CN-121831839-B - Beidou wave buoy signal unlocking artifact detection and spectrum consistency restoration method

CN121831839BCN 121831839 BCN121831839 BCN 121831839BCN-121831839-B

Abstract

The invention relates to the technical field of ocean observation and discloses a Beidou wave buoy signal unlocking artifact detection and spectrum consistency restoration method which comprises the steps of extracting an original measurement data sequence, generating a synchronous state data matrix, separating a three-dimensional trend removal displacement sequence, constructing a signal quality sign sequence, carrying out single wave segmentation, calculating a kinematic characteristic parameter based on a wave height comparison result, comparing a dynamic limit according to the kinematic characteristic parameter and a state sign, intercepting a positioning artifact wave band, solving an autoregressive model coefficient, carrying out bidirectional prediction weighted fusion, replacing abnormal data to generate a time domain reconstruction vertical sequence, solving an energy attenuation slope test theoretical deviation for the time domain reconstruction vertical sequence, calculating a spectrum verification factor mean square error by combining horizontal components, and enabling the error to be lower than an external output parameter in a specified upper limit. The invention combines dynamic physical limit checking artifact cause, and ensures the integrity and reliability of wave data through bidirectional deduction and frequency domain double check.

Inventors

ZHANG YONGQIANG
ZHANG GUANJUN
LIU JIANQIANG

Assignees

自然资源部第一海洋研究所

Dates

Publication Date: 20260508
Application Date: 20260311

Claims (10)

