radar radial flow data of a single station is polar coordinates distributed according to the radar element, and the first observation result within the 3×3 radar elements adjacent to the target radar element is searched. In one embodiment, for the median filter test of the spatial neighborhood, if the observation value of the target radar element exceeds a certain value of the median, such as |observation value-median|>0.5m / s, the test result is judged as erroneous data and marked as 4. The ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar wave data are uniform grid point data (rectangular coordinates) of the synthetic station, and the second observation result within the 3×3 grid points adjacent to the target grid point is searched. In one embodiment, the time window used for the Rhineda test in the time neighborhood is 6 hours before and after the data time point. In the visual inspection, in order to quantitatively judge the rationality of the quality control method and process, the concepts of spatial misjudgment rate and temporal misjudgment rate are introduced respectively.

[0101] Table 1 Quality control methods and quality identification instructions

[0102]

[0103]

[0104] The evaluation module 3 is configured to select different evaluation methods to perform quality evaluation on different types of data in the quality-controlled ground wave radar ocean environment data to obtain corresponding evaluation conclusions.

[0105] In a specific embodiment, different evaluation methods are selected for different types of data in the quality-controlled ground wave radar ocean environment data to perform quality evaluation, and corresponding evaluation conclusions are obtained, specifically including:

[0106] The evaluation method used for the ground wave radar radial flow data after quality control is as follows: statistics on the distribution of spatial coverage of all grid points in the ground wave radar radial flow data after quality control within the first preset period, the spatial coverage of the grid points is the number of radar radial flow grid point samples in the first preset period / the number of radar theoretical sampling events in the first preset period, and the number of radar radial flow grid point samples is the number of credible data;

[0107] Further calculate the angle average spatial coverage rate and the distance average spatial coverage rate. The angle average spatial coverage rate is obtained by dividing the detection viewing angle according to preset intervals, averaging the spatial coverage rates of the grid points within the same detection viewing angle interval, and obtaining the angle average spatial coverage rate. The distance average spatial coverage rate is obtained by dividing the detection distance according to preset intervals, averaging the spatial coverage rates of the grid points within the same detection distance interval, and obtaining the distance average spatial coverage rate. Analyze the changing characteristics of the angle average spatial coverage rate with the detection viewing angle and the changing characteristics of the distance average spatial coverage rate with the detection distance. If the angle average spatial coverage rate gradually decreases with the increase of the detection viewing angle and the distance average spatial coverage rate gradually decreases with the increase of the detection distance, then the detection conditions of the ground wave radar radial flow data after quality control and the corresponding ground wave radar vector flow data after quality control within the first preset time period are normal, otherwise they are abnormal.

[0108] The evaluation method used for the ground wave radar vector flow data after quality control is as follows: the stable observation area of ​​the ground wave radar vector flow data after quality control within the second preset time period is subjected to ocean current reconciliation analysis, the stable observation area is defined as the area where the spatial coverage of the ground wave radar vector flow data is above 70%, the main tidal components are separated according to the tidal characteristics of the sea area, and the tidal ellipse elements of each main tidal component are calculated, the tidal ellipse elements include the maximum tidal component, the minimum tidal component, the ellipticity and the maximum tidal direction, and the numerical value of the tidal ellipse element is compared with the historical numerical value; if the difference between the numerical value of the tidal ellipse element and the historical numerical value is less than or equal to the element threshold area, the detection of the ground wave radar vector flow data after quality control within the second preset time period is normal, and if the numerical value of the tidal ellipse element and the historical numerical value are greater than the element threshold area, the detection of the ground wave radar vector flow data after quality control within the second preset time period is abnormal;

[0109] The evaluation method used for the quality-controlled ground wave radar sea surface wind data is as follows: the quality-controlled ground wave radar sea surface wind data is vector averaged with reference to the temporal and spatial resolution of the ERA5 reanalysis wind field to obtain data consistent with the temporal and spatial resolution of the ERA5 wind field, and then the time series diagrams of wind speed and wind direction of the ERA5 and quality-controlled ground wave radar sea surface wind data in the corresponding grid are compared, and the credibility of the quality-controlled ground wave radar sea surface wind data in the grid point is judged by analyzing whether the change trends between the two are consistent;

