CN115422746B - Internal solitary wave parameter extraction algorithm based on underwater glider - Google Patents

Internal solitary wave parameter extraction algorithm based on underwater glider Download PDF

Info

Publication number
CN115422746B
CN115422746B CN202211064070.4A CN202211064070A CN115422746B CN 115422746 B CN115422746 B CN 115422746B CN 202211064070 A CN202211064070 A CN 202211064070A CN 115422746 B CN115422746 B CN 115422746B
Authority
CN
China
Prior art keywords
data
depth
solitary wave
density
internal solitary
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202211064070.4A
Other languages
Chinese (zh)
Other versions
CN115422746A (en
Inventor
王智峰
崔俊男
夏浩峰
过武宏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ocean University of China
PLA Navy Submarine College
Original Assignee
Ocean University of China
PLA Navy Submarine College
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ocean University of China, PLA Navy Submarine College filed Critical Ocean University of China
Priority to CN202211064070.4A priority Critical patent/CN115422746B/en
Publication of CN115422746A publication Critical patent/CN115422746A/en
Application granted granted Critical
Publication of CN115422746B publication Critical patent/CN115422746B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)

Abstract

An internal solitary wave parameter extraction algorithm based on an underwater glider belongs to the technical field of marine internal solitary wave observation, and adopts a glider matrix to observe, and comprises the following steps: step 1, extracting space-time data observed by an underwater glider, and analyzing the space-time data by adopting a hierarchical clustering analysis method; step 2, carrying out equal-depth interval interpolation on the data in the simultaneous empty data set, calculating Euclidean distance, and judging; step 3, extracting suspicious sequences with equal-close surfaces with maximum depth difference larger than 25m and depth where the maximum depth difference is larger than 20m, and carrying out data and calculation on the machine number, the section number and the average density of the suspicious sequences to obtain ocean layering conditions, floating frequency and internal solitary wave amplitude; and 4, selecting a proper internal solitary wave theory to invert the influence of the internal solitary wave on the equal-density surface according to the optimal application range of the internal solitary wave theory, and outputting a density profile containing the information of the internal solitary wave. The invention can obviously improve the accuracy of the inversion result of the internal solitary wave.