1. The method for detecting the out-of-lock artifact of the Beidou wave buoy signal and repairing the spectrum consistency is characterized by comprising the following steps of: Reading the observation message, extracting an original measurement data sequence, and aligning the time stamps to generate a synchronous state data matrix; performing local tangent plane conversion and high-pass filtering according to the synchronous state data matrix to separate a three-dimensional trending displacement sequence; Constructing a signal quality sign sequence based on the three-dimensional trending displacement sequence, performing single wave segmentation, comparing the single wave height with the background significant wave height, and calculating a kinematic characteristic parameter based on a comparison result; Performing logic judgment on the dynamic limit according to the kinematic characteristic parameters and the corresponding state identifiers, intercepting a positioning artifact wave band, resolving an autoregressive model coefficient, and performing bidirectional prediction weighted fusion to replace original abnormal data so as to repair signal unlocking artifacts, and generating a time domain reconstruction vertical sequence; And carrying out fast Fourier transform on the time domain reconstruction vertical sequence to obtain the theoretical deviation of the energy attenuation slope test, extracting horizontal components in the three-dimensional detrending displacement sequence and calculating a spectrum verification factor mean square error of the time domain reconstruction vertical sequence, and outputting high-quality ocean wave parameters outwards when the error is lower than a specified upper limit.
2. The method for detecting the out-of-lock artifact of the Beidou wave buoy signal and repairing the spectrum consistency according to claim 1, wherein the method is characterized in that in the steps of reading an observation message, extracting an original measurement data sequence, aligning a time stamp to generate the synchronous state data matrix, carrying out local tangential plane conversion and high-pass filtering according to the synchronous state data matrix, and separating out the three-dimensional detrack displacement sequence: If and only if the difference value of the adjacent time stamps is strictly greater than zero, mapping and aligning the extracted original measurement data sequence to a corresponding sampling time node by adopting a linear interpolation algorithm, and merging and constructing to obtain the synchronous state data matrix; The geodetic coordinate system parameters in the synchronous state data matrix are called to perform local tangential plane conversion, a north displacement sequence and an east displacement sequence are separated, a vertical elevation trend removal formula is adopted to subtract the high arithmetic mean value of ellipsoids in an observation time window from the current ellipsoids, and an original vertical displacement component is extracted; and performing high-pass filtering on the north displacement sequence, the east displacement sequence and the original vertical displacement component to eliminate extremely low-frequency drift, and separating the three-dimensional trending displacement sequence comprising a trending north displacement sequence, a trending east displacement sequence and a trending vertical displacement sequence.
3. The method for detecting the out-of-lock artifact of the Beidou wave buoy signal and repairing the spectrum consistency according to claim 1, wherein the step of constructing a signal quality sign sequence and performing single-wave segmentation based on the three-dimensional detrending displacement sequence specifically comprises the following steps: extracting state parameters under the corresponding time stamps, and comprehensively judging by adopting multi-element logical OR operation to generate a signal quality mark sequence; Obtaining vertical components in the three-dimensional trending displacement sequence, and forcedly skipping zero crossing interpolation when absolute values of continuous sampling points are smaller than a preset minimum floating point limit, and performing single-wave segmentation by taking adjacent zero crossing moments as time fences to obtain independent wave period sections; and the time period corresponding to the independent wave period is projected and combined onto the signal quality sign sequence to obtain an independent wave unit set.
4. The method for detecting out-of-lock artifacts and repairing spectral consistency of Beidou wave buoy signals according to claim 3, wherein the step of comparing the single wave height with the background significant wave height further comprises the following steps: The current analysis wave in the independent wave unit set is taken as a center to extend forwards and backwards bidirectionally to construct a statistical time window, the variance of wave surface displacement in the statistical time window is calculated to deduce a macroscopic background reference, and background significant wave height is obtained; Measuring the wave band extremum span aiming at the wave units in the independent wave unit set to obtain a single wave height; Comparing the single wave height with the background significant wave height, and executing extraction when the single wave height is larger than the product of a significant wave height multiple threshold value and the background significant wave height, so as to extract a suspected abnormal candidate wave set; And the significant wave height multiple threshold is a dimensionless multiplier proportion limit for judging that the transient abrupt wave height deviates from the sea state background energy level.
5. The method for detecting the out-of-lock artifact of the Beidou wave buoy signal and repairing the spectrum consistency according to claim 4, wherein the step of calculating the kinematic feature parameters based on the comparison result specifically comprises the following steps: Based on the comparison result, aiming at displacement sampling points in the suspected abnormal candidate wave set, obtaining a transient derivative by adopting a vertical acceleration five-point center difference formula, and solving to obtain vertical acceleration; based on ocean dynamics principle, obtaining a wavelength space scale by adopting a deep water dispersion relation wavelength calculation formula, and calculating wave steepness by adopting a wave steepness calculation formula in combination with the single wave height; Combining and extracting the vertical acceleration and the steep wave to form the kinematic characteristic parameter; and carrying out limit checking on the kinematic characteristic parameters, and generating a physical abnormal triggering state when the morphological distortion degree breaks through the dynamic limit.
