Argo buoy real-time quality control method and system

Abstract

The application relates to an Argo buoy real-time quality control method and system, the method combines the best matching profile of the historical trajectory of the to-be-detected Argo buoy with WOA climatic data to calculate the probability distribution of observation data, the two data and the corresponding data extraction method are used, the positioning of surrounding historical observation elements of the moving Argo buoy and the extraction of the corresponding verification data set can be effectively carried out, and the accuracy of subsequent use of the probability distribution to judge suspicious profiles can be effectively improved. Moreover, the akima interpolation method is used to construct the probability distribution in the final maximum and minimum boundary including the measurement area and the non-measurement area, so that the constructed probability distribution is more in line with the actual situation, and the reliability of subsequent use of the probability distribution to judge suspicious profiles can be effectively improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of marine data processing, in particular to an Argo buoy real-time quality control method and system. BACKGROUND

[0002] Argo buoys can capture real-time data of water pressure, salinity and temperature of ocean currents within 2000 meters of the upper ocean, which helps to further understand the structure and characteristics of seawater. The data detected by Argo buoys is of great significance to understanding the changes of marine environment, ocean circulation and climate change, understanding and predicting the changes of climate system, understanding the interaction between ocean and atmosphere, and verifying and improving the ocean model. Since the measurement sensors carried by Argo buoys are easily affected by abnormal conditions such as biological contamination and biological pesticide leakage, resulting in abnormal values of measurement results, it is necessary to use quality control means to identify and mark suspicious profiles.

[0003] The existing buoy quality control methods mainly include two methods: delay quality control and real-time quality control. In the delay quality control, data analysts perform various correction algorithms and methods to identify and correct potential problems or errors, which may involve analyzing and processing abnormal values, drift, sensor calibration problems and other data. At the same time, delay quality detection may also need to be compared and verified with other data sources to ensure the consistency and accuracy of the data. However, due to the large number of processes involved, the human resources are consumed greatly, and it often requires very high labor cost and time consumption, which is difficult to achieve efficient business purposes. Real-time quality detection is a quality control method that is performed in real time during data collection. It monitors and checks data in real time through sensors and measurement equipment on the buoy to identify potential data problems or errors. Real-time quality detection helps to quickly find data anomalies and can take timely measures to repair or exclude abnormal data. However, due to the timeliness of real-time quality detection, simple threshold judgment is often used to process the data to be detected, which has a certain scientific nature and has reached a relatively mature business, but its principle is mostly for single profile and often difficult to consider multiple factors, especially the profile characteristics from the ground, so there is still a lot of room for improvement.

[0004] In summary, according to the existing Argo buoy quality control methods, real-time quality control has the problems of simple method and insufficient consideration of problems, which can only make preliminary judgments; delay quality control has the problems of insufficient timeliness and large consumption of human resources. Therefore, it is very important to find a new method that has good timeliness and can consider relatively comprehensive characteristics. SUMMARY

[0005] The following is a summary of the subject matter described in detail in this document. This summary is not intended to limit the scope of protection of the claims.

[0006] The main purpose of the embodiment of the present application is to provide an Argo buoy real-time quality control method, which can effectively improve the accuracy and reliability of real-time quality control of Argo buoy.

[0007] To achieve the above purpose, the first aspect of the embodiment of the present application provides an Argo buoy real-time quality control method, which comprises:

[0008] Obtaining a WOA climatological dataset, a profile of an Argo buoy to be measured, and a historical observation dataset of the Argo buoy to be measured;

[0009] Finding a corresponding best matching profile from the historical observation dataset according to the observation data in the profile of the Argo buoy to be measured, and determining the extreme value boundary of the best matching profile, wherein the extreme value boundary is a boundary for characterizing whether the observation data in the best matching profile is 0;

[0010] Finding a corresponding target climatological data from the WOA climatological dataset according to the observation data in the profile of the Argo buoy to be measured, and determining a climatological extreme value boundary according to the target climatological data;

[0011] Comparing the extreme value boundary of the best matching profile and the climatological extreme value boundary, and determining a final extreme value boundary according to the comparison result;

[0012] According to the final extreme value boundary and the frequency number distribution of the observation data in the best matching profile, the probability distribution of the observation data in the best matching profile is calculated by akima interpolation;

[0013] According to the probability distribution, a discrimination boundary is calculated, and whether the profile of the Argo buoy to be measured is a suspicious profile is judged according to the observation data in the profile of the Argo buoy to be measured and the discrimination boundary.

[0014] In some embodiments, after finding the corresponding best matching profile from the historical observation dataset, the Argo buoy real-time quality control method further comprises:

[0015] The observation data in the best matching profile is interpolated to each depth by akima interpolation;

[0016] The observation data of the best matching profile in adjacent two water layers is fused.

[0017] In some embodiments, the determination of the extreme value boundary of the best matching profile comprises:

[0018] Obtaining the measurement statistical result of the observation data of the best matching profile in each water layer;

[0019] frequency normalization is performed on the measurement statistics, and the maximum and minimum values of the observation data in each water layer of the best matching profile are extracted;

[0020] According to the maximum and minimum values of the observation data extracted in each water layer, the peak interval with the highest statistical frequency of the observation data in each water layer is determined;

[0021] The difference between the peak interval and the first end maximum value and the difference between the peak interval and the second end maximum value are determined, wherein the first end maximum value and the second end maximum value are the two end maximum values in the same water layer in opposite directions.

[0022] The first end maximum value is extended outward by a first distance, and the second end maximum value is extended outward by a second distance, to obtain the maximum and minimum value boundaries of the best matching profile in each water layer, wherein the first distance is equal to the first difference multiplied by a predetermined proportion, and the second distance is equal to the second difference multiplied by the proportion.

[0023] In some embodiments, the corresponding target climatological data is found from the WOA climatological data set according to the observation data in the to-be-tested Argo float profile, and the climatological maximum and minimum value boundaries are determined according to the target climatological data, comprising:

[0024] Data interpolation is performed on the WOA climatological data set to obtain a full map data set;

[0025] According to the month, longitude and latitude of the observation data in the to-be-tested Argo float profile, the corresponding target climatological data is found from the full map data set;

[0026] The average value and standard deviation of the observation data corresponding to the target climatological data are calculated;

[0027] The average value is added or subtracted by several times the standard deviation to obtain the climatological maximum and minimum value;

[0028] The climatological maximum and minimum value boundary is generated according to the calculated climatological maximum and minimum value.

[0029] In some embodiments, the best matching profile maximum and minimum value boundary and the climatological maximum and minimum value boundary are compared, and the final maximum and minimum value boundary is determined according to the comparison result, comprising:

[0030] Each first maximum value constituting the best matching profile maximum and minimum value boundary and each second maximum value constituting the climatological maximum and minimum value boundary are determined

[0031] In each water layer, the two maximum values located at both ends of the first maximum value and the second maximum value are found.