[0110] The evaluation method used for the quality-controlled ground wave radar wave data is as follows: the quality-controlled ground wave radar wave data are vector averaged with reference to the spatiotemporal resolution of the ERA5 reanalysis wind field to obtain data consistent with the spatial resolution of the ERA5 wind field. The quality-controlled ground wave radar wave data in the corresponding grid point are then compared with the time series diagram of the ERA5 wind vector. The credibility of the quality-controlled ground wave radar wave data in the grid is judged by analyzing whether the change trends between the two are consistent.

[0111] Specifically, refer to Figure 5 , for the calculation formula of the spatial coverage rate of the grid points in the evaluation method used for the ground wave radar radial flow data after quality control, taking the first preset period of statistics as 1 day as an example, the ground wave radar environmental data is collected once every 10 minutes, and the theoretical sampling times of the first preset period of statistics are 144. The angle average spatial coverage rate is to divide the detection angle into intervals of 5° (such as 0-5°, 5°-10°, etc.), and average the spatial coverage rates of the grid points within the same detection angle interval to obtain the angle average spatial coverage rate; the distance average spatial coverage rate is to divide the detection distance into intervals of 5km (such as 0-5km, 5-10km), and average the spatial coverage rates of the grid points within the same detection distance interval to obtain the distance average spatial coverage rate. If the angle average spatial coverage rate gradually decreases with the increase of the detection angle (the detection angle at the normal direction is 0°) and the distance average spatial coverage rate gradually decreases with the increase of the detection distance, then the detection of the ground wave radar radial flow data and the corresponding ground wave radar vector flow data in the statistical period is normal, otherwise it is abnormal.

[0112] In the evaluation method adopted for the ground wave radar vector flow data after quality control, the second preset period is used for statistics as 1 month, and the ocean current harmonization analysis is performed on the stable observation area of ​​the ground wave radar vector flow data after quality control. According to the tidal characteristics of the sea area, M 2 , S 2 , K 1 , O 1 、M 4 ,MS 4 The main tidal components are calculated, and the tidal ellipse elements of each tidal component, such as the maximum tidal component, the minimum tidal component, the ellipticity, the maximum tidal direction, etc., are calculated, and the numerical values ​​of these tidal ellipse elements are compared with the historical values. Considering the non-uniformity of the plane distribution of the accuracy of ground wave radar vector flow data, that is, the ground wave radar vector flow data has the characteristics of variable accuracy in different detection areas, by comparing the numerical values ​​of the tidal ellipse elements of the ground wave radar vector flow data with the historical values, the detection accuracy of the ground wave radar is higher in areas close to the historical values, and the detection accuracy of the ground wave radar is lower in areas with large numerical differences. This method can comprehensively evaluate the overall quality of the data during the statistical period and the reliability of the ground wave radar vector flow data in different observation areas.

[0113] In the evaluation method used for the quality-controlled ground wave radar sea surface wind data, the quality-controlled ground wave radar sea surface wind data is vector averaged according to the spatial resolution of 25km and the temporal resolution of 1h to obtain data consistent with the temporal and spatial resolution of the ERA5 wind field. Then, the time series of wind speed and wind direction of ERA5 and ground wave radar in the corresponding grid are compared. The credibility of the radar wind field in the grid is judged by analyzing whether the change trends between the two are consistent. Correlation analysis is performed on the ERA5 data with wind speeds above 5m / s. If the correlation coefficient is above 0.75, the quality-controlled ground wave radar sea surface wind data in the grid is more credible. If the correlation coefficient is between 0.5 and 0.75, the quality-controlled ground wave radar sea surface wind data in the grid should be used with caution. If the correlation coefficient is below 0.5, the quality-controlled ground wave radar sea surface wind data in the grid is unusable.