Description

Internal solitary wave parameter extraction algorithm based on underwater glider
Technical Field
The invention belongs to the technical field of ocean solitary wave observation, and particularly relates to an internal solitary wave parameter extraction algorithm based on an underwater glider.
Background
The traditional method for observing the internal solitary wave mainly comprises the steps of submerged buoy laying, shipborne instrument observation, argo buoy and the like, precious first hand data can be obtained through field observation, sea phenomena can be reflected most truly, and the defects are obvious: such as huge instrument input cost, high labor and time costs for sailing observation, difficulty in realizing large-area observation, and the like. Meanwhile, the method can also be used for observing based on satellite remote sensing, and the method can accurately obtain the information such as the space position, the propagation direction and the like of the internal solitary wave, but cannot obtain the amplitude of the internal solitary wave and the influence of the amplitude of the internal solitary wave on the sea temperature salt and the density jump layer.
As unmanned observation platforms mature gradually, underwater gliders show significant advantages, particularly in: the instrument is simple to collect and release, the data time is continuous, the position area is flexible, and the labor cost is reduced. Meanwhile, the use efficiency of the glider in observing data can be effectively improved by utilizing the research of the underwater glider on the internal wave characteristic parameters and the three-dimensional warm salt. The glider is utilized to carry out ocean observation, the internal wave characteristics are extracted, and the sea area temperature salt fine structure is analyzed and researched, so that the method is beneficial to calculating an underwater sound field, and the implementation of underwater engineering or military operation is ensured.
Since the motion of the underwater glider in the sea is mainly driven by gravity and buoyancy, it is easily affected by the change of the flow field generated by the ocean phenomenon. By utilizing this characteristic of the glider, some students reverse the internal wave characteristic parameters encountered by the glider through the motion condition of the underwater glider. Seven glider observation experiments were performed on the igneous strait by Rudnick et al (2013) from april 2007 to july 2008, and the occurrence of high-frequency internal waves was observed with a period smaller than the acquisition period (3-6 hours) of the glider for the sea water profile information. They estimate the vertical flow rate caused by the internal wave by the pressure measured by the glider and the posture of the glider, and propose an algorithm for calculating the vertical flow rate by the motion equation of the glider and estimating the vertical flow rate by the high-pass filter. During this time, they observed a maximum vertical flow rate exceeding 0.2m/s, and a maximum vertical amplitude of thermocline approaching 200m. Palm et al (2015) have conducted research into marine microstructure in keltersea in 2012 in summer using underwater gliders. They found that the motion inside sea water is mainly dependent on widely distributed ocean internal waves, and that the maximum internal wave amplitude observed exceeds 50m, corresponding to 1/3 of the total water depth. The density jump mixing caused by the internal wave is the main reason for the density reduction of the bottom layer of the seawater, which means that the movement of the density jump is an important mechanism for regulating the seawater exchange in continental shelf areas and deep sea areas. The occurrence of internal solitary waves was observed in south China sea using underwater gliders in August 2017, ma et al (2018). In combination with the mesoscale resolution satellite ocean images, it was demonstrated that the motion of the south sea thermocline is largely dependent on the effects of internal waves. They estimated the period and amplitude of the isolated wave in addition to the inter-range by the vertical velocity and motion characteristics of the glider, 24 minutes and 127 meters respectively.
However, the current research is in the starting stage, the theory and the method in various aspects are still immature, and a great amount of improvement room still exists. In the aspect of observing the internal solitary wave by utilizing the underwater glider, the theoretical basis of the internal solitary wave is lacking, the internal solitary wave parameters are difficult to accurately invert, and the traditional method adopts a single glider to observe, so that the influence of random factors on the motion of the glider is difficult to overcome, thereby affecting the observation precision and being incapable of accurately determining the information such as the amplitude of the internal solitary wave.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention discloses an internal solitary wave parameter extraction algorithm based on an underwater glider, which aims to solve the problems of poor maneuverability, high observation cost and low data precision of the traditional observation means by performing data analysis by combining a space-time analysis method of ocean data, an ocean vertical speed model and an internal solitary wave theoretical model and extracting internal solitary wave characteristic parameters and reversing the influence of the internal solitary wave on ocean temperature and density through the underwater glider matrix observation.
In order to achieve the above purpose, the technical scheme of the invention is as follows:
an internal solitary wave parameter extraction algorithm based on an underwater glider, wherein the extraction algorithm adopts a glider matrix for observation, and the glider matrix refers to: a plurality of underwater gliders are put in at different places in a space matrix form at the same time, and the underwater gliders are used for simultaneous observation, comprising the following steps:
step 1, extracting space-time data observed by an underwater glider, wherein the space-time data comprises a glider machine number, an observation section number, a measuring point longitude and latitude, an observation time, a temperature, a salinity, a density and a water depth, and analyzing the space-time data by adopting a hierarchical clustering analysis method to obtain a simultaneous space-time data set;
step 2, carrying out equal-depth interval interpolation on the data in the simultaneous space data set to obtain equal-depth average data, then comparing the single-section data with the equal-depth average data, calculating Euclidean distance, and judging; the single profile data refers to: when each group of data obtained by hierarchical clustering analysis has a plurality of section data of different gliders at different times, single section information in the same group is single section data;
step 3, extracting suspicious sequences with equal-close surfaces with maximum depth difference larger than 25m and depth where the maximum depth difference is larger than 20m, and carrying out data and calculation on the machine number, the section number and the average density of the suspicious sequences to obtain ocean layering conditions, floating frequency and internal solitary wave amplitude;
and 4, selecting a proper internal solitary wave theory to invert the influence of the internal solitary wave on the isopipe surface according to the ocean layering condition and the internal solitary wave amplitude and the optimal application range of the internal solitary wave theory, combining the internal solitary wave theory with the original profile, and finally outputting a density profile containing the internal solitary wave information.
Preferably, in the step 1, a bottom-up (agglomerate) hierarchical clustering method is adopted, when calculating the distance, firstly, the euclidean distance (Euclidean distance) between samples is calculated, and after the cluster is formed, the distance between the classes is calculated; the class-to-class distance calculation method is a shortest distance method (single link) or a longest distance method (complex link), or a class average distance method (average link) or a center distance method (centroid link).
Preferably, in the step 1, all sections observed by all gliders in the same task are extracted first, and machine numbers n corresponding to each observed section are extracted respectively M Cross section number n P Longitude and latitude lon and lat of water inlet point and average time parameter of profile
Figure BDA0003827081940000031
And carrying out dimensionless treatment on the time, as shown in formula (1), so that the time meets the space-time clustering analysis order-of-magnitude requirement:
Figure BDA0003827081940000032
in the formula (1), the components are as follows,
Figure BDA0003827081940000033
is the average value of single profile observation time, and is expressed in seconds, r t Is the cluster analysis spatial threshold c S With time threshold c T Ratio of (r), i.e. r t ＝c S /c T ；
Then, a three-dimensional matrix of spatio-temporal data is constructed, having:
Figure BDA0003827081940000034
suppose that a three-dimensional matrix is divided into m disjoint classes C according to a clustering criterion 1 ，C 2 ，...，C m Two of them are C α And C β The distance of (2) is expressed as:
Figure BDA0003827081940000035
/>
in the formula (3), delta is a distance operator, delta a，b Representing the Euclidean distance from sample a to sample b;
Figure BDA0003827081940000036
representation class C α Longitude and latitude and average time parameters of the ith sample;
in the calculation, the shortest distance method is expressed as:
Figure BDA0003827081940000037
the longest distance method is expressed as:
Figure BDA0003827081940000038
the average distance-like method is expressed as:
Figure BDA0003827081940000039
the center distance method is expressed as:
Figure BDA00038270819400000310
in the formula (7), the amino acid sequence of the compound,
Figure BDA00038270819400000311
and->
Figure BDA00038270819400000312
Respectively refer to class C α And C β Center of (C) α And C β Average of samples in (a).
Preferably, in the step 1, the shortest distance method is selected to analyze the time data, and when the shortest distance method is used for analyzing, the original time-space data is divided into M groups of data according to the standard that the longitude and latitude and the average time parameter distance are smaller than respective threshold values, wherein each group contains M profile information; when the same group of spatiotemporal data is the spatiotemporal consistent data, the group of glider-in-flight machine spatiotemporal data has similar observation results; after grouping the data, the group number N of the grouping is output G Machine number N M Cross section number N P : as shown in formula (8), formula (8) is the output data of step one and the input data of step two:
Data 1 ＝[N G ，N M ，N P ] (8)。
preferably, in the step 2, according to the obtained group number, machine number and profile number, the operation time t, depth dp, temperature tp, salinity st and density ds of all the machine numbers and profile numbers in the same group are extracted for analysis; extracting the maximum submerging depth and the minimum floating depth in different profile observations from the same set of data as interpolation depth intervals of the set of data; the following formula is shown:
Figure BDA0003827081940000041
in the formula (9):
Figure BDA0003827081940000042
a depth interval representing the interpolation of the set of data; dp (dp) down Represents the submerging depth sequence, dp up Representing a floating depth sequence;
then, carrying out equal-depth interpolation on different section data in the same group of data to obtain interpolation results of temperature, salinity and density of each section in the same depth interval, wherein the interpolation results are shown in the following formula:
Figure BDA0003827081940000043
and then, the temperature and salt density results of the sections with the same depth and different sections are averaged to obtain an average sequence, wherein the average sequence is shown in the following formula:
Figure BDA0003827081940000044
in the formula (11), j represents the cross-section number in the same group, and i represents the data number in the same cross-section.
Preferably, in the step 3, the euclidean distance between each profile data and the average data is compared in the same depth interval, and the profile with the largest euclidean distance is selected as the suspicious sequence to perform the next analysis, where the following formula is shown:
Figure BDA0003827081940000045
in the formula (12): d (D) Eucl (A, B) representing Euclidean distance calculation of the sequences A and B, wherein the specific calculation formula is shown on the right side of the equation;
Figure BDA0003827081940000051
the density sequence of the j-th section is calculated by a Newton interpolation method according to the measured density sequence in the formula (10),
Figure BDA0003827081940000052
representing the average density sequence of the group;
the suspicious section is analyzed to find the maximum value diff of the difference diff between the equal close face and the average equal close face max Its corresponding depth dp md Finally, only extracting the profile data with the equal-density surface difference value larger than 25m and the maximum depth difference value larger than 20m to perform the next analysis, and outputting the machine number N of the suspicious sequence M Cross section number N P Equal-surface-to-average equal-surface maximum difference diff max Maximum difference depth dp md Floating frequency N, average depth interval
Figure BDA0003827081940000053
Equal depth average Density->
Figure BDA0003827081940000054
Upper and lower layer thickness h 1 、h 2 Density ρ of upper and lower layers 1 、ρ 2 ；
Based on the output data of the suspicious sequence, the suspicious sequence is subjected to the next step of analysis: firstly, local water depth data are found according to the space coordinates of suspicious sections, and then the buoyancy frequency is calculated according to the measured density data
Figure BDA0003827081940000057
Frequency), as shown in the following equation:
Figure BDA0003827081940000055
in the formula (13): g represents gravitational acceleration, ρ represents sea water density;
finding the depth with the largest buoyancy frequency as the depth d of the density jump layer pyc The maximum diff of the density difference appears in the previous analysis max Corresponding depth d md Substituting formula (14), and estimating the maximum internal solitary wave amplitude at the density jump layer through linear interpolation or other interpolation modes:
Figure BDA0003827081940000056
in formula (14): a represents amplitude, diff max Represents the maximum value of the density difference, d md Represents the depth of occurrence of the maximum value of the density difference, d tot Represents the total water depth d pyc Indicating the depth of the dense jump.
Preferably, in the step 4, a suitable internal solitary wave theory is selected to calculate the characteristic parameters of the internal solitary wave according to the thickness of the upper and lower layers and the amplitude of the internal solitary wave (Cui et al, 2021), a suitable internal solitary wave theory is selected to reconstruct the temperature salt profile at the moment of occurrence of the internal solitary wave, wherein the change form of the amplitude along the depth is calculated in a linear fitting mode or in an exponential fitting mode according to the actual situation, so as to form an equal-density change Δp caused by the internal solitary wave, and the equal-density change Δp is compared with the initial density profile P 0 And combining to obtain density profile data which is finally output and contains the influence of the internal solitary wave on the isopipe, wherein the density profile data is shown in the following formula:
P＝P 0 +ΔP (15)
in formula (15): p represents the density profile of the output, P 0 Representing the raw density profile, Δp represents the calculated intra-solitary wave-induced isopycnic face change according to the algorithm.
The internal solitary wave parameter extraction algorithm based on the underwater glider has the beneficial effects that:
1. the algorithm for extracting the internal solitary wave characteristic parameters through the glider matrix observation data greatly improves the maneuverability of the internal solitary wave observation, improves the probability of capturing the internal solitary wave data, and simultaneously improves the accuracy of inverting the internal solitary wave characteristic parameters.
2. The invention adopts matrix observation data, can more accurately measure the space-time distribution of local temperature salt, and is more beneficial to the extraction of internal solitary wave characteristic parameters.
3. The invention adopts hierarchical clustering analysis and considers space-time factors, and the data grouping method is closer to the actual situation.
4. The invention adopts the internal solitary wave theory to invert the characteristic parameters, and the inversion result is more accurate.
Drawings
FIG. 1, a flow chart of the present invention;
and 2, an internal solitary wave theory optimal application range diagram.
Detailed Description
The following description is of the preferred embodiments of the invention and is not intended to limit the scope of the invention, but is intended to cover any modifications, equivalents, and improvements within the spirit and principles of the invention.