[0032] The final maximum boundary is formed based on the two maximum values ​​found in each water layer.

[0033] In some embodiments, calculating the determination boundary of the Argo float profile to be measured according to the probability distribution includes:

[0034] According to the probability distribution of each water layer, selection is made based on the probability from high to low, and a discriminant number is assigned to the observation data according to the total probability of the selection to obtain a first assignment result;

[0035] According to the probability distribution of each water layer, the probability distribution is accumulated from small to large based on the concentration. When the total probability reaches 50%, it is used as the intermediate starting cumulative probability calculation position, and the probabilities on the left and right sides of the intermediate starting cumulative probability calculation position are compared. Based on the cumulative probabilities in the order of the comparison results, the cumulative probabilities from the middle to both sides are obtained; based on the final cumulative probability, the observed data is assigned a discriminant number to obtain a second assignment result;

[0036] The first assignment result and the second assignment result are each weighted 50% and averaged to obtain a third assignment result;

[0037] The observation data whose third assignment result is less than the preset first threshold is used as the judgment boundary.

[0038] In some embodiments, judging whether the Argo float profile to be measured is a suspicious profile based on the observation data in the Argo float profile to be measured and the judgment boundary includes:

[0039] If the observation data in the Argo float profile to be measured is outside the determination boundary, determining that the measurement point where the observation data in the Argo float profile to be measured is located is a suspicious measurement point;

[0040] Determining a control depth of the suspicious measuring point, wherein the control depth is half of the sum of a depth difference between the suspicious measuring point and the previous measuring point plus a depth difference between the suspicious measuring point and the next measuring point;

[0041] If a plurality of consecutive measuring points are all the suspicious measuring points and / or the sum of the number of the controlled depths reaches a second threshold, the Argo float profile to be measured is determined to be a suspicious profile.

[0042] A second aspect of an embodiment of the present invention provides an Argo float real-time quality control system, the Argo float real-time quality control system comprising:

[0043] Data acquisition unit, used to obtain WOA climate state data set, Argo float profile to be measured and historical observation data set of Argo float to be measured;

[0044] The first extreme boundary calculation unit is configured to find a corresponding best matching profile from the historical observation data set according to the observation data in the to-be-tested Argo float profile, and determine an extreme boundary of the best matching profile, wherein the extreme boundary is a boundary for representing whether the possibility of the observation data in the best matching profile is 0.

[0045] The second extreme boundary calculation unit is configured to find corresponding target climatological data from the WOA climatological data set according to the observation data in the to-be-tested Argo float profile, and determine a climatological extreme boundary according to the target climatological data.

[0046] The third extreme boundary calculation unit is configured to compare the extreme boundary of the best matching profile and the climatological extreme boundary, and determine a final extreme boundary according to a comparison result.

[0047] The probability distribution calculation unit is configured to calculate a probability distribution of the observation data in the best matching profile by using akima interpolation according to the final extreme boundary and a frequency number distribution of the observation data in the best matching profile.

[0048] The suspicious profile detection unit is configured to calculate a judgment boundary according to the probability distribution, and judge whether the to-be-tested Argo float profile is a suspicious profile according to the observation data in the to-be-tested Argo float profile and the judgment boundary.

[0049] To achieve the above object, a third aspect of embodiments of the present application provides an electronic device, comprising: at least one control processor and a memory connected in communication with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the Argo float real-time quality control method.

[0050] To achieve the above object, a fourth aspect of embodiments of the present application provides a computer readable storage medium, which stores computer executable instructions for enabling a computer to execute the Argo float real-time quality control method.

[0051] An embodiment of the present application provides an Argo buoy real-time quality control method, which combines the best matching profile of a to-be-detected Argo buoy historical trajectory and WOA climatological data to calculate the probability distribution of observation data, uses the two data and corresponding data extraction methods, can effectively position the surrounding historical observation elements of the moving Argo buoy and extract the corresponding verification data set, and can effectively improve the accuracy of subsequent use of the probability distribution to judge suspicious profiles. Moreover, the akima interpolation method is used to construct the probability distribution in the final maximum and minimum boundary including the measurement area and the non-measurement area, so that the constructed probability distribution is more in line with the actual situation, and the reliability of subsequent use of the probability distribution to judge suspicious profiles can be effectively improved.

[0052] It can be understood that the beneficial effects of the second aspect to the fourth aspect and related technologies compared with the first aspect and related technologies are the same, and can be referred to the related description in the first aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments or related technical description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0054] Figure 1 is a flowchart of an Argo buoy real-time quality control method provided by an embodiment of the present application;

[0055] Figure 2 is a flowchart of the best matching profile processing provided by an embodiment of the present application;

[0056] Figure 3 is Figure 1 is a flowchart of determining the maximum and minimum boundary of the best matching profile in step S120 in the embodiment;

[0057] Figure 4 is Figure 1 is a flowchart of generating the climatological maximum and minimum boundary in step S130 in the embodiment;

[0058] Figure 5 is Figure 1 is a flowchart of the final maximum and minimum boundary in step S140 in the embodiment;

[0059] Figure 6 is Figure 1 is a flowchart of calculating the discrimination boundary in step S160 in the embodiment;

[0060] Figure 7 is Figure 1 A flowchart of calculating whether the profile of the Argo buoy to be measured is a suspicious profile in step S160 in the method;

[0061] Figure 8 is a schematic diagram of finding the best matching profile of the Argo buoy profile project_29010 provided by an embodiment of the present application;

[0062] Figure 9 is a data statistical diagram in the Argo buoy profile project_29010 provided by an embodiment of the present application;

[0063] Figure 10 is a data statistical diagram of the best matching profile of the Argo buoy profile project_29010 provided by an embodiment of the present application;

[0064] Figure 11 is a schematic diagram of the extreme boundary of the Argo buoy profile project_29010 provided by an embodiment of the present application;

[0065] Figure 12 is a schematic diagram of the final extreme boundary corresponding to two Argo buoy profile projects provided by an embodiment of the present application;

[0066] Figure 13 is a schematic diagram of the probability distribution of the cross-section calculation of the Argo buoy profile project_29077 at-100dbar and the probability distribution of the cross-section calculation of the Argo buoy profile project_290406 at-400dbar provided by an embodiment of the present application;

[0067] Figure 14 is a schematic diagram of assigning the discriminant number of the Argo buoy profile project_29044 provided by an embodiment of the present application;

[0068] Figure 15 is a schematic diagram of the discrimination result of the continuous point discrimination method provided by an embodiment of the present application;

[0069] Figure 16 is a schematic diagram of the discrimination result and effect of the control depth discrimination method provided by an embodiment of the present application;

[0070] Figure 17 is a flowchart of an Argo buoy real-time quality control method provided by another embodiment of the present application;

[0071] Figure 18 is a structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0072] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not intended to limit the present application.