[0114] In the evaluation method used for the quality-controlled ground wave radar wave data, when the wind speed is greater than a certain value, there is a good correlation between the wave height and wind speed, and between the wave vector (the wave height and wave direction are regarded as "wave vector") and the wind vector. Therefore, the quality-controlled ground wave radar wave data are vector averaged according to the spatial resolution of 25km and the temporal resolution of 1h to obtain data consistent with the spatial resolution of the ERA5 wind field. Then, the time series diagrams of the radar wave vector and the ERA5 wind vector in the corresponding grid are compared. The reliability of the quality-controlled ground wave radar wave data in the grid is judged by analyzing whether the change trends between the two are consistent. The correlation analysis is performed on the ERA5 data with wind speeds above 5m / s. If the correlation coefficient is above 0.75, the quality-controlled ground wave radar wave data in the grid is relatively reliable. If the correlation coefficient is between 0.5 and 0.75, the quality-controlled ground wave radar wave data in the grid should be used with caution. If the correlation coefficient is below 0.5, the quality-controlled ground wave radar wave data in the grid is unusable.

[0115] The scheme of the present invention is further described below with specific embodiments.

[0116] Taking the ground wave radar vector flow data of Dongshan and Longhai radar stations in Fujian at 14:20 on August 6, 2021 as an example, the comparison of data before and after quality control is shown. Figure 6 shown.

[0117] Taking the results of ground wave radar vector flow data from Dongshan and Longhai radar stations in Fujian from August 6 to August 31, 2021 as an example, the tidal ellipses of M2, S2, K1 and O1 tides are obtained as follows: Figure 7 As shown in the figure, by comparing with historical values, the evaluation conclusion of this period shows that for observation areas with a spatial coverage rate of more than 70% during the statistical period, the ground wave radar vector flow data is more reliable.

[0118] The quality-controlled ground wave radar sea surface wind data are vector averaged with reference to the spatiotemporal resolution of the ERA5 reanalysis wind field, that is, the quality-controlled ground wave radar sea surface wind data are vector averaged according to the spatial resolution of 25km and the temporal resolution of 1h to obtain data consistent with the spatiotemporal resolution of the ERA5 wind field, and then the time series diagrams of wind speed and wind direction of the ERA5 and ground wave radar sea surface wind data in the corresponding grid are compared. The credibility of the quality-controlled ground wave radar sea surface wind data in the grid is judged by analyzing whether the change trends between the two are consistent. Correlation analysis is performed on the ERA5 data with wind speeds above 5m / s, such as Figure 8 and Fig. 9 As shown in the figure, if the correlation coefficient is above 0.75, the surface wave radar sea surface wind data after quality control in the grid is more reliable; if the correlation coefficient is between 0.5 and 0.75, the surface wave radar sea surface wind data and / or the surface wave radar sea surface wind data after quality control in the grid should be used with caution; Fig.10 and Fig.11 As shown in Figure 1, if the correlation coefficient is below 0.5, the quality-controlled ground wave radar sea surface wind data in the grid is unusable.

[0119] The evaluation method of the ground wave radar wave data after quality control and the evaluation method of the ground wave radar sea surface wind data after quality control will not be repeated here.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-dimensional data quality control and evaluation system for ground wave radar based on radio waves and ocean characteristics, characterized in that: include: A data acquisition module is configured to acquire different types of ground wave radar ocean environment data obtained through preprocessing, wherein the data types of the ground wave radar data include ground wave radar radial flow data, ground wave radar vector flow data, ground wave radar sea surface wind data and ground wave radar sea wave data; A data quality control module is configured to respectively test different types of data in the ground wave radar ocean environment data using different quality control methods and processes, and separate the quality-controlled ground wave radar ocean environment data with quality identification according to the discrimination criteria set by the quality control method, wherein the quality control method includes format test, time range test, position test, radar effective observation range test, threshold test, geometric precision factor test, median filter test, Rheinland test, gradient test, correlation test and visualization test; The data evaluation module is configured to select different evaluation methods to perform quality evaluation on different types of data in the quality-controlled ground wave radar ocean environment data to obtain corresponding evaluation conclusions.