In an initial embodiment, the invention provides an extraction algorithm of internal solitary wave parameters based on an underwater glider, wherein the extraction algorithm adopts a glider matrix for observation, and the glider matrix refers to: and a plurality of underwater gliders are put in different places in a space matrix form at the same time (namely, the plurality of underwater gliders are arranged in a space matrix form), and the temperature salt space-time variation data of the target sea area can be more accurately captured through the simultaneous observation of the plurality of underwater gliders, so that the characteristic parameters of the internal solitary waves can be more accurately estimated. As shown in fig. 1 and 2, the method comprises the following steps:
step 1, extracting space-time data observed by an underwater glider, wherein the space-time data comprises a glider machine number, an observation section number, a measuring point longitude and latitude, an observation time, a temperature, a salinity, a density and a water depth, and analyzing the space-time data by adopting a hierarchical clustering analysis method to obtain a simultaneous space-time data set;
step 2, carrying out equal-depth interval interpolation on the data in the simultaneous space data set to obtain equal-depth average data, then comparing the single-section data with the equal-depth average data, calculating Euclidean distance, and judging; the single profile data refers to: when each group of data obtained by hierarchical clustering analysis has a plurality of section data of different gliders at different times, single section information in the same group is single section data; comparing the data with average data in the same group to reflect the difference between a single section and the average data;
step 3, extracting suspicious sequences with equal-close surfaces with maximum depth difference larger than 25m and depth where the maximum depth difference is larger than 20m, and carrying out data and calculation on the machine number, the section number and the average density of the suspicious sequences to obtain ocean layering conditions, floating frequency and internal solitary wave amplitude;
and 4, selecting a proper internal solitary wave theory to invert the influence of the internal solitary wave on the isopipe surface according to the ocean layering condition and the internal solitary wave amplitude and the optimal application range of the internal solitary wave theory, combining the internal solitary wave theory with the original profile, and finally outputting a density profile containing the internal solitary wave information.
In a further embodiment, as shown in fig. 1 and 2, in the step 1, a bottom-up (agglemerable) hierarchical clustering method is adopted, and the basic principle is that each sample in data is first treated as a class, then the classes are searched for according to the distances, the samples are combined step by step upwards, finally a total class group is formed, when the distances are calculated, firstly, euclidean distances (Euclidean distance) between the samples are calculated, and after the class group is formed, the distances between the classes are calculated; the class-to-class distance calculation method is a shortest distance method (single link) or a longest distance method (complex link), or a class average distance method (average link) or a center distance method (centroid link).
In a further embodiment, as shown in fig. 1 and 2, in the step 1, all sections observed by all gliders in the same task are extracted first, and machine numbers n corresponding to each of the observed sections are extracted respectively M Cross section number n P Longitude and latitude lon and lat of water inlet point and average time parameter of profile
Figure BDA0003827081940000071
And carrying out dimensionless treatment on the time, as shown in formula (1), so that the time meets the space-time clustering analysis order-of-magnitude requirement:
Figure BDA0003827081940000072
in the formula (1), the components are as follows,
Figure BDA0003827081940000073
is the average value of single profile observation time, and is expressed in seconds, r t Is the cluster analysis spatial threshold c S With time threshold c T Ratio of (r), i.e. r t ＝c S /c T ；
Then, a three-dimensional matrix of spatio-temporal data is constructed, having:
Figure BDA0003827081940000074
it is assumed that the threshold values of the cluster analysis time and the spatial distance, i.e. c mentioned above, are manually set by the data analyst according to the cluster standard (here the cluster standard refers to the threshold value of the cluster analysis time and the spatial distance S And c T ) The three-dimensional matrix is divided into m disjoint classes C 1 ，C 2 ，...，C m Two of them are C α And C β The distance of (2) is expressed as:
Figure BDA0003827081940000081
in the formula (3), delta is a distance operator, delta a，b Representing the Euclidean distance from sample a to sample b;
Figure BDA0003827081940000082
representation class C α Longitude and latitude and average time parameters of the ith sample;
in the calculation, the shortest distance method is expressed as:
Figure BDA0003827081940000083
the longest distance method is expressed as:
Figure BDA0003827081940000084
the average distance-like method is expressed as:
Figure BDA0003827081940000085
the center distance method is expressed as:
Figure BDA0003827081940000086
in the formula (7), the amino acid sequence of the compound,
Figure BDA0003827081940000087
and->
Figure BDA0003827081940000088
Respectively refer to class C α And C β Center of (C) α And C β Average of samples in (a).
In a further embodiment, as shown in fig. 1 and 2, in the step 1, the shortest distance method is selected to analyze the space-time data, and when the shortest distance method is analyzed, the original space-time data is divided into M groups of data according to the criteria that the longitude and latitude and the average time parameter distance are smaller than respective threshold values, wherein each group contains M profile information; when the same group of spatiotemporal data is the spatiotemporal consistent data, the group of glider-in-flight machine spatiotemporal data has similar observation results; after grouping the data, the group number N of the grouping is output G Machine number N M Cross section number N P : as shown in formula (8), formula (8) is the output data of step one and the input data of step two:
Data 1 ＝[N G ，N M ，N P ] (8)。
in a further embodiment, as shown in fig. 1 and 2, in the step 2, according to the obtained group number, machine number and profile number, the working time t, depth dp, temperature tp, salinity st and density ds of all the machine numbers and profile numbers in the same group are extracted for analysis; extracting the maximum submerging depth and the minimum floating depth in different profile observations from the same set of data as interpolation depth intervals of the set of data; the following formula is shown:
Figure BDA0003827081940000089
in the formula (9):
Figure BDA00038270819400000810
a depth interval representing the interpolation of the set of data; dp (dp) down Represents the submerging depth sequence, dp up Representing a floating depth sequence;
then, carrying out equal-depth interpolation on different section data in the same group of data to obtain interpolation results of temperature, salinity and density of each section in the same depth interval, wherein the interpolation results are shown in the following formula:
Figure BDA0003827081940000091
and then, the temperature and salt density results of the sections with the same depth and different sections are averaged to obtain an average sequence, wherein the average sequence is shown in the following formula:
Figure BDA0003827081940000092
in the formula (11), j represents the cross-section number in the same group, and i represents the data number in the same cross-section.
In a further embodiment, as shown in fig. 1 and 2, in the step 3, euclidean distances between each profile data and the average data are compared in the same depth interval, and a profile with the largest euclidean distance is selected as a suspicious sequence to perform a next analysis, where the following formula is shown:
Figure BDA0003827081940000093
in the formula (12): d (D) Eucl (A, B) representing Euclidean distance calculations for sequences A and B, the specific calculation formulas being not shown to the right of the equation;
Figure BDA0003827081940000094
the density sequence of the j-th section is calculated by a Newton interpolation method according to the measured density sequence in the formula (10),
Figure BDA0003827081940000095
representing the average density sequence of the group;
the suspicious section is analyzed to find the maximum value diff of the difference diff between the equal close face and the average equal close face max Its corresponding depth dp md Finally, only extracting the profile data with the equal-density surface difference value larger than 25m and the maximum depth difference value larger than 20m to perform the next analysis, and outputting the machine number N of the suspicious sequence M Cross section number N P Equal-surface-to-average equal-surface maximum difference diff max Maximum difference depth dp md Floating frequency N, average depth interval
Figure BDA0003827081940000096
Equal depth average Density->
Figure BDA0003827081940000097
Upper and lower layer thickness h 1 、h 2 Density ρ of upper and lower layers 1 、ρ 2 ；
Based on the output data of the suspicious sequence, the suspicious sequence is subjected to the next step of analysis: firstly, local water depth data are found according to the space coordinates of suspicious sections, and then the buoyancy frequency is calculated according to the measured density data
Figure BDA0003827081940000098
Frequency), as shown in the following equation:
Figure BDA0003827081940000101
in the formula (13): g represents gravitational acceleration, ρ represents sea water density;
finding the depth with the largest buoyancy frequency as the depth d of the density jump layer pyc The maximum diff of the density difference appears in the previous analysis max Corresponding depth d md Substituting formula (14), and estimating the maximum internal solitary wave amplitude at the density jump layer through linear interpolation or other interpolation modes:
Figure BDA0003827081940000102
in formula (14): a represents amplitude, diff max Represents the maximum value of the density difference, d md Represents the depth of occurrence of the maximum value of the density difference, d tot Represents the total water depth d pyc Indicating the depth of the dense jump.
In a further embodiment, as shown in fig. 1 and 2, in the step 4, a suitable internal solitary wave theory is selected to calculate the characteristic parameters of the internal solitary wave according to the thickness of the upper and lower layers and the amplitude of the internal solitary wave (Cui et al, 2021): the following table shows:
table 1 in solitary wave theory optimum range
Figure BDA0003827081940000103
Selecting a proper internal solitary wave theory to reconstruct a warm salt profile at the occurrence time of the internal solitary wave, wherein the change form of the amplitude along the depth is calculated in a linear fitting mode or in an exponential fitting mode according to the actual situation, so as to form an equal-density surface change delta P caused by the internal solitary wave, and the equal-density surface change delta P is matched with an initial density profile P 0 And combining to obtain density profile data which is finally output and contains the influence of the internal solitary wave on the isopipe, wherein the density profile data is shown in the following formula:
P＝P 0 +ΔP (15)
in formula (15): p represents the density profile of the output, P 0 Representing the raw density profile, Δp represents the calculated intra-solitary wave-induced isopycnic face change according to the algorithm.
Appendix: the expression of the internal solitary wave theory exemplifies:
a) KdV theory its wave front can be expressed as:
Figure BDA0003827081940000111
b) The internal solitary wave surface under eKdV theory:
Figure BDA0003827081940000112
c) Wave surface under mKdV theory:
Figure BDA0003827081940000113
d) The wave surface of the internal solitary wave under the MCC theory:
Figure BDA0003827081940000114
/>