[0073] It should be noted that although the functional modules are divided in the device schematic diagram, and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a manner different from the module division in the device or the sequence in the flowchart. The terms "first", "second", etc. in the specification and claims and the above-described drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0074] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0075] Argo floats can capture real-time data of water pressure, salinity and temperature of ocean currents within 2000 meters of the upper ocean, which helps to further understand the structure and characteristics of the ocean. The data detected by Argo floats is of great significance to understanding the changes of marine environment, ocean circulation and climate change, understanding and predicting the changes of climate system, understanding the interaction between ocean and atmosphere, and verifying and improving the ocean model. Since the measurement sensors carried by Argo floats are easily affected by abnormal conditions such as biological contamination and biological pesticide leakage, resulting in abnormal values of measurement results, it is necessary to use quality control means to identify and mark suspicious profiles.

[0076] The existing buoy quality control methods mainly include two methods: delayed quality control and real-time quality control. In the delayed quality control, data analysts perform various correction algorithms and methods to identify and correct potential problems or errors, which may involve analyzing and processing outliers, drifts, sensor calibration problems, and the like of the data. Meanwhile, the delayed quality control may also need to be compared and verified with other data sources to ensure the consistency and accuracy of the data. However, due to the large number of processes involved, the human resources are consumed greatly, and it often requires extremely high labor cost and time consumption, which is difficult to achieve efficient business purposes. Real-time quality control is a quality control method performed in real time during data collection. It monitors and checks the data in real time through sensors and measuring devices on the buoy to identify potential data problems or errors. Real-time quality control helps to quickly find data anomalies and can take timely measures to repair or exclude abnormal data. However, due to the timely nature of real-time quality control, it often uses simple threshold judgment to process the data to be detected, which has a certain scientific nature and has reached a relatively mature business, but its principle is mostly for a single profile and often difficult to consider multiple factors, especially the profile characteristics from the ground, so there is still a lot of room for improvement.

[0077] In summary, according to the existing Argo buoy quality control methods, real-time quality control has the problems of simple method and insufficient consideration, which can only make preliminary judgments; delayed quality control has the problems of insufficient timeliness and large consumption of human resources. Therefore, it is very important to find a new method that has good timeliness and can consider relatively comprehensive characteristics.

[0078] In order to solve the above-mentioned defects, with reference to Figure 1 An embodiment of the present application provides an Argo buoy real-time quality control method, which comprises the following steps S110-S160:

[0079] Step S110, obtaining a WOA climatological dataset, a profile to be detected of an Argo buoy, and a historical observation dataset of the Argo buoy to be detected.

[0080] Step S120, finding a corresponding best matching profile from the historical observation dataset according to the observation data in the profile to be detected of the Argo buoy, and determining a maximum value boundary of the best matching profile, wherein the maximum value boundary is a boundary used to represent whether the possibility of the observation data in the best matching profile is 0.

[0081] Step S130, finding a corresponding target climatological data from the WOA climatological dataset according to the observation data in the profile to be detected of the Argo buoy, and determining a climatological maximum value boundary according to the target climatological data.

[0082] Step S140, comparing the extremum boundary of the best matching profile and the climatological extremum boundary, and determining the final extremum boundary according to the comparison result.

[0083] Step S150, calculating the probability distribution of the observation data in the best matching profile according to the final extremum boundary and the frequency distribution of the observation data in the best matching profile by using akima interpolation.

[0084] Step S160, calculating the discrimination boundary according to the probability distribution, and judging whether the to-be-tested Argo float profile is a suspicious profile according to the observation data in the to-be-tested Argo float profile and the discrimination boundary.

[0085] The following describes step S110:

[0086] The WOA climatological dataset can be downloaded at a specific website: https: / / www.ncei.noaa.gov / access / world-ocean-atlas-2023 / bin / woa23.pl .

[0087] The to-be-tested Argo float measures the profile data of the observation elements in the sea at different observation profile positions (determined according to the latitude and longitude) in the sea, where the observation elements include but are not limited to water pressure, salinity, temperature, etc., and the following embodiments will use salinity for description.

[0088] One profile measured by the to-be-tested Argo float contains observation data of observation elements in different water layers (or depths), that is, there are n measurement points in the profile, and each measurement point has corresponding observation data, for example, the salinity value measured at the water layer of 300 meters is 33.5, and the salinity value measured at the water layer of 330 meters is 33.6. As the to-be-tested Argo float moves in the sea, it will measure multiple to-be-tested Argo float profiles, and each to-be-tested Argo float profile has multiple observation data (measured at the measurement points).

[0089] The historical observation data set of the to-be-tested Argo float reflects the observation data collected in the marine moving history track of the to-be-tested Argo float.

[0090] The following describes step S120:

[0091] This embodiment uses the historical observation data set to find the best matching profile best matching the to-be-tested Argo float profile from the historical observation data set, which mainly uses the historical "best matching" profile method. The matching method is well known in the field, and will not be described here.

[0092] Then, according to the observation data in the best matching profile, the extremum boundary of the best matching profile is determined. The extremum refers to the maximum value and the minimum value of the observation element (such as salinity) in a water layer (a level divided according to depth). For example, if the Argo float to be measured measures multiple Argo float profiles to be measured, then multiple best matching profiles corresponding to the historical observation data set need to be found, the maximum value of salinity in the same water layer in the multiple best matching profiles is max, and the minimum value of salinity is min, and then max and min are taken as the extremum of the same water layer. The extremum boundary of the multiple best matching profiles is obtained by combining the extremums of different water layers. Obviously, the probability of the observation data being outside the extremum boundary is 0.

[0093] Referring to Figure 2 In some embodiments, after step S120 finds the corresponding best matching profile from the historical observation data set, the method further includes steps S210-S220:

[0094] Step S210, the observation data in the best matching profile is interpolated to each depth by using akima interpolation.

[0095] Step S220, the observation data in the adjacent two water layers of the best matching profile is fused.

[0096] Through such processing, the found best matching profile is suitable for the full depth and contains the continuity characteristics of different water layers.

[0097] Referring to Figure 3 In some embodiments, step S120 of determining the extremum boundary of the best matching profile includes steps S250-S290:

[0098] Step S250, the statistical result of the measure of the observation data in each water layer of the best matching profile is obtained.

[0099] Step S260, the statistical result of the measure is frequency normalized, and the extremum of the observation data in each water layer of the best matching profile is extracted.

[0100] Step S270, according to the extracted extremum of the observation data in each water layer, the peak interval with the highest statistical frequency of the observation data in each water layer is determined.