2. The ground wave radar multi-dimensional data quality control and evaluation system based on radio waves and ocean characteristics according to claim 1 is characterized in that: Different types of data in the ground wave radar ocean environment data are inspected using different quality control methods and processes, and the ground wave radar ocean environment data after quality control with quality identification is separated according to the discrimination criteria set by the quality control method, specifically including: The quality control method used for the ground wave radar radial flow data includes format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, median filter inspection, Rheinland inspection, gradient inspection and visualization inspection, so as to obtain the ground wave radar radial flow data after quality control; The quality control method adopted by the ground wave radar vector flow data includes format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection, geometric precision factor inspection, median filtering inspection, Rheinland inspection, gradient inspection and visualization inspection, wherein the flow velocity and flow direction of the ground wave radar vector flow data are firstly subjected to format inspection, time range inspection, position inspection, radar effective observation range inspection, threshold inspection and geometric precision factor inspection respectively to obtain the ground wave radar vector flow data after the first inspection, and then the ground wave radar vector flow data after the first inspection is decomposed into the east component and the north component; the flow velocity, the east component and the north component in the ground wave radar vector flow data after the first inspection are respectively subjected to the median filtering inspection and Rheinland inspection of the spatial neighborhood and the Rheinland inspection and gradient inspection of the temporal neighborhood to obtain the ground wave radar vector flow data after the second inspection, and finally the ground wave radar vector flow data after the second inspection is subjected to visualization inspection to obtain the ground wave radar vector flow data after quality control; The quality control method for the ground wave radar sea surface wind data includes format check, time range check, position check, radar effective observation range check, threshold check, median filtering check, Grubbs' test, gradient check, and visualization check; among which, the wind speed and wind direction in the ground wave radar sea surface wind data are first respectively subjected to format check, time range check, position check, radar effective observation range check, and threshold check to obtain the ground wave radar sea surface wind data after the first check. Then, the ground wave radar sea surface wind data after the first check is decomposed into east component and north component. Then, the median filtering check and Grubbs' test in the spatial neighborhood and the Grubbs' test and gradient check in the time neighborhood are respectively performed on the wind speed, east component, and north component in the ground wave radar sea surface wind data after the first check to obtain the ground wave radar sea surface wind data after the second check. Finally, the visualization check is performed on the ground wave radar sea surface wind data after the second check to obtain the quality-controlled ground wave radar sea surface wind data; The quality control method for the ground wave radar sea wave data includes format check, time range check, position check, radar effective observation range check, threshold check, median filtering check, Grubbs' test, gradient check, correlation check, and visualization check; among which, the significant wave height and wave direction in the ground wave radar sea wave data are first respectively subjected to format check, time range check, position check, radar effective observation range check, and threshold check to obtain the ground wave radar sea wave data after the first check. Then, the ground wave radar sea wave data after the first check is decomposed into east component and north component. Then, the median filtering check and Grubbs' test in the spatial neighborhood and the Grubbs' test and gradient check in the time neighborhood are respectively performed on the significant wave height, east component, and north component in the ground wave radar sea wave data after the first check to obtain the ground wave radar sea wave data after the second check. Based on the wind speed in the quality-controlled ground wave radar sea surface wind data, the correlation check is performed on the significant wave height in the ground wave radar sea wave data after the second check to obtain the ground wave radar sea wave data after the third check. Finally, the visualization check is performed again on the ground wave radar sea wave data after the third check to obtain the quality-controlled ground wave radar sea wave data.

3. The ground wave radar multi-dimensional data quality control and evaluation system based on radio waves and ocean characteristics according to claim 2 is characterized in that: The quality identifier includes identifiers for marking reliable data, suspicious data, error data, and missing data.