6. The method for detecting and repairing the out-of-lock artifact of a Beidou wave buoy signal according to claim 5, wherein in the step of executing logic judgment on a dynamic limit according to the kinematic characteristic parameter and the corresponding state identification, intercepting a positioning artifact wave band and solving an autoregressive model coefficient: executing multidimensional logic comprehensive judgment by combining the physical abnormal triggering state and the state identifier corresponding to the suspected abnormal candidate wave set, and intercepting a positioning artifact wave band; Measuring the signal missing span of the positioning artifact wave band, and extracting the outside normal observation sampling points to form a sample sequence; and (3) pre-calculating the variance of the sample sequence, judging that the covariance matrix is close to degradation when the variance is lower than a preset machine floating point minimum limit, carrying out void filling by adopting low-order polynomial smooth interpolation to degrade the bypass autoregressive resolving process, and resolving to obtain an autoregressive model coefficient when the covariance matrix is not degraded.
7. The method for detecting and repairing the signal loss-of-lock artifact of a Beidou wave buoy according to claim 1, wherein in the step of implementing bidirectional predictive weighted fusion to replace original abnormal data to repair the signal loss-of-lock artifact and generating the time domain reconstruction vertical sequence: invoking the autoregressive model coefficient, deducing the positive data at the left end at the outer side to the right side as a reference, and reversely deducing the positive data at the right end at the outer side to the left side as a reference by the same principle to obtain a positive and negative two-way predictive value; And extracting the missing span in the positioning artifact wave band, distributing inverse proportion weights based on time intervals of distances between each target point to be estimated in the missing span and the left and right real data boundaries, and carrying out weighted fusion on the forward and reverse predicted values to replace original abnormal data so as to generate the time domain reconstruction vertical sequence.
8. The method for detecting out-of-lock artifacts and repairing spectral consistency of Beidou wave buoy signals according to claim 1, wherein in the step of performing fast fourier transform on the time domain reconstructed vertical sequence to obtain theoretical deviation of energy attenuation slope test: Performing fast Fourier transform optimization on the time domain reconstruction vertical sequence to locate the spectrum peak frequency corresponding to the energy main peak, and extending, cutting and extracting high-frequency balance frequency band data in the high-frequency direction by taking the set multiple of the spectrum peak frequency as a starting point; Projecting the high-frequency balance frequency band data into a double-logarithmic coordinate system, performing unitary linear regression fitting to obtain an energy attenuation slope, and calculating a spectral slope deviation between the energy attenuation slope and an attenuation slope theoretical value; When the spectrum slope deviation is smaller than a set spectrum slope deviation threshold, judging that the time domain reconstruction vertical sequence accords with theoretical deviation test; the attenuation slope theoretical value is an ideal index limit which is preset and used for representing the attenuation rule of the high-frequency energy of the deep water gravity wave along with the frequency; the spectral slope deviation threshold is an allowable floating point error limit for determining that the high-frequency energy attenuation characteristic of the reconstructed wave surface deviates from an ideal state.
9. The method for detecting the out-of-lock artifact of the Beidou wave buoy signal and repairing the spectrum consistency according to claim 1, wherein in the step of extracting horizontal components in the three-dimensional detrending displacement sequence and calculating a spectrum verification factor mean square error by the time domain reconstruction vertical sequence, the error is lower than a specified upper limit, and outputting high-quality ocean wave parameters outwards, the method is characterized in that: Extracting a trending east displacement sequence and a trending north displacement sequence in the three-dimensional trending displacement sequence, calculating power spectrum density, and implementing linear equal weight superposition to construct a horizontal total displacement power spectrum; Obtaining the time domain reconstruction vertical sequence to derive a vertical hydrodynamic power spectrum density, and combining and calculating the vertical hydrodynamic power spectrum density and the horizontal total displacement power spectrum to obtain a spectrum domain physical consistency check factor; a core frequency band containing main wave energy is framed to evaluate the global error to obtain a spectrum checking factor mean square error, and when the spectrum checking factor mean square error is lower than the prescribed upper limit, a high-quality ocean wave parameter is externally output; And, the prescribed upper limit is a maximum tolerance limit for characterizing the degree of coincidence of horizontal dynamics with vertical kinematic two-dimensional spectral energy.
10. The Beidou wave buoy signal out-of-lock artifact detection and spectrum consistency restoration system is characterized by being applied to the Beidou wave buoy signal out-of-lock artifact detection and spectrum consistency restoration method as claimed in any one of claims 1-9, and comprising the following steps: The data acquisition and preprocessing module (10) is used for analyzing the original observation message and executing coordinate system conversion and trending filtering to generate a three-dimensional trending displacement sequence; The anomaly detection and feature extraction module (20) establishes data connection with the data acquisition and preprocessing module (10) through a data link, and the anomaly detection and feature extraction module (20) is used for receiving the three-dimensional trending displacement sequence and segmenting single waves, further comparing the single wave height with the background significant wave height, and outputting a suspected anomaly candidate wave set and corresponding kinematic feature parameters based on a comparison result; the judging and reconstructing module (30) establishes data connection with the abnormality detection and feature extraction module (20) through a data link, and the judging and reconstructing module (30) is used for executing joint condition judgment and interpolation substitution according to the kinematic feature parameters to generate a time domain reconstruction vertical sequence; the verification and output module (40) establishes data connection with the judgment and reconstruction module (30) through a data link, and the verification and output module (40) is used for converting the time domain reconstruction vertical sequence into a frequency domain rechecking physical rule and outputting high-quality ocean wave parameters.