[0101] Step S280, the difference between the peak interval and the two end extremums in the same water layer is determined, to obtain a first difference between the peak interval and a first end extremum and a second difference between the peak interval and a second end extremum; wherein the first end extremum and the second end extremum are two end extremums in the same water layer and in opposite directions.

[0102] Step S290, extending the first end extreme value outward by a first distance and extending the second end extreme value outward by a second distance to obtain the extreme value boundary of the best matching profile in each water layer, wherein the first distance is equal to the first difference multiplied by a preset ratio, and the second distance is equal to the second difference multiplied by the ratio.

[0103] In step S250, the observation data of the observation elements (such as salinity) is statistically processed to obtain the observation data of each water layer, for example, the salinity value of each layer. In step S260, because the frequency of measurement of each water layer is different, the frequency of the measurement statistical result is normalized here, and the extreme values (maximum and minimum) of the observation data of the best matching profile in each water layer are extracted. In step S270, the observation data is divided into intervals, for example, every 0.1 interval is divided, and assuming that the salinity value between 33.5 and 34 is divided into an interval every 0.1, the interval 33.9-34 with the largest frequency is found as the peak interval in the intervals 33.5-33.6, 33.6-33.7, …, 33.9-34. In step S280, the difference between the peak interval and the two end extreme values in the same water layer is determined to obtain the first difference between the peak interval and the first end extreme value and the second difference between the peak interval and the second end extreme value; wherein the first end extreme value and the second end extreme value are the two end extreme values in the same water layer in opposite directions. For example: the first end extreme value and the second end extreme value are 33.2 and 34.8 respectively, then the first difference between the peak interval 33.9-34 and the first end extreme value 33.2 is 33.9-33.2=0.7, and the second difference between the peak interval 33.9-34 and the second end extreme value 34.8 is 34.8-34=0.8. In step S290, continuing the above example, the first end extreme value 33.2 is extended by 1 / n*0.7, and the second end extreme value 34.8 is extended by 1 / n*0.8, the value of n can be set according to the empirical value, generally, the value of n is selected to be 10, then the extended extreme values are equal to: 33.2-0.07=33.13 and 34.88. Finally, the extreme value boundary is obtained by combining the extreme values of each water layer.

[0104] The step S130 is described below:

[0105] Referring to Figure 4 In some embodiments, the step includes the following steps S310-S350:

[0106] Step S310, data interpolation is performed on the WOA climate data set to obtain a full map data set.

[0107] Step S320, according to the month and the longitude and latitude of the observation data in the profile of the to-be-measured Argo float, the corresponding target climate data is found in the full map data set.

[0108] Step S330, calculate the average value and standard deviation of the observation data corresponding to the target climatological data.

[0109] Step S340, add or subtract several times of the standard deviation to the average value to obtain the climatological extreme value.

[0110] Step S350, generate the climatological extreme value boundary according to the calculated climatological extreme value.

[0111] In general, 3 times of the standard deviation is added in step S340 to obtain the climatological extreme value boundary.

[0112] The following describes step S140:

[0113] Referring to Figure 5 In some embodiments, the steps include steps S410-S430 as follows:

[0114] Step S410, determine each first extreme value constituting the extreme value boundary of the best matching profile, and each second extreme value constituting the climatological extreme value boundary.

[0115] Step S420, find two extreme values located at both ends among the first extreme value and the second extreme value in each water layer.

[0116] Step S430, constitute the final extreme value boundary according to the two extreme values found in each water layer.

[0117] The extreme value boundary of the best matching profile and the climatological extreme value boundary are compared, because for a water layer, the extreme value boundary of the best matching profile has two extreme values, and the climatological extreme value boundary also has two extreme values, and by comparing the two extreme values at both ends, one extreme value is selected at one end.

[0118] The following describes step S150:

[0119] Step S150 is to obtain the probability distribution of the observation data of different water layers.

[0120] According to the final extreme value boundary and the frequency distribution of the observation data in the water layer, the akima interpolation is used to calculate the probability distribution of all intervals (including intervals without observation data and intervals with observation data) in the boundary, so that the constructed probability distribution conforms to the actual situation.

[0121] Because the best matching profile is obtained above, the frequency distribution of the observation data of the best matching profile in the same water layer can be obtained here, and the upper envelope fitting of the frequency distribution can be solved by using the akima interpolation combined with the maximum and minimum boundaries to obtain a function, which is taken as the probability distribution. For example, in some embodiments, the number of intervals crossed by the observation data is calculated according to the frequency distribution of the same water layer, where the interval has been described in the above embodiments, and the step length in the water layer can be calculated according to the number of intervals crossed combined with the related calculation formula, and then the extreme value position in each step length range is calculated by using the step length of each layer, and finally the upper envelope is drawn to obtain a function, which represents the probability distribution.

[0122] With reference to Figure 6 and Figure 7 , the following describes step S160:

[0123] Step S160 is divided into two steps, the first step is to calculate the discrimination boundary, and the second step is to determine whether the observation data in the to-be-tested Argo float profile is a suspicious profile according to the discrimination boundary and the observation data in the to-be-tested Argo float profile.

[0124] The first step is introduced here, and in some embodiments, the implementation process includes steps S510-S540:

[0125] Step S510, according to the probability distribution of each water layer, the observation data is discriminated and valued based on the total probability from high to low according to the selected probability, to obtain a first valuation result.

[0126] Step S520, according to the probability distribution of each water layer, the observation data is discriminated and valued based on the total probability from small to large according to the cumulative probability distribution, and when the total probability sum reaches 50%, the intermediate starting cumulative probability calculation position is taken as the intermediate starting cumulative probability calculation position. The probability size on the left and right of the intermediate starting cumulative probability calculation position is compared, and the cumulative probability from the middle to both sides is obtained according to the comparison result; the observation data is discriminated and valued according to the final cumulative probability, to obtain a second valuation result.

[0127] Step S530, the first valuation result and the second valuation result are each weighted by 50% to obtain a third valuation result.

[0128] Step S540, the observation data with the third valuation result less than the preset first threshold value is taken as the discrimination boundary.

[0129] The second step is introduced here, and in some embodiments, the implementation process includes steps S610-S630:

[0130] Step S610, if the observation data in the to-be-tested Argo float profile is located outside the discrimination boundary, the observation data in the to-be-tested Argo float profile is determined as a suspicious measurement point.

[0131] Step S620, determining the control depth of the suspicious measurement point, wherein the control depth is the depth difference between the suspicious measurement point and the last measurement point plus half of the sum of the depth differences between the suspicious measurement point and the next measurement point.

[0132] Step S630, if the consecutive measurement points are all suspicious measurement points and / or the number of control depths reaches the second threshold value, judging that the to-be-measured Argo buoy profile is a suspicious profile.

[0133] The first distinguishing mode is that the consecutive measurement points are all suspicious measurement points.