4. The multi-dimensional data quality control and evaluation system of ground wave radar based on radio waves and ocean characteristics according to claim 3 is characterized in that: The steps of the geometric dilution of precision (GDOP) check are as follows: Calculate the geometric dilution of precision using the following formula: Among them, α represents the angle between the line connecting the observation target and the middle position of the two radar stations and the east direction, 2θ represents the angle between the observation target and the line connecting the two radar stations, GDOP, GDOP n and GDOP e Respectively represent the values ​​of the geometric dilution of precision and its north component and east component; When GDOP < 4, the test result is reliable data; When 4 < GDOP ≤ 6, the test result is suspicious data; When GDOP > 6, the test result is error data; The steps of the median filtering check in the spatial neighborhood are as follows: For the ground wave radar radial flow data: searching for a first observation result of the target radar element at a detection angle of view ±1.5×angle resolution and a detection distance ±1.5×distance resolution, calculating a median of the first observation result, if the absolute value of the difference between the observation value of the target radar element and the median of the first observation result is greater than a threshold, the test result is judged to be erroneous data; if the absolute value of the difference between the observation value of the target radar element and the median of the first observation result is less than or equal to the threshold, the test result is judged to be credible data; For the ground wave radar vector flow, ground wave radar sea surface wind data and ground wave radar wave data: search for the second observation result of the target grid point at the detection angle of view ±1.5×spatial resolution, calculate the median of the second observation result, if the absolute value of the difference between the observation value of the target grid point and the median of the second observation result is greater than a threshold, the test result is judged to be erroneous data, if the absolute value of the difference between the observation value of the target grid point and the median of the second observation result is less than or equal to the threshold, the test result is judged to be credible data; The steps of the Rheinland test for spatial neighborhood are as follows: For the ground wave radar radial flow data: calculating the mean value and standard deviation of the first observation result; if the difference between the observation value of the target radar element and the mean value of the first observation result is greater than or equal to three times the standard deviation of the first observation result, the test result is erroneous data; if the difference between the observation value of the target radar element and the mean value of the first observation result is greater than or equal to two times the standard deviation of the first observation result and less than three times the standard deviation of the first observation result, the test result is suspicious data; if the difference between the observation value of the target radar element and the mean value of the first observation result is less than two times the standard deviation of the first observation result, the test result is credible data; For the ground wave radar vector flow data, the ground wave radar sea surface wind data and the ground wave radar wave data: calculate the mean value and standard deviation of the second observation result; if the difference between the observation value of the target grid point and the mean value of the second observation result is greater than or equal to three times the standard deviation of the second observation result, the test result is erroneous data; if the difference between the observation value of the target grid point and the mean value of the second observation result is greater than or equal to two times the standard deviation of the second observation result and less than three times the standard deviation of the second observation result, the test result is suspicious data; if the difference between the observation value of the target grid point and the mean value of the second observation result is less than two times the standard deviation of the second observation result, the test result is credible data; The steps of the Rheinland test for the time neighborhood are as follows: For a certain specified target radar element or target grid point at a certain moment, search for the third observation result within a certain time window before and after the data time point of the target radar element or target grid point, calculate the average value and standard deviation of the third observation result, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is greater than or equal to three times the standard deviation of the third observation result, then the test result is erroneous data, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is greater than or equal to two times the standard deviation of the third observation result and less than three times the standard deviation of the third observation result, then the test result is suspicious data, if the difference between the observation value of the target radar element or target grid point and the average value of the third observation result is less than two times the standard deviation of the third observation result, then the test result is credible data; The steps of the association test are as follows: Establish the fitting relationship between wind speed and significant wave height in the target sea area; Calculate the 95% prediction interval in the fitting relationship, and infer the reasonable range of the significant wave height data in the ground wave radar sea wave data after the second inspection in combination with the wind speed in the ground wave radar sea surface wind data after quality control; The visualization test involves spatial misjudgment rate and temporal misjudgment rate. The spatial misjudgment rate is for the ocean environment data field at each moment, and the temporal misjudgment rate is for the ocean environment data field that has misjudgment within the first preset period. The specific definitions are as follows: Spatial misjudgment rate = number of misjudged data points in each marine environment data field / total number of valid data points in each marine environment data field × 100%; If the spatial misjudgment rate is lower than the first misjudgment threshold, the data of the marine environment data field is regarded as a normal field; if the spatial misjudgment rate is higher than the first misjudgment threshold, the data of the marine environment data field is regarded as a misjudgment field; Time misjudgment rate = number of misjudged fields in a certain period of time / total number of marine environment data fields in the same period of time × 100%; If the time misjudgment rate is lower than the second misjudgment threshold, the corresponding quality control method is effective. If the time misjudgment rate is higher than the second misjudgment threshold, the corresponding quality control method needs to be optimized until the requirements are met.