Description

Beidou wave buoy signal unlocking artifact detection and spectrum consistency restoration method Technical Field The invention relates to the technical field of ocean observation, in particular to a Beidou wave buoy signal unlocking artifact detection and spectrum consistency restoration method. Background Ocean wave observation is the basis of ocean science research and ocean resource development. The traditional wave observation mostly adopts a large buoy based on an accelerometer, and obtains displacement by measuring motion acceleration and integrating, so that the device does not depend on external signals, and data acquisition is relatively stable. With the development of satellite navigation technology, a small drifting buoy based on Beidou positioning is an important means for acquiring sea state data due to easy deployment and wide coverage range. Different from the measuring principle of the accelerometer buoy, the Beidou wave buoy directly acquires the position coordinates by receiving satellite signals. When the buoy is subjected to wave slapping or instant submergence, the antenna enters water to cause the Beidou signal to lose lock temporarily. The signal recovery moment generates positioning coordinate jump, and the jump data is subjected to band-pass filtering treatment to form positioning artifacts with larger amplitude and similar shape to actual waves in the displacement time sequence. Such artifacts are covert and can cause the final output wave statistics to deviate from true values. At present, the processing mode of such positioning artifacts of Beidou buoys mainly depends on direct data rejection or statistical threshold judgment. Part of the conventional schemes directly reject abnormal time periods only according to state parameters such as the number of visible satellites or position precision factors, and adopt linear interpolation for filling. The operation can cause a large amount of effective observation data to be lost under the sea condition that signals are frequently interrupted, so that the time continuity of wave time sequence is affected, and the data integrity is reduced. Meanwhile, part of schemes directly follow the quality control standard of the accelerometer buoy, and abnormal data are identified by setting a fixed wave height statistical threshold. Since the magnitude of the positioning artifact depends on the out-of-lock time and the degree of position jump, there is no deterministic correlation with the background sea state, and a fixed statistical threshold makes it difficult to distinguish the positioning artifact from normal transient waves. The judging method depending on the single statistical characteristic does not combine the special cause of the Beidou artifact, and can cause missed detection and false deletion of normal wave data. In addition, the existing processing means stay at the marking or discarding level for the abnormal wave band, and lack a verification means and a repairing mechanism combined with the wave hydrodynamic boundary. After filling the abnormal data, the conventional method does not establish a verification criterion of the frequency domain dimension, and cannot verify whether the reconstructed wave surface sequence accords with the physical rule of deep water wave evolution on the frequency domain characteristic. This lack of physical consistency constraints in the repair results in a lack of reliability in the final output wave energy spectrum analysis results. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a Beidou wave buoy signal unlocking artifact detection and spectrum consistency restoration method, which solves the problems that when the Beidou signal is unlocked in a short time or distorted due to external interference, the conventional data processing means are difficult to accurately identify an artifact interval, and the restored wave surface sequence is separated from the ocean dynamics rule to cause the distortion of the wave spectrum characteristics in the existing wave measurement data. In order to achieve the purpose, the invention is realized by the following technical scheme that the method for detecting the unlocking artifact of the Beidou wave buoy signal and repairing the spectrum consistency comprises the following steps: Reading the observation message, extracting an original measurement data sequence, and aligning the time stamps to generate a synchronous state data matrix; Performing local tangent plane conversion and high-pass filtering according to the synchronous state data matrix to separate a three-dimensional trending displacement sequence; Constructing a signal quality sign sequence based on the three-dimensional trending displacement sequence, performing single wave segmentation, comparing the single wave height with the background significant wave height, and calculating kinematic characteristic parameters based on a comparison result; Performing logic ju