[0134] If the observation data measured at a measurement point in the to-be-measured Argo buoy profile is outside the distinguishing boundary, the observation point is a suspicious measurement point. If the consecutive measurement points are all suspicious measurement points, it is proved that the to-be-measured Argo buoy profile is a suspicious profile.

[0135] The second distinguishing mode is that the number of control depths reaches the second threshold value. For a measurement point, the depth distance from the nearest measurement point is closest to the property, and since the specific information of the water body between two measurement points and the upper and lower measurement points depends almost only on the depth distance between them, the control depth of each measurement point is half of the distance from the last measurement point plus half of the distance from the next measurement point. Assuming that a profile has n measurement points, the control depth of the ith measurement point is:

[0136]

[0137] wherein, is the control depth of the ith measurement point, is the depth difference between the 1th measurement point and the 2th measurement point, is the depth difference between the 1th measurement point and the 2th measurement point, is the depth difference between the 1th measurement point and the 2th measurement point, is the depth difference between the 1th measurement point and the 2th measurement point, is the depth difference between the 1th measurement point and the 2th measurement point. is the depth difference between the 1th measurement point and the 2th measurement point, is the depth difference between the 1th measurement point and the 2th measurement point. is the depth difference between the 1th measurement point and the 2th measurement point. is the depth difference between the 1th measurement point and the 2th measurement point. The control depth of the suspicious measurement point is regarded as a suspicious control depth, and when the cumulative suspicious control depth exceeds a certain depth (referred to as the second threshold value in the embodiment), the profile is regarded as a suspicious profile. It is worth noting that the value of the certain depth can be set according to the empirical value, which will be described in detail in the subsequent embodiment.

[0138] The embodiment has the following beneficial effects:

[0139] The embodiment has the following beneficial effects: ​

[0140] In the process of calculating the probability distribution, the best matching profile of the historical trajectory of the Argo buoy to be tested is combined with the WOA climatic data. By using the two data and their corresponding data extraction methods, the historical observation elements (such as salinity) around the moving Argo buoy can be effectively positioned and the corresponding verified data set can be extracted, which can effectively improve the accuracy of real-time quality control of the Argo buoy. Moreover, the akima interpolation method is used to construct the probability distribution in the unmeasured area within the boundary, so that the constructed probability distribution conforms to the actual situation, which can effectively improve the reliability of real-time quality control of the Argo buoy.

[0141] In the above steps S510 to S540, the characteristics of the 3sigma discrimination method itself are used to extract: selecting features of normal distribution, in 3sigma discrimination, according to the characteristics of normal distribution, there are the following main features: 1) selecting probability content value from large to small 2) selecting probability from middle to both sides 3) selecting 3sigma threshold value containing cumulative probability. By using these characteristics of normal distribution, the distribution is extended from normal distribution to all distributions by S510 and S520, and then by means of discrimination number method, the two perspectives are integrated into one perspective by using the average value in step S530, so that the probability distribution of marine elements in most non-normal distribution cases in the ocean can be discriminated by using the existing means, making the result more reliable and objective.

[0142] In the above steps S610 to S630, the discrimination of local features is added on the premise of the traditional discrimination method, and the cross verification of continuous suspicious point discrimination and suspicious point control depth discrimination is used to convert the judgment result from suspicious point to suspicious profile, so as to achieve the purpose of quality control.

[0143] In order to solve the above technical defects, with reference to Figure 8 to 16 An embodiment of the present application provides an Argo buoy real-time quality control method, which comprises the following steps:

[0144] Step S910, obtaining the measurement items of a single Argo buoy to be tested, the measurement items containing the profile observation positions of the buoy to be tested, and using the historical "best matching" profile method to extract the data set of the best matching profile suitable for full depth and containing continuity characteristics corresponding to the to-be-tested data in the to-be-tested Argo buoy profile from the reliable historical observation data set of the to-be-tested Argo buoy (containing a plurality of corresponding best matching profiles).

[0145] The specific process of the data set of the best matching profile is as follows:

[0146] Step S911, since the observation depth of the historical observation data set is not uniform, the data set of the best matching profile is interpolated to each depth by using the akima interpolation method.

[0147] Step S912, each water layer in the data set of the best matching profile is fused with the observation data of the adjacent water layer, so as to contain the data distribution characteristics of the upper and lower water layers, that is, to contain the continuity characteristics.

[0148] Referring to Figure 8 to Figure 10 , Figure 8 is the profile position of the Argo float of the Argo float project_29010 and the best matching profile position. Figure 9 is Figure 8 the observation profile cluster of the Argo float in the Argo float project_29010 converted from the profile position of the Argo float. Figure 10 is Figure 8 the observation profile cluster of the best matching profile converted from the best matching profile position in the Argo float project_29010, wherein Figure 10 (b) is Figure 10 the statistical result at the position of -200dbar (unit of pressure) in (a), Figure 10 (c) is Figure 10 the statistical result at -800dbar in (a), there are 2 to 3 observation peaks at -200dbar, there are two observation peaks at -800dbar, the float may cross two water masses, and the statistical result can better reflect the element distribution at this depth.

[0149] Step S920, in the data set of the best matching profile obtained in step S910, the statistical result of the observation data of the best matching profile in each water layer is obtained, and due to the measurement frequency of each water layer, frequency normalization of the statistical result is required, that is, the best matching profile corresponding to the depth is statistically processed according to temperature 0.1℃ and salinity 0.05, the frequency distribution graph corresponding to the accuracy is obtained, and the total frequency is the same, so that the frequency in the corresponding data interval will be enlarged at different scales, and the magnification is .

[0150] Step S930, calculate the final extreme boundary.

[0151] The definition of the extreme boundary is that the interval outside the extreme boundary is an interval with a probability of 0 of appearing observation values. In this step, the climatic data and the data set of the best matching profile obtained in step S910 are used to calculate the final extreme boundary of the data set of the best matching profile.

[0152] Step S931, collect the WOA climatological dataset, and perform data interpolation on the WOA climatological dataset to obtain a full map dataset; according to the month and latitude and longitude of the to-be-tested data, locate the corresponding target climatological data from the full map dataset, calculate the average value and standard deviation of the observation data corresponding to the month of the target climatological data, and assume that the original distribution is a normal distribution, and consider that the climatological maximum boundary of the climatological data is reached when the three standard deviations are reached, and the average value of the month plus three times the standard deviation is taken as the climatological maximum boundary of the target climatological data.

[0153] Step S932, after the frequency normalization of the measurement statistical results of the observation data of each water layer of the best matching profile, according to the maximum value of the observation data of each water layer, extract the peak interval with the highest statistical frequency of each water layer, calculate the difference between the median value of the peak interval and the two end maximum values, obtain the difference result, and extend the peak interval outward by 10% according to the difference result to obtain the maximum boundary of the best matching profile.