5. The ground wave radar multi-dimensional data quality control and evaluation system based on radio waves and ocean characteristics according to claim 3 is characterized in that: Different evaluation methods are selected for different types of data in the quality-controlled ground wave radar ocean environment data to perform quality evaluation, and corresponding evaluation conclusions are obtained, including: The evaluation method adopted for the quality-controlled ground wave radar radial flow data is as follows: statistics on the distribution of spatial coverage rates of all grid points in the quality-controlled ground wave radar radial flow data within a first preset period, the spatial coverage rate of the grid points is the number of radar radial flow grid point samples within the first preset period / the number of radar theoretical sampling events within the first preset period, and the number of radar radial flow grid point samples is the number of credible data; Further calculate the angle average spatial coverage rate and the distance average spatial coverage rate. The angle average spatial coverage rate is obtained by dividing the detection viewing angle according to preset intervals, averaging the spatial coverage rates of the grid points within the same detection viewing angle interval, and obtaining the angle average spatial coverage rate. The distance average spatial coverage rate is obtained by dividing the detection distance according to preset intervals, averaging the spatial coverage rates of the grid points within the same detection distance interval, and obtaining the distance average spatial coverage rate. Analyze the changing characteristics of the angle average spatial coverage rate with the detection viewing angle and the changing characteristics of the distance average spatial coverage rate with the detection distance. If the angle average spatial coverage rate gradually decreases with the increase of the detection viewing angle and the distance average spatial coverage rate gradually decreases with the increase of the detection distance, then the detection conditions of the ground wave radar radial flow data after quality control and the corresponding ground wave radar vector flow data after quality control within the first preset time period are normal, otherwise they are abnormal. The evaluation method adopted by the quality-controlled ground wave radar vector flow data is as follows: the stable observation area of ​​the quality-controlled ground wave radar vector flow data within the second preset time period is subjected to ocean current reconciliation analysis, the stable observation area is defined as the area where the spatial coverage rate of the ground wave radar vector flow data is above 70%, the main tidal components are separated according to the tidal characteristics of the sea area, and the tidal ellipse elements of each main tidal component are calculated, the tidal ellipse elements include the maximum tidal component, the minimum tidal component, the ellipticity and the maximum tidal direction, and the numerical value of the tidal ellipse element is compared with the historical numerical value; if the difference between the numerical value of the tidal ellipse element and the historical numerical value is less than or equal to the element threshold area, the quality-controlled ground wave radar vector flow data within the second preset time period is detected normally, and if the numerical value of the tidal ellipse element and the historical numerical value are greater than the element threshold area, the quality-controlled ground wave radar vector flow data within the second preset time period is detected abnormally; The evaluation method used for the quality-controlled ground wave radar sea surface wind data is as follows: the quality-controlled ground wave radar sea surface wind data is vector averaged with reference to the spatiotemporal resolution of the ERA5 reanalysis wind field to obtain data consistent with the spatiotemporal resolution of the ERA5 wind field, and then the time series diagrams of wind speed and wind direction of the ERA5 and quality-controlled ground wave radar sea surface wind data in the corresponding grid are compared, and the credibility of the quality-controlled ground wave radar sea surface wind data in the grid point is judged by analyzing whether the change trends between the two are consistent; The evaluation method used for the quality-controlled ground wave radar wave data is as follows: the quality-controlled ground wave radar wave data is vector averaged with reference to the spatiotemporal resolution of the ERA5 reanalysis wind field to obtain data consistent with the spatial resolution of the ERA5 wind field, and then the quality-controlled ground wave radar wave data in the corresponding grid point is compared with the time series diagram of the ERA5 wind vector, and the credibility of the quality-controlled ground wave radar wave data in the grid is judged by analyzing whether the change trends between the two are consistent.

6. The ground wave radar multi-dimensional data quality control and evaluation system based on radio waves and ocean characteristics according to claim 1 is characterized in that: The preprocessing process includes checking the readability, completeness and format standardization of the ground wave radar ocean environment data, decoding, sorting and deduplication processing, and format conversion according to the data file format required for quality control to obtain a data file with a unified format.

Images