Claims (7)

1. an internal solitary wave parameter extraction algorithm based on an underwater glider is characterized in that: the extraction algorithm adopts a glider matrix to observe, and the glider matrix refers to: a plurality of underwater gliders are put in at different places in a space matrix form at the same time, and the underwater gliders are used for simultaneous observation, comprising the following steps:
step 1, extracting space-time data observed by an underwater glider, wherein the space-time data comprises a glider machine number, an observation section number, a measuring point longitude and latitude, an observation time, a temperature, a salinity, a density and a water depth, and analyzing the space-time data by adopting a hierarchical clustering analysis method to obtain a simultaneous space-time data set;
step 2, carrying out equal-depth interval interpolation on the data in the simultaneous space data set to obtain equal-depth average data, then comparing the single-section data with the equal-depth average data, calculating Euclidean distance, and judging; the single profile data refers to: when each group of data obtained by hierarchical clustering analysis has a plurality of section data of different gliders at different times, single section information in the same group is single section data;
step 3, extracting suspicious sequences with equal-close surfaces with maximum depth difference larger than 25m and depth where the maximum depth difference is larger than 20m, and carrying out data and calculation on the machine number, the section number and the average density of the suspicious sequences to obtain ocean layering conditions, floating frequency and internal solitary wave amplitude;
and 4, selecting a proper internal solitary wave theory to invert the influence of the internal solitary wave on the isopipe surface according to the ocean layering condition and the internal solitary wave amplitude and the optimal application range of the internal solitary wave theory, combining the internal solitary wave theory with the original profile, and finally outputting a density profile containing the internal solitary wave information.
2. The internal solitary wave parameter extraction algorithm based on an underwater glider as claimed in claim 1, wherein: in the step 1, a bottom-up hierarchical clustering analysis method is adopted, when the distance is calculated, euclidean distance between samples is calculated firstly, and after a class group is formed, the distance between classes is calculated again; the class-to-class distance calculation method is a shortest distance method, a longest distance method, a class average distance method or a center distance method.
3. The internal solitary wave parameter extraction algorithm based on an underwater glider as claimed in claim 2, wherein: in the step 1, all sections observed by all gliders in the same task are extracted first, and the machine number n corresponding to each observed section is extracted respectively M Cross section number n P Longitude and latitude lon and lat of water inlet point and average time parameter of profile
Figure QLYQS_1
And carrying out dimensionless treatment on the time, as shown in formula (1), so that the time meets the space-time clustering analysis order-of-magnitude requirement:
Figure QLYQS_2
in the formula (1), the components are as follows,
Figure QLYQS_3
is the average value of single profile observation time, and is expressed in seconds, r t Is the cluster analysis spatial threshold c s With time threshold c T Ratio of (r), i.e. r t ＝c s /c T ；
Then, a three-dimensional matrix of spatio-temporal data is constructed, having:
Figure QLYQS_4
suppose that a three-dimensional matrix is divided into m disjoint classes C according to a clustering criterion 1 ，C 2 ，...，C m Two of them are C α And C β The distance of (2) is expressed as:
Figure QLYQS_5
in the formula (3), delta is a distance operator, delta a，b Representing the Euclidean distance from sample a to sample b;
Figure QLYQS_6
representation class C α Longitude and latitude and average time parameters of the ith sample;
in the calculation, the shortest distance method is expressed as:
Figure QLYQS_7
the longest distance method is expressed as:
Figure QLYQS_8
the average distance-like method is expressed as:
Figure QLYQS_9
the center distance method is expressed as:
Figure QLYQS_10
in the formula (7), the amino acid sequence of the compound,
Figure QLYQS_11
and->
Figure QLYQS_12
Respectively refer to class C α And C β Center of (C) α And C β Average of samples in (a).
4. An internal solitary wave parameter extraction algorithm based on an underwater glider as claimed in claim 3 wherein: in the step 1, the shortest distance method is selected to analyze the time-space data, and when the shortest distance method is analyzed, the original time-space data is divided into M groups of data according to the standard that the longitude and latitude and the average time parameter distance are smaller than respective threshold values, wherein each group of data comprises M section information; when the same group of spatiotemporal data is the spatiotemporal consistent data, the group of glider-in-flight machine spatiotemporal data has similar observation results; after grouping the data, the group number N of the grouping is output G Machine number N M Cross section number N P : as shown in formula (8):
Data 1 ＝[N G ，N M ，N P ] (8)。
5. the internal solitary wave parameter extraction algorithm based on an underwater glider as claimed in claim 4, wherein: in the step 2, according to the obtained group number, machine number and section number, extracting the operation time t, depth dp, temperature tp, salinity st and density ds of all the machine numbers and section numbers in the same group for analysis; extracting the maximum submerging depth and the minimum floating depth in different profile observations from the same set of data as interpolation depth intervals of the set of data; the following formula is shown:
Figure QLYQS_13
in the formula (9):
Figure QLYQS_14
a depth interval representing the interpolation of the set of data; dp (dp) down Represents the submerging depth sequence, dp up Representing a floating depth sequence;
then, carrying out equal-depth interpolation on different section data in the same group of data to obtain interpolation results of temperature, salinity and density of each section in the same depth interval, wherein the interpolation results are shown in the following formula:
Figure QLYQS_15
and then, the temperature and salt density results of the sections with the same depth and different sections are averaged to obtain an average sequence, wherein the average sequence is shown in the following formula:
Figure QLYQS_16
in the formula (11), j represents the cross-section number in the same group, and i represents the data number in the same cross-section.
6. The internal solitary wave parameter extraction algorithm based on the underwater glider as claimed in claim 5, wherein: in the step 3, the euclidean distance between each profile data and the average data is compared in the same depth interval, and the profile with the largest euclidean distance is selected as the suspicious sequence to be analyzed in the next step, wherein the suspicious sequence is represented by the following formula:
Figure QLYQS_17
in the formula (12): d (D) Eucl (A, B) representing Euclidean distance calculations for sequences A and B, the specific calculation formulas being not shown to the right of the equation;
Figure QLYQS_18
a density sequence representing the jth cross section calculated by Newton's interpolation method of the measured density sequence according to formula (10)>
Figure QLYQS_19
Representing the average density sequence of the group;
the suspicious section is analyzed to find the maximum value diff of the difference diff between the equal close face and the average equal close face max Its corresponding depth dp md Finally, only extracting the profile data with the equal-density surface difference value larger than 25m and the maximum depth difference value larger than 20m to perform the next analysis, and outputting the machine number N of the suspicious sequence M Cross section number N P Equal-surface-to-average equal-surface maximum difference diff max Maximum difference depth dp md Floating frequency N, average depth interval
Figure QLYQS_20
Equal depth average Density->
Figure QLYQS_21
Upper and lower layer thickness h 1 、h 2 Density ρ of upper and lower layers 1 、ρ 2 ；
Based on the output data of the suspicious sequence, the suspicious sequence is subjected to the next step of analysis: firstly, local water depth data are found according to the space coordinates of the suspicious section, and then the buoyancy frequency is calculated according to the measured density data, wherein the buoyancy frequency is shown in the following formula:
Figure QLYQS_22
in the formula (13): g represents gravitational acceleration, ρ represents sea water density;
finding the depth with the greatest buoyancy frequency as the depth of the density jump layerDegree d pyc The maximum diff of the density difference appears in the previous analysis max Corresponding depth d md Substituting formula (14), and estimating the maximum internal solitary wave amplitude at the density jump layer through linear interpolation or other interpolation modes:
Figure QLYQS_23
in formula (14): a represents amplitude, diff max Represents the maximum value of the density difference, d md Represents the depth of occurrence of the maximum value of the density difference, d tot Represents the total water depth d pyc Indicating the depth of the dense jump.
7. The internal solitary wave parameter extraction algorithm based on an underwater glider as claimed in claim 6, wherein: in the step 4, according to the thickness of the upper and lower layers and the amplitude of the internal solitary wave, selecting a proper internal solitary wave theory to calculate the characteristic parameters of the internal solitary wave, selecting a proper internal solitary wave theory to reconstruct the temperature salt profile at the occurrence time of the internal solitary wave, wherein the amplitude is calculated in a linear fitting mode along the change form of the depth or in an exponential fitting mode according to the actual situation, so as to form an equal-density surface change delta P caused by the internal solitary wave, and then the equal-density surface change delta P is matched with the initial density profile P 0 And combining to obtain density profile data which is finally output and contains the influence of the internal solitary wave on the isopipe, wherein the density profile data is shown in the following formula:
P＝P 0 +ΔP (15)
in formula (15): p represents the density profile of the output, P 0 Representing the raw density profile, Δp represents the calculated intra-solitary wave-induced isopycnic face change according to the algorithm.
CN202211064070.4A 2022-09-01 2022-09-01 Internal solitary wave parameter extraction algorithm based on underwater glider Active CN115422746B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211064070.4A CN115422746B (en) 2022-09-01 2022-09-01 Internal solitary wave parameter extraction algorithm based on underwater glider