[0154] Step S933, according to the climatological maximum boundary obtained in step S931 and the maximum boundary of the best matching profile obtained in step S932, obtain the final maximum boundary.

[0155] Reference Figure 11 , the maximum value difference of the best matching profile is generated based on the maximum value of the single water layer of the Argo buoy project_29010. The peak interval is the column marked with a dashed line in the figure, the difference between the peak interval and the left maximum value is calculated, and the difference between the peak interval and the right maximum value is calculated, and then the difference between the peak interval and the left maximum value is pushed to the left by 10%, and the difference between the peak interval and the right maximum value is pushed to the right by 10%. The difference between the two maximum values before the push is 90.91% after the push.

[0156] Reference Figure 12 , the final maximum boundary obtained by Argo buoy project_2900433( Figure 12 (a) and Argo buoy project_project_29054( Figure 12 (b) are shown in the figure, wherein the green line is the median line of each water layer of the peak interval corresponding to Figure 11 , the blue line represents the maximum boundary presented based on the best matching profile, the cyan line represents the maximum boundary of the average value plus or minus three times the standard deviation of the entire observation data according to the corresponding month of the climatological data, and the red line represents the selected final maximum position.

[0157] Step S940, according to the frequency distribution diagram obtained in step S920 and the final maximum boundary obtained in step S930, use the data span interval number , according to the formula:

[0158]

[0159] wherein is the number of data span intervals of the element value of each water layer, is the step length for finding the extreme value;

[0160] The corresponding step length of each layer is calculated, and then the extreme value position in each step length range is calculated using the step length of each layer. For the extreme value and the corresponding final extreme boundary, the akima interpolation is used to obtain the probability distribution function corresponding to the layer.

[0161] Referring to Figure 13 , the final extreme boundary and the frequency cumulative value are given. According to the span of the observation data of each water layer, different step lengths are selected, and the maximum value in the corresponding interval is selected as the interpolation data point (it is considered that the larger the value, the closer it is to the actual sea area distribution in the sea area) according to the step length selection. The akima interpolation calculation result is taken as the probability distribution of the water layer. Figure 13 (a) is the probability distribution diagram of Argo buoy project_29077 at-100dbar, Figure 13 (b) is the probability distribution diagram of Argo buoy project_290406 at-400dbar.

[0162] Step S950, extend the 3sigma (σ) discrimination of normal distribution, and use the discrimination number assignment method to judge the threshold extraction processing of the obtained data result. The specific steps are as follows:

[0163] Step S951, probability-oriented assignment. Select each layer probability distribution from high to low based on probability size. The discrimination number assignment result is related to the sum of the total probability of selection. The total probability sum and the corresponding assignment are shown in Table 1.

[0164]

[0165] Table 1

[0166] Step S952, 50% probability extrapolation assignment. Based on the probability distribution accumulation from small to large concentration, when the probability sum reaches 50%, it is taken as the intermediate starting cumulative probability calculation position. Compare the probability size on the left and right of the starting calculation position. According to the cumulative probability of the comparison result, the cumulative probability from the middle to both sides is finally obtained. The discrimination number assignment result is related to the sum of the total probability of selection. The total probability sum and the corresponding assignment are shown in Table 2.

[0167]

[0168] Table 2

[0169] Step S953: Average the results of S5.1 and S5.2 (50%-50% weighting) to obtain the final judgment assignment result, and use the position where the average assignment result is less than 1.5 as the judgment boundary. The results of the averaged judgment number assignment of the two methods are shown in Table 3 below:

[0170]

[0171] Table 3

[0172] Reference Figure 14 , gives the results of different discriminant number assignment methods for Argo float project_29044. The distribution of probability-oriented discriminant number assignment results ( Figure 14 (a) Considering the high probability of normal distribution, probability-oriented assignment is adopted, and each layer of probability distribution is selected from high to low based on the probability size. The discriminant number assignment result is related to the sum of the total probabilities of the selections; the distribution of the discriminant number assignment result based on the 50% probability quantile extrapolation ( Figure 14 (b)), taking into account the concentration of the normal distribution, based on the accumulation of probability distribution from small to large concentration, when the total probability reaches 50%, it is used as the middle starting cumulative probability calculation position, and the probabilities on the left and right of the starting calculation position are compared. According to the cumulative probabilities of the comparison results, the cumulative probabilities from the middle to both sides are finally obtained. The discriminant number assignment result is related to the sum of the total probabilities of the selection; the difference in the assignment results of the two methods ( Figure 14 (c)), the two methods have different focuses, resulting in different assignment results. The probability-oriented method focuses on high-probability selection, and its assigned large value area is affected by the probability distribution. The 50% quantile extrapolation method has a higher discriminant value assignment in the data distribution close to the center area. The two methods can complement each other; the two methods average the discriminant value assignment results ( Figure 14 (d) Based on the comparison of the two assignment methods, the final judgment assignment result was averaged (50%-50% weighting). The position where the average assignment result was less than 1.5 was used as the outer boundary of the final judgment threshold. This method simultaneously considers the characteristics of the 3sigma discrimination from the center to the sides and from high probability to low probability in the normal distribution, and from the center to the ends, successfully extending the 3sigma discrimination to non-normal probability distributions.

[0173] Step S960: perform boundary determination on the observation data in the Argo float profile to be measured in step S910, and identify points on the boundary as suspicious measurement points.

[0174] Step S970, the suspicious measurement points outside the threshold are analyzed by using the continuous point discrimination and the control depth discrimination, when any one of the two discrimination methods is not satisfied, the profile where the suspicious measurement point is located is regarded as a suspicious profile, and the following is the specific implementation steps:

[0175] Step S971, continuous point discrimination: when two or more continuous suspicious measurement points appear in the same profile, the profile is regarded as a suspicious profile.

[0176] Step S972, control depth discrimination: the control depth of the suspicious measurement point is regarded as a suspicious control depth, when the cumulative suspicious control depth exceeds 200m, that is, more than 10% of the usual Argo measurement depth 2000m, that is, the profile is regarded as a suspicious profile.

[0177] Referring to Figure 15 , the continuous point discrimination method discrimination result schematic diagram is given, and four figures show the results of different depths of different profiles of different observation buoys after continuous suspicious points are defined as suspicious profiles. According to the judgment result, it can be seen that the continuous point discrimination method can effectively label and process the suspicious profile of the suspicious buoy.

[0178] Referring to Figure 16 , the control depth discrimination method discrimination result (left figure in Figure 16 ) and effect schematic diagram (right figure in Figure 16 ) for salt drift phenomenon are given. It can be seen that the control depth discrimination method can effectively identify most of the profiles with salt drift problems and label them, which has good accuracy.

[0179] The embodiment has the beneficial effects:

[0180] In the process of calculating the probability distribution, the best matching profile of the historical trajectory of the to-be-measured Argo buoy and the WOA climatic data are combined, and the two data and the corresponding data extraction method are used to effectively locate the surrounding historical observation elements (such as salinity) of the moving Argo buoy and extract the corresponding verification data set, which can effectively improve the accuracy of real-time quality control of the Argo buoy. Moreover, the akima interpolation method is used to construct the probability distribution in the boundary without measurement area, so that the constructed probability distribution conforms to the actual situation, which can effectively improve the reliability of real-time quality control of the Argo buoy.

[0181] The characteristics of the 3sigma discrimination method are used for feature extraction: selecting features for normal distribution, and in 3sigma discrimination, the following main features are selected according to the characteristics of normal distribution: 1) selecting probability content values from large to small 2) selecting probability from the middle to both sides 3) selecting 3sigma threshold value containing cumulative probability of about 99.73%. By using these characteristics of normal distribution, the distribution is extended from normal distribution to all distributions by using S510 and S520, and then by using the discrimination number method, the two perspectives are integrated into one perspective by using the average value in step S530, so that the probability distribution of the marine elements in the vast majority of non-normal distribution cases in the sea can be discriminated by using the existing means, and the result is more reliable and objective.

[0182] The discrimination of local features is added to the traditional discrimination method, and the cross verification of the continuous point discrimination and the control depth discrimination is used to convert the judgment result from the suspicious measurement point to the suspicious profile, so as to achieve the purpose of quality control.

[0183] An embodiment of the present application provides an Argo buoy real-time quality control system, the Argo buoy real-time quality control system comprising:

[0184] A data acquisition unit is configured to acquire a WOA climatic data set, a to-be-tested Argo buoy profile, and a historical observation data set of the to-be-tested Argo buoy.

[0185] A first extreme boundary calculation unit is configured to find a corresponding best matching profile from the historical observation data set according to observation data in the to-be-tested Argo buoy profile, and determine an extreme boundary of the best matching profile, wherein the extreme boundary is a boundary for representing whether the observation data in the best matching profile is 0.

[0186] A second extreme boundary calculation unit is configured to find corresponding target climatic data from the WOA climatic data set according to the observation data in the to-be-tested Argo buoy profile, and determine a climatic extreme boundary according to the target climatic data.

[0187] A third extreme boundary calculation unit is configured to compare the extreme boundary of the best matching profile and the climatic extreme boundary, and determine a final extreme boundary according to a comparison result.

[0188] A probability distribution calculation unit is configured to calculate a probability distribution of the observation data in the best matching profile by using akima interpolation according to the final extreme boundary and a frequency distribution of the observation data in the best matching profile.

[0189] A suspicious profile detection unit is configured to calculate a discrimination boundary according to the probability distribution, and determine whether the to-be-tested Argo buoy profile is a suspicious profile according to the observation data in the to-be-tested Argo buoy profile and the discrimination boundary.

[0190] It should be noted that the Argo buoy real-time quality control system provided by the embodiment is based on the same inventive concept as the Argo buoy real-time quality control method embodiments described above, and therefore the related content of the Argo buoy real-time quality control method embodiments described above also applies to the Argo buoy real-time quality control system embodiment, which will not be described in detail here.

[0191] As Figure 18 The embodiment of the present application also provides an electronic device, and the electronic device comprises:

[0192] at least one memory;

[0193] at least one processor;

[0194] at least one program;

[0195] The program is stored in the memory, and the processor executes the at least one program to implement the Argo buoy real-time quality control method described above.

[0196] The electronic device can be any intelligent terminal including a mobile phone, a tablet computer, a personal digital assistant (PDA), a vehicle-mounted computer, etc.

[0197] The electronic device of the embodiment of the present application will be described in detail below.

[0198] The processor 1600 can be implemented in the form of a general central processing unit (CPU), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the embodiments of the present application.

[0199] The memory 1700 can be implemented in the form of a read only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM), etc. The memory 1700 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are saved in the memory 1700 and are called and executed by the processor 1600 to implement the Argo buoy real-time quality control method of the embodiments of the present application.

[0200] The input / output interface 1800 is used to realize information input and output.

[0201] The communication interface 1900 is configured to realize the communication interaction between the device and other devices, and can realize the communication through a wired manner (for example, a USB, a network cable, etc.) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, etc.).

[0202] The bus 2000 is configured to transmit information between various components (for example, the processor 1600, the memory 1700, the input / output interface 1800 and the communication interface 1900) of the device.

[0203] The processor 1600, the memory 1700, the input / output interface 1800 and the communication interface 1900 are connected to each other through the bus 2000 to realize the communication connection between the device.

[0204] The embodiment of the present application further provides a storage medium, which is a computer readable storage medium, and stores computer executable instructions for causing a computer to execute the Argo buoy real-time quality control method.

[0205] The memory is a non-transitory computer readable storage medium, and can be used to store a non-transitory software program and a non-transitory computer executable program. In addition, the memory can include a high-speed random access memory, and can further include a non-transitory memory, for example, at least one magnetic disk storage device, a flash memory device or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and the remote memory can be connected to the processor through a network. Examples of the network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0206] The embodiments described in the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0207] Those skilled in the art can understand that the technical solutions shown in the drawings do not constitute a limitation on the embodiments of the present application, and can include more or fewer steps than the drawings, or combine certain steps or different steps.

[0208] The device embodiments described above are only schematic, and the units described as separate components can be or can not be physically separated, that is, can be located in one place or can be distributed on multiple network units. According to actual needs, part or all of the modules can be selected to realize the purposes of the embodiments of the present application.

[0209] Those skilled in the art can understand that all or some of the steps in the method disclosed above, the functional modules / units in the system and the device can be implemented by software, firmware, hardware or a proper combination thereof.

[0210] The terms "first", "second", "third", "fourth" and the like in the description of the application and in the claims hereof, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of the terms so termed is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of orderly or chronological permutations. Moreover, the terms "comprise", "have" and any variations thereof are intended to cover a non-exclusive inclusion, for example, a process, method, article, or apparatus that comprises a list of steps or units can not necessarily be limited to those steps or units, but can include additional steps or units not expressly listed or inherent to such process, method, article, or apparatus.

[0211] It should be understood that, in the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the relationship between associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that there are three cases: only A, only B, and A and B at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0212] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed objects can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0213] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0214] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0215] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a plurality of instructions for causing an electronic device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and various program storage media.

[0216] The above is a specific description of the preferred implementation of the embodiments of the present application, but the embodiments of the present application are not limited to the above implementation. Those skilled in the art can make various equivalent modifications or replacements without departing from the spirit of the embodiments of the present application, and these equivalent modifications or replacements are all included in the scope defined by the claims of the embodiments of the present application.

Claims

1. An Argo buoy real-time quality control method, characterized in that, The Argo buoy real-time quality control method comprises: obtaining a WOA climatological dataset, a profile of a to-be-tested Argo buoy and a historical observation dataset of the to-be-tested Argo buoy; finding a corresponding best matching profile from the historical observation dataset according to observation data in the profile of the to-be-tested Argo buoy, and determining extreme value boundaries of the best matching profile, wherein the extreme value boundaries are boundaries for characterizing whether the observation data in the best matching profile is 0; finding corresponding target climatological data from the WOA climatological dataset according to the observation data in the profile of the to-be-tested Argo buoy, and determining climatological extreme value boundaries according to the target climatological data; comparing the extreme value boundaries of the best matching profile and the climatological extreme value boundaries, and determining final extreme value boundaries according to a comparison result; calculating a probability distribution of the observation data in the best matching profile by using akima interpolation according to the final extreme value boundaries and a frequency number distribution of the observation data in the best matching profile; calculating a discrimination boundary according to the probability distribution, and judging whether the profile of the to-be-tested Argo buoy is a suspicious profile according to the observation data in the profile of the to-be-tested Argo buoy and the discrimination boundary.

2. The Argo buoy real-time quality control method of claim 1, wherein, After finding the corresponding best matching profile from the historical observation dataset, the Argo buoy real-time quality control method further comprises: interpolating the observation data in the best matching profile to each depth by using akima interpolation; fusing the observation data of the best matching profile in adjacent two water layers.

3. The Argo buoy real-time quality control method according to claim 2, wherein, The determination of the extreme value boundaries of the best matching profile comprises: obtaining a measurement statistical result of the observation data in each water layer of the best matching profile; performing frequency normalization on the measurement statistical result, and extracting extreme values of the observation data in each water layer of the best matching profile; determining a peak interval with the highest statistical frequency of the observation data in each water layer according to the extracted extreme values of the observation data in each water layer; determining a first difference value between the peak interval and a first end extreme value and a second difference value between the peak interval and a second end extreme value by determining differences between the peak interval and two end extreme values in the same water layer, wherein the first end extreme value and the second end extreme value are two end extreme values in the same water layer and in opposite directions; extending the first end extreme value outward by a first distance and extending the second end extreme value outward by a second distance to obtain the extreme value boundaries of the best matching profile in each water layer, wherein the first distance is equal to the first difference value multiplied by a preset ratio, and the second distance is equal to the second difference value multiplied by the ratio.

4. The Argo buoy real-time quality control method according to claim 2, wherein, The finding of the corresponding target climatological data from the WOA climatological dataset according to the observation data in the profile of the to-be-tested Argo buoy, and the determination of the climatological extreme value boundaries according to the target climatological data comprise: performing data interpolation on the WOA climatological dataset to obtain a full map dataset; finding corresponding target climatological data from the full map dataset according to the month, the latitude and the longitude of the observation data in the profile of the to-be-tested Argo buoy; calculating a mean value and a standard deviation of observation data corresponding to the target climatological data; adding and subtracting several times the standard deviation to the mean value to obtain climatological maximum and minimum values; generating a climatological maximum and minimum value boundary according to the calculated climatological maximum and minimum values.

5. The Argo buoy real-time quality control method according to any one of claims 1 to 4, characterized in that, The comparison of the maximum and minimum value boundary of the best matching profile and the climatological maximum and minimum value boundary includes: determining each first maximum and minimum value constituting the maximum and minimum value boundary of the best matching profile, and each second maximum and minimum value constituting the climatological maximum and minimum value boundary finding two maximum and minimum values located at both ends of the first maximum and minimum values and the second maximum and minimum values in each water layer; composing a final maximum and minimum value boundary according to the two maximum and minimum values found in each water layer.

6. The Argo buoy real-time quality control method according to claim 5, wherein, The calculation of the discrimination boundary of the to-be-tested Argo float profile according to the probability distribution includes: selecting from high to low according to the probability of each water layer, assigning a discrimination number to the observation data according to the total probability of the selected probability, to obtain a first assignment result; comparing the probability size of the left and right of the intermediate starting accumulation probability calculation position according to the probability distribution of each water layer, and obtaining the accumulation probability from the middle to both sides according to the comparison result; assigning a discrimination number to the observation data according to the final accumulation probability, to obtain a second assignment result; averaging the first assignment result and the second assignment result with a weight of 50%, to obtain a third assignment result; the observation data less than a preset first threshold value is taken as the discrimination boundary.

7. The Argo buoy real-time quality control method according to claim 6, wherein, The determination of whether the to-be-tested Argo float profile is a suspicious profile according to the observation data in the to-be-tested Argo float profile and the discrimination boundary includes: if the observation data in the to-be-tested Argo float profile is outside the discrimination boundary, determining that the observation data in the to-be-tested Argo float profile is a suspicious measurement point; determining the control depth of the suspicious measurement point, wherein the control depth is the depth difference between the suspicious measurement point and the previous measurement point plus half the sum of the depth difference between the suspicious measurement point and the next measurement point; if a plurality of consecutive measurement points are the suspicious measurement points and / or the number of control depths reaches a second threshold value, determining that the to-be-tested Argo float profile is a suspicious profile.

8. An Argo buoy real-time quality control system characterized by, The Argo float real-time quality control system includes: a data acquisition unit for acquiring a WOA climatological data set, a to-be-tested Argo float profile, and a historical observation data set of a to-be-tested Argo float; a first maximum and minimum value boundary calculation unit for finding a corresponding best matching profile from the historical observation data set according to the observation data in the to-be-tested Argo float profile, and determining a maximum and minimum value boundary of the best matching profile, wherein the maximum and minimum value boundary is a boundary for representing whether the possibility of the observation data in the best matching profile is 0; a second extreme boundary calculation unit, configured to find corresponding target climatological data from the WOA climatological data set according to the observation data in the to-be-tested Argo float profile, and determine a climatological extreme boundary according to the target climatological data; a third extreme boundary calculation unit, configured to compare the extreme boundary of the best matching profile and the climatological extreme boundary, and determine a final extreme boundary according to a comparison result; a probability distribution calculation unit, configured to calculate a probability distribution of the observation data in the best matching profile according to the final extreme boundary and a frequency number distribution of the observation data in the best matching profile, and obtain the probability distribution by using akima interpolation; a suspicious profile detection unit, configured to determine whether the to-be-tested Argo float profile is a suspicious profile according to the observation data in the to-be-tested Argo float profile and a judgment boundary calculated according to the probability distribution.

9. An electronic device, comprising: comprise: at least one control processor and a memory connected in communication with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to perform the Argo float real-time quality control method of any one of claims 1 to 7.

10. A computer readable storage medium characterized by the computer readable storage medium stores computer executable instructions for causing a computer to perform the Argo float real-time quality control method of any one of claims 1 to 7.

Images