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211064070.4A CN115422746B (en) 2022-09-01 2022-09-01 Internal solitary wave parameter extraction algorithm based on underwater glider

Publications (2)

Publication Number Publication Date
CN115422746A CN115422746A (en) 2022-12-02
CN115422746B true CN115422746B (en) 2023-04-25

Family

ID=84200278

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211064070.4A Active CN115422746B (en) 2022-09-01 2022-09-01 Internal solitary wave parameter extraction algorithm based on underwater glider

Country Status (1)

Country Link
CN (1) CN115422746B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110222306A (en) * 2019-06-06 2019-09-10 大连理工大学 A kind of improvement mode decomposition method analyzed and reconstructed suitable for interior estimates experimental flow field
CN113093605A (en) * 2021-03-31 2021-07-09 天津大学 Underwater glider networking control system and method for internal solitary wave observation
CN113640800A (en) * 2021-08-25 2021-11-12 中国人民解放军海军潜艇学院 Inversion method for inverting isolated wave data in ocean

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109253726B (en) * 2018-06-22 2020-05-05 东南大学 Underwater glider navigation positioning system and floating precision correction method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110222306A (en) * 2019-06-06 2019-09-10 大连理工大学 A kind of improvement mode decomposition method analyzed and reconstructed suitable for interior estimates experimental flow field
CN113093605A (en) * 2021-03-31 2021-07-09 天津大学 Underwater glider networking control system and method for internal solitary wave observation
CN113640800A (en) * 2021-08-25 2021-11-12 中国人民解放军海军潜艇学院 Inversion method for inverting isolated wave data in ocean

Also Published As

Publication number Publication date
CN115422746A (en) 2022-12-02

Similar Documents

Publication Publication Date Title
Yu et al. The application of autonomous underwater vehicles for interdisciplinary measurements in Massachusetts and Cape Cod Bays
CN111307266B (en) Sound velocity obtaining method and global ocean sound velocity field construction method based on same
CN111637874B (en) Multi-AUV layered detection system and detection method for red tide sea area
Callaghan et al. Two regimes of laboratory whitecap foam decay: Bubble-plume controlled and surfactant stabilized
Lucas et al. Measuring turbulent dissipation using a tethered ADCP
CN109376589A (en) ROV deformation target and Small object recognition methods based on convolution kernel screening SSD network
CN108985304B (en) Automatic sedimentary layer structure extraction method based on shallow profile data
CN108562891B (en) Sound source depth autonomous real-time tracking method under deep sea low signal-to-noise ratio condition
CN112348280B (en) Method for performing EOF analysis by using improved Argo buoy data
CN107515390B (en) Aerial target positioning method based on single vector sensor
CN115422746B (en) Internal solitary wave parameter extraction algorithm based on underwater glider
CN107942316A (en) Concentrate suspension movement velocity method of estimation in a kind of water based on multibeam sonar beamformer output signal
Shao et al. Verification of echosounder measurements of thickness and spatial distribution of kelp forests
CN113533188B (en) Real sea evaluation method for antifouling property of waterline area coating
CN112085779B (en) Wave parameter estimation method and device
CN114417601A (en) Method for quickly estimating mesoscale eddy underwater sound velocity field based on satellite altimeter data
Yoon et al. Behavior and frequency analysis of Aurelia aurita by using in situ target strength at a port in Southwestern Korea
Trump et al. Mapping small-scale along-front structure using ADCP acoustic backscatter range-bin data
Tang et al. Towards an integrated study of subglacial conditions in Princess Elizabeth Land, East Antarctica
CN113268068B (en) Bionic submersible vehicle-based mixed intelligent autonomous detection method for deep sea area
CN115079258B (en) On-line identification method for ocean bottom seismic signals based on wavelet analysis
CN104677414B (en) CTD data processing methods based on AUV
Paul The role of Zostera noltii in wave attenuation
CN108647692B (en) Ocean layer knot extraction method based on LH histogram
CN116562110A (en) Underwater target V-shaped internal wave wake simulation method integrating marine measured data

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant