Ocean gravity data error processing method based on adaptive variational mode decomposition

Abstract

The application discloses a marine gravity data error processing method based on adaptive variational mode decomposition, relates to the technical field of marine gravity data processing, and comprises the following steps: acquiring shipborne gravity data and corresponding reference data, performing position correspondence and scale analysis, and obtaining scale data; extracting shipborne change data and reference change data based on the scale data, and identifying a protected area; performing adaptive variational mode decomposition and regional mode extraction on the shipborne gravity data, combining the protected area, and reconstructing to obtain regional background data; extracting local residual data from the shipborne gravity data in the protected area, decomposing the local residual data, and combining to obtain mode data; performing mode attribution processing on the mode data, obtaining attribution data, and performing hierarchical reconstruction based on the attribution data and the mode data, to obtain marine gravity data after error processing; and the application can retain local real gravity anomaly information in marine gravity data when the scale of reference data does not match.

Description

Technical Field

[0001] This application relates to the field of marine gravity data processing technology, and more specifically, to a method for processing marine gravity data errors based on adaptive variational mode decomposition. Background Technology

[0002] Marine gravity data can reflect seafloor topographic relief, seafloor faults, local density bodies, sedimentary basin boundaries, and seamounts, among other seafloor structural features. In shipborne marine gravity measurements, the survey vessel continuously collects gravity data along a pre-defined survey line. Shipborne gravity data typically has a high sampling density along the survey line, capable of recording short-distance gravity anomalies caused by local seafloor structures. However, due to the continuous movement of the survey vessel at sea, shipborne gravity data is easily contaminated with errors caused by hull heave, roll, pitch, speed changes, instrument vibration, positioning errors, and sea state disturbances. Therefore, shipborne gravity data needs to undergo error processing before being used for seafloor structural interpretation, seafloor topography inversion, marine geodesy, or resource surveys to reduce error interference and retain as much geologically significant gravity anomaly information as possible.

[0003] Existing methods for processing ocean gravity data typically incorporate satellite gravity data, historical gravity grid data, or regional gravity models as benchmark data to evaluate or constrain the error processing results of shipborne gravity data. Some methods employ variational mode decomposition (VMD) or adaptive variational mode decomposition (ADMD) to decompose shipborne gravity data into multiple modal components. These components are then filtered and reconstructed based on their frequency, energy, statistical characteristics, or differences from the benchmark data. While these methods can mitigate random and navigational disturbances to some extent, the benchmark data and shipborne gravity data do not share the same spatial scale. The spatial resolution scale of the benchmark data usually differs from the sampling scale along the survey line of the shipborne gravity data, making it more suitable for representing regional or long-wavelength gravity variations. Shipborne gravity data, acquired continuously along the survey line, more readily reflects local gravity anomalies corresponding to small-scale structures such as seamount margins, submarine fault zones, localized high-density rock masses, and trench slope breaks.

[0004] When the baseline data cannot fully represent the true local gravity anomalies, the local anomaly mode components obtained through variational mode decomposition or adaptive variational mode decomposition may exhibit significant differences from the baseline data, low correlation, or cause large deviations in the overall error evaluation index. If the overall consistency between the reconstructed results and the baseline data is primarily used as the criterion for mode selection, local true gravity anomaly modes are easily classified as error modes and weakened or eliminated in subsequent reconstruction processing. After such processing, although the marine gravity data may be smoother overall or closer to the baseline data in statistical indicators, the local anomaly information with actual geological significance in the shipborne gravity data will be weakened.

[0005] When local true gravity anomaly information is weakened, the application value of shipborne gravity data in identifying small-scale seafloor structures, judging local density anomalies, determining seafloor fault boundaries, and resource surveys is reduced. This leads to the underestimation of true seafloor structural anomalies, obscuring the location of anomaly boundaries, and even causing subsequent seafloor topography or density inversion results to deviate from actual geological conditions. Therefore, the fundamental flaw of the above processing method lies in its failure to adequately distinguish the spatial scale differences between the benchmark data and shipborne gravity data when using benchmark data for modal screening and reconstruction. It also lacks a basis for identifying and retaining local true gravity anomalies that the benchmark data cannot fully represent. This mismatch in scale from the benchmark data can easily mislead shipborne gravity data, causing local true gravity anomaly modes to be treated as error modes. Therefore, in the error processing of marine gravity data using benchmark data for modal screening and reconstruction, it is still necessary to address the problem of misidentification of local true gravity anomaly modes due to the mismatch in the spatial scale of the benchmark data.

[0006] In view of this, this application proposes an error processing method for ocean gravity data based on adaptive variational mode decomposition to solve the above problems. Summary of the Invention

[0007] To overcome the aforementioned shortcomings of the prior art and achieve the above objectives, this application provides the following technical solution: a method for processing ocean gravity data errors based on adaptive variational mode decomposition, comprising:

[0008] Obtain shipboard gravity data and corresponding baseline data, perform positional correspondence and scale analysis on the shipboard gravity data and baseline data to obtain scale data;

[0009] Based on scale data, continuous changes in shipborne gravity data and baseline data are extracted to obtain shipborne change data and baseline change data. The differences in changes between shipborne change data and baseline change data are then identified to obtain the protected area.

[0010] Based on scale data, adaptive variational mode decomposition is performed on shipborne gravity data to obtain survey line mode components. Based on scale data, regional mode extraction is performed on the survey line mode components to obtain regional mode components. Based on the protected area, the regional mode components are reconstructed into segments to obtain regional background data.

[0011] Based on regional background data, local residuals are extracted from shipborne gravity data within the protected area to obtain local residual data. Adaptive variational mode decomposition is then performed on the local residual data to obtain local modal components. Finally, the survey line modal components and local modal components are combined to obtain modal data.

[0012] Based on shipborne variation data, baseline variation data, protected areas, and modal data, modal data is processed for modal attribution to obtain attribution data;

[0013] Based on the attribution data and modal data, hierarchical reconstruction processing is performed to obtain ocean gravity data after error processing.

[0014] Furthermore, the method for obtaining scale data by performing positional correspondence and scale analysis on shipborne gravity data and reference data includes:

[0015] The positions of the measuring points in the shipborne gravity data and the reference positions in the reference data are processed to obtain position correspondence data.

[0016] Based on the location-corresponding data, the shipborne gravity data and the reference data are extracted using a unified sampling interval, reference resolution scale, survey line analysis scale, and local residual scale.

[0017] The regional background scale is calculated based on the baseline resolution scale and the survey line analysis scale to obtain the regional background scale.

[0018] Based on the unified sampling interval, survey line analysis scale, regional background scale, and local residual scale, the protection zone determination window length, minimum continuous length of the protection zone, range of candidate modes, range of candidate penalty factors, and regional mode screening scale are generated to obtain scale data.

[0019] Furthermore, the method for identifying the differences between shipboard change data and baseline change data to obtain the protected area includes:

[0020] Extract the direction of shipborne change, the amplitude of shipborne change, and the continuous length of shipborne change from the shipborne change data; extract the reference change direction, the reference change amplitude, and the reference change continuous length from the reference change data.

[0021] A consistency comparison is performed between the ship's load change direction and the reference change direction to obtain the direction comparison result;

[0022] The difference between the shipborne variation amplitude and the baseline variation amplitude is calculated to obtain the variation difference value;

[0023] The average and standard deviation of multiple variation values ​​within the same survey line are statistically analyzed to obtain the protection difference threshold.

[0024] Based on the direction comparison results, change difference values, protection difference thresholds, continuous length of shipborne change, continuous length of benchmark change, and minimum continuous length of the protection zone, the same range of locations along the survey line corresponding to the shipborne change data and benchmark change data is segmented to obtain the protection zone.

[0025] Furthermore, the method for calculating the difference between the shipborne variation amplitude and the reference variation amplitude to obtain the variation difference value includes:

[0026] Extract the ship-borne variation amplitude and the reference variation amplitude within the same location range along the survey line;

[0027] The absolute value of the difference between the shipborne variation amplitude and the reference variation amplitude is calculated to obtain the variation amplitude difference.

[0028] A uniform sampling interval is extracted from the scale data, and the average absolute value of the difference between adjacent shipborne gravity observations and adjacent reference gravity values ​​in the corresponding reference data is calculated. The larger of the two average values ​​is taken as the allowable deviation of the variation amplitude.

[0029] Using the allowable deviation of the change amplitude as the normalization scale, the difference in change amplitude is normalized to obtain the change difference value.

[0030] Furthermore, the method for obtaining the survey line modal components by performing adaptive variational mode decomposition on shipborne gravity data based on scale data includes:

[0031] Extract the range of candidate modal numbers and the range of candidate penalty factors from the scale data;

[0032] Based on the range of candidate modes and the range of candidate penalty factors, candidate variational mode decomposition is performed on the shipborne gravity data to obtain the candidate decomposition results;

[0033] The candidate decomposition results are evaluated for modal center frequency interval, modal reconstruction error, and residual energy to obtain adaptive variational mode decomposition parameters.

[0034] Adaptive variational mode decomposition is performed on the shipborne gravity data according to the adaptive variational mode decomposition parameters to obtain the survey line mode components.

[0035] Furthermore, the method for obtaining regional background data by segment reconstruction of regional modal components based on protected areas includes:

[0036] Within the protected area, the regional modal components are superimposed and reconstructed to obtain the regional background reconstruction value within the protected area;

[0037] Based on the regional background reconstruction value at the boundary of adjacent protected areas, the non-protected sections between adjacent protected areas are reconstructed proportionally along the survey line.

[0038] Based on the regional background reconstruction values ​​at the boundaries of adjacent protected areas, the end constraint values ​​and translation amounts of the end unprotected sections are determined, and the end unprotected sections located at the beginning or end of the survey line are translated and reconstructed to obtain the regional background data.

[0039] Furthermore, the method for performing adaptive variational mode decomposition on the local residual data to obtain local modal components includes:

[0040] Based on the local residual scale in the scale data, configure the range of local candidate modes and the range of local penalty factor candidates for the local residual data.

[0041] Based on the range of local candidate modes and the range of local penalty factor candidates, local candidate variational mode decomposition is performed on the local residual data to obtain the local candidate decomposition results;

[0042] Local candidate decomposition results are filtered by local center frequency interval, local reconstruction error and local residual energy to obtain local adaptive variational mode decomposition parameters;

[0043] Adaptive variational mode decomposition is performed on the local residual data according to the local adaptive variational mode decomposition parameters to obtain local modal components.

[0044] Furthermore, the method for performing modal attribution processing on the modal data to obtain attribution data includes:

[0045] Modal variation data is obtained by extracting the modal variation data from the transect modal components and local modal components in the modal data.

[0046] Based on the protected area, the modal change data corresponding to the modal components of the survey line and the baseline change data are continuously compared to obtain the regional background attribution results.

[0047] Based on the protected area, the modal change data, shipborne change data and baseline change data corresponding to the local modal components are continuously compared to obtain the local signal attribution results.

[0048] Based on modal variation data, protected area and shipborne gravity data, error attribution is determined for modal data, and error attribution results are obtained;

[0049] Prioritize the regional background attribution results, local signal attribution results, and error attribution results to obtain the attribution data.

[0050] Furthermore, the method for determining the error attribution of modal data based on modal variation data, protected area data, and shipborne gravity data, and obtaining the error attribution result, includes:

[0051] Based on the modal change data, random error assignment is determined for modal components whose central scale is smaller than the local residual scale and whose switching ratio between the rising and falling directions satisfies the alternating change condition, and the random error assignment results are obtained.

[0052] Based on the acquisition time and measurement point location in the shipborne gravity data, the locations of course changes, speed changes, and measurement interruptions are identified. The navigation error attribution is determined for modal components with concentrated energy in the vicinity that are inconsistent with the baseline change data, and the navigation error attribution result is obtained.

[0053] Based on the boundary of the protected area, the transition error attribution is determined for the modal components with concentrated energy near the boundary and discontinuous modal reconstruction values ​​on both sides of the boundary, and the transition error attribution result is obtained.

[0054] The results of random error assignment, navigation error assignment, and transition error assignment are summarized to obtain the error assignment results.

[0055] Furthermore, the method for obtaining error-processed ocean gravity data by performing hierarchical reconstruction based on attribution data and modal data includes:

[0056] Based on the attribution data, the survey line modal components attribution to the regional background are extracted from the modal data and superimposed to obtain the regional background layer data.

[0057] Based on the attribution data, the local modal components of the local signal attribution are extracted from the modal data and superimposed according to the protected area to obtain the local signal layer data;

[0058] Modal components for random error assignment, navigation error assignment, and transient error assignment are reconstructed and excluded.

[0059] The regional background layer data and local signal layer data are overlaid, and the location of the protected area boundary and the location of the measuring points near the protected area boundary are reconstructed to obtain the ocean gravity data after error processing.

[0060] Compared with existing technologies, the technical effects and advantages of the ocean gravity data error processing method based on adaptive variational mode decomposition proposed in this application are as follows:

[0061] By acquiring shipborne gravity data and corresponding baseline data, and performing positional correspondence and scale analysis on the shipborne gravity data and baseline data, scale data can be obtained for subsequent continuous variation extraction and mode decomposition processing. Scale data provides scale constraints for subsequent continuous variation extraction, adaptive variational mode decomposition, regional mode extraction, and local residual extraction, reducing the risk of misjudgment caused by scale mismatch when directly using baseline data to constrain mode screening and reconstruction.

[0062] The protected area is determined by the identification results of the variation differences between shipborne and baseline variation data. It is used to identify sections where shipborne gravity data and baseline data have a continuous variation correspondence within the same location along the survey line. The protected area provides a section basis for subsequent section reconstruction processing of regional modal components and extraction of local residuals within the protected area, improving the pertinence of local residual extraction within the protected area and enhancing the accuracy of the protected area in representing the continuous variation correspondence between shipborne gravity data and baseline data.

[0063] In the process of constructing the regional background data, adaptive variational mode decomposition is performed on the shipborne gravity data to obtain the survey line modal components. Regional modal components are then extracted based on scale data, and the regional modal components are reconstructed in segments in conjunction with the protected area data to obtain the regional background data. The regional background data is reconstructed by combining the regional background reconstruction values ​​at the protected area boundary, the proportion of the distance along the survey line in the non-protected area, and the end constraint values ​​and translations of the end non-protected area segments. This improves the continuity of the regional background data at the protected area boundary and in the non-protected area segments, and reduces the possibility of confusion between regional background changes and local anomalies.

[0064] For shipborne gravity data within the protected area, local residual data is first extracted based on regional background data. Then, adaptive variational mode decomposition is performed on the local residual data to obtain local modal components. The survey line modal components and local modal components are combined into modal data. Local changes within the protected area after subtracting the regional background can be expressed through local modal components, alleviating the problem of unclear mode attribution caused by the overlap of regional background, local anomalies, and error components in single-scale mode decomposition.

[0065] Modal attribution processing combines shipborne variation data, baseline variation data, protected area data, and modal data to determine the attribution of survey line modal components and local modal components, resulting in attribution data. This attribution data can distinguish between regional background attribution, local signal attribution, and error attribution, reducing the influence of relying solely on modal frequency, energy, or overall differences from baseline data for modal selection, and lowering the possibility of local true gravity anomaly modes being misclassified as error modes.

[0066] The layered reconstruction process, based on assigned and modal data, uses the modal components of the survey lines assigned to the regional background and the local modal components assigned to local signals for reconstruction. It excludes the modal components assigned to random errors, navigation errors, and transition errors from being included in the overlay reconstruction as regional background layer data or local signal layer data. This reduces error interference, preserves geologically significant local gravity anomalies in the shipborne gravity data, and improves the reliability of the error-processed marine gravity data in identifying small-scale seafloor structures, determining local density anomalies, and identifying seafloor fault boundaries. Attached Figure Description

[0067] Figure 1 This is a flowchart of the ocean gravity data error processing method based on adaptive variational mode decomposition according to an embodiment of this application;

[0068] Figure 2 This is a schematic diagram of the reconstruction of the regional background data segment in an embodiment of this application. Detailed Implementation

[0069] The technical solutions of this application will be described in detail, clearly, and completely below with reference to the accompanying drawings of the embodiments. It should be particularly noted that the specific embodiments described below are only used to better illustrate and explain the technical solutions of this application, and are intended to enable those skilled in the art to better understand and implement this application, and should not be construed as limiting the scope of protection of this application. Without departing from the spirit and substance of this application, those skilled in the art can modify, adjust, or make equivalent substitutions based on the content disclosed in this application, and these modifications, adjustments, or equivalent substitutions should all be considered within the scope of protection of this application.

[0070] Example 1:

[0071] Please see Figure 1 As shown, this embodiment provides a method for processing ocean gravity data errors based on adaptive variational mode decomposition, including:

[0072] S101, acquire shipboard gravity data and corresponding baseline data, perform positional correspondence and scale analysis on the shipboard gravity data and baseline data to obtain scale data.

[0073] In practice, shipborne gravity data consists of gravity observation data continuously recorded along the ocean survey line. The shipborne gravity data includes the survey line number, acquisition time, measurement point location, and shipborne gravity observation value. The survey line number is used to distinguish different measurement segments, the acquisition time is used to indicate the recording time of the shipborne gravity observation value, the measurement point location is used to indicate the spatial location of the shipborne gravity observation value in the ocean survey line, and the shipborne gravity observation value is used to indicate the gravity observation result at the measurement point location.

[0074] The reference data corresponding to the shipborne gravity data is the reference gravity data that has a spatial correspondence with the shipborne gravity data. The reference data includes the reference position, reference gravity value, and reference resolution. The reference data can be historical ocean gravity survey line data, ocean gravity reference network data, or satellite altimetry inversion gravity anomaly data. The reference position is used to establish a spatial correspondence with the measurement point position. The reference gravity value is used to provide the reference gravity result at the reference position. Multiple reference gravity values ​​form the reference gravity variation according to the distance along the survey line. The reference resolution is used to represent the spatial resolution capability of the reference data.

[0075] When matching the shipboard gravity data with the reference data, the positions of the measuring points and the reference positions are unified under the same coordinate reference, and the distance along the measuring line is established according to the measuring line sequence of the shipboard gravity data; the distance along the measuring line is used to represent the positional relationship between different shipboard gravity observations in the same measuring line.

[0076] For each shipborne gravity observation, the nearest reference position is found in the reference data based on the measurement point location. When the distance between the nearest reference position and the measurement point location is less than or equal to the corresponding distance threshold, the reference gravity value corresponding to the nearest reference position is taken as the reference gravity value corresponding to the shipborne gravity observation. The corresponding distance threshold is determined based on the reference resolution scale and the average distance along the measurement line between adjacent measurement point locations. The corresponding distance threshold is used to limit the maximum spatial distance between the reference position and the measurement point location that can be directly corresponded. The smaller value between the reference resolution scale and the average distance along the measurement line between adjacent measurement point locations is determined as the corresponding distance threshold.

[0077] When the distance between the nearest reference position and the measuring point position is greater than the corresponding distance threshold, multiple neighboring reference positions are determined in the reference data with the measuring point position as the center and the interpolation search distance as the search radius. The interpolation search distance is determined based on the average distance along the survey line between the reference resolution scale and the adjacent measuring point positions. The interpolation search distance is not less than the larger of the average distance along the survey line between the reference resolution scale and the adjacent measuring point positions, and the interpolation search distance is greater than the corresponding distance threshold. When the number of neighboring reference positions within the interpolation search distance is less than two, at least two reference positions are selected as neighboring reference positions in ascending order of the distance between the reference position and the measuring point position. Based on the distances between multiple neighboring reference positions and the measuring point position, distance-weighted interpolation is performed on the reference gravity values ​​corresponding to the multiple neighboring reference positions to obtain the reference gravity value corresponding to the shipborne gravity observation value.

[0078] After completing the location mapping, scale analysis is performed on the shipborne gravity data and the baseline data. The scale analysis includes: determining the uniform sampling interval based on the distance along the survey line between adjacent measuring points; determining the baseline resolution scale based on the baseline resolution; determining the survey line analysis scale based on the survey line length and the uniform sampling interval; determining the regional background scale based on the baseline resolution scale and the survey line analysis scale; determining the fluctuation range of adjacent measuring points based on the absolute value of the difference between adjacent shipborne gravity observations within the uniform sampling interval; and determining the local residual scale based on the uniform sampling interval and the fluctuation range of adjacent measuring points.

[0079] The length of the protected area decision window, the minimum continuous length of the protected area, the range of candidate modes, the range of candidate penalty factors, and the regional mode screening scale are determined based on the regional background scale and the local residual scale. The length of the protected area decision window corresponds to a length along the survey line that is less than or equal to the regional background scale, the minimum continuous length of the protected area is greater than or equal to the local residual scale, and the range of candidate modes is determined based on the scale ratio between the survey line analysis scale and the regional background scale. When the scale ratio between the survey line analysis scale and the regional background scale increases, the upper limit of the range of candidate modes increases. The range of candidate penalty factors is determined based on the uniform sampling interval and the local residual scale. When the local residual scale decreases, the upper limit of the range of candidate penalty factors increases. The regional mode screening scale is determined based on the regional background scale, and the regional mode screening scale is less than or equal to the regional background scale.

[0080] The scale data includes uniform sampling interval, benchmark resolution scale, survey line analysis scale, regional background scale, local residual scale, protected area determination window length, minimum continuous length of protected area, candidate mode number range, penalty factor candidate range, and regional mode screening scale; the scale data is used for subsequent continuous change extraction, adaptive variational mode decomposition, regional mode extraction, and local residual extraction.

[0081] S102, based on scale data, continuously changes in shipborne gravity data and baseline data are extracted to obtain shipborne change data and baseline change data. The differences in changes between shipborne change data and baseline change data are then identified to obtain the protected area.

[0082] In practice, based on the length of the protected area determination window in the scale data, the shipborne gravity data is processed into a sliding window according to the distance along the survey line. Within each sliding window, the difference between the shipborne gravity observation value at the end of the sliding window and the shipborne gravity observation value at the beginning of the sliding window is calculated to obtain the shipborne change difference.

[0083] The change direction threshold is determined based on the absolute value of the difference between adjacent shipborne gravity observations within a unified sampling interval and the absolute value of the difference between adjacent reference gravity values ​​after the position is completed. Specifically, the larger of the average value of the absolute value of the difference between adjacent shipborne gravity observations within a unified sampling interval and the average value of the absolute value of the difference between adjacent reference gravity values ​​after the position is completed is taken as the allowable deviation of the change amplitude, and the allowable deviation of the change amplitude is determined as the change direction threshold.

[0084] When the difference in ship load variation is greater than the threshold for the direction of change, the direction of ship load variation is determined to be the rising direction; when the difference in ship load variation is less than the negative of the threshold for the direction of change, the direction of ship load variation is determined to be the falling direction; when the difference in ship load variation is greater than or equal to the negative of the threshold for the direction of change and less than or equal to the threshold for the direction of change, the direction of ship load variation is determined to be the stable direction; the absolute value of the difference in ship load variation is determined to be the amplitude of ship load variation, and the length along the survey line where the direction of ship load variation remains consistent is determined to be the continuous length of ship load variation; the direction of ship load variation, the amplitude of ship load variation, and the continuous length of ship load variation constitute the ship load variation data.

[0085] Based on the protection zone determination window length in the scale data, the reference data is processed by sliding window according to the distance along the survey line; within the same survey line range corresponding to the location of the shipborne gravity data, the difference between the reference gravity value at the end position of the sliding window and the reference gravity value at the beginning position of the sliding window is calculated to obtain the reference change difference.

[0086] When the baseline change difference is greater than the change direction threshold, the baseline change direction is determined to be the rising direction; when the baseline change difference is less than the opposite of the change direction threshold, the baseline change direction is determined to be the falling direction; when the baseline change difference is greater than or equal to the opposite of the change direction threshold and less than or equal to the change direction threshold, the baseline change direction is determined to be the stable direction; the absolute value of the baseline change difference is determined to be the baseline change amplitude, and the length along the survey line where the baseline change direction remains consistent is determined to be the baseline change continuity length; the baseline change direction, baseline change amplitude, and baseline change continuity length constitute the baseline change data.

[0087] When identifying the difference between ship-borne change data and reference change data, the direction of ship-borne change and the direction of reference change within the same range along the survey line are compared, and the difference between the amplitude of ship-borne change and the amplitude of reference change is calculated. The difference value is used to indicate the degree of deviation between the ship-borne gravity data and the reference data within the same range along the survey line.

[0088] The variation difference value is obtained by normalizing the absolute value of the difference between the shipborne variation amplitude and the reference variation amplitude. The scale used for normalization is the allowable deviation of the variation amplitude. The allowable deviation of the variation amplitude is the larger of the average absolute value of the difference between adjacent shipborne gravity observations within a uniform sampling interval and the average absolute value of the difference between adjacent reference gravity values ​​after the position is matched. When the variation difference value is a dimensionless value after normalization, the protection difference threshold is a dimensionless threshold. The protection difference threshold is determined based on the statistical distribution of multiple variation difference values ​​within the same survey line after the position is matched. Specifically, the sum of the average value and standard deviation of multiple variation difference values ​​within the same survey line after the position is matched is determined as the protection difference threshold. The protection difference threshold is used to limit the allowable deviation between the shipborne variation amplitude and the reference variation amplitude.

[0089] When the same location range along the survey line simultaneously meets the following conditions: the direction of ship-borne change is consistent with the direction of reference change, the difference value of change is less than or equal to the protection difference threshold, the continuous length of ship-borne change is greater than or equal to the minimum continuous length of the protection zone, and the continuous length of reference change is greater than or equal to the minimum continuous length of the protection zone, the same location range along the survey line is determined as the initial protection zone.

[0090] When an initial protection zone is not determined, the range along the survey line in both the shipborne variation data and the baseline variation data where the shipborne variation direction is consistent with the baseline variation direction, the variation difference is minimal, the continuous length of the shipborne variation is greater than or equal to the minimum continuous length of the protection zone, and the continuous length of the baseline variation is greater than or equal to the minimum continuous length of the protection zone, is identified as a candidate protection zone and is then designated as the protection zone. When an initial protection zone is determined, it is designated as the protection zone. The protection zone represents a section where the shipborne gravity data and the baseline data have a continuous variation relationship within the same range along the survey line.

[0091] S103. Based on scale data, adaptive variational mode decomposition is performed on shipborne gravity data to obtain survey line mode components. Based on scale data, regional mode extraction is performed on the survey line mode components to obtain regional mode components. Based on the protected area, the regional mode components are reconstructed into segments to obtain regional background data.

[0092] Please see Figure 2 In specific implementation, the parameters for adaptive variational mode decomposition are determined for shipborne gravity data based on the range of candidate modes and the range of candidate penalty factors in the scale data. For each candidate mode in the range of candidate modes and each candidate penalty factor in the range of candidate penalty factors, candidate variational mode decomposition is performed on the shipborne gravity data according to the number of candidate modes and the candidate penalty factors to obtain the candidate decomposition results. The candidate decomposition results include multiple candidate modal components.

[0093] The candidate decomposition results are evaluated by determining the modal center frequency interval, modal reconstruction error, and residual energy. Multiple candidate modal components are sorted from low to high center frequency, with the modal center frequency interval being the minimum difference between the center frequencies of adjacent candidate modal components after sorting. The modal reconstruction error is the average of the absolute values ​​of the differences between the reconstructed value (after superimposing multiple candidate modal components at each measurement point) and the shipborne gravity observation value. The residual energy is the sum of squares of the differences between the shipborne gravity observation value and the reconstructed value (after superimposing multiple candidate modal components at each measurement point).

[0094] The center frequency interval threshold is determined based on the uniform sampling interval and the range of candidate modes, and the reconstruction error threshold is determined based on the fluctuation range of adjacent measurement points. The overall energy of the shipborne gravity observation is the sum of the squares of the differences between multiple shipborne gravity observations and the average value of the shipborne gravity observations in the shipborne gravity data. The allowable residual ratio is determined based on the scale ratio between the local residual scale and the measurement line analysis scale. The allowable residual ratio is used to limit the proportion of the remaining unreconstructed energy in the candidate decomposition results to the overall energy of the shipborne gravity observation. The residual energy threshold is determined by the product of the overall energy of the shipborne gravity observation and the allowable residual ratio.

[0095] When the center frequency interval of the modes corresponding to the number of candidate modes and the candidate penalty factor is greater than or equal to the center frequency interval threshold, the mode reconstruction error is less than or equal to the reconstruction error threshold, and the residual energy is less than or equal to the residual energy threshold, the number of candidate modes and the candidate penalty factor are used as optional decomposition parameters; when there are multiple sets of optional decomposition parameters, the set of optional decomposition parameters with the smallest mode reconstruction error is determined as the adaptive variational mode decomposition parameters.

[0096] When there are no optional decomposition parameters, the candidate decomposition result with the smallest residual energy is selected from the candidate decomposition results corresponding to the number of candidate modes and the candidate penalty factor. When there are multiple candidate decomposition results with the smallest residual energy, the candidate decomposition result with the largest mode center frequency interval is selected from the multiple candidate decomposition results with the smallest residual energy, and the number of candidate modes and the candidate penalty factor corresponding to the selected candidate decomposition result are determined as the adaptive variational mode decomposition parameters.

[0097] Adaptive variational mode decomposition (ADMD) parameters were used to perform ADMD on the shipborne gravity data to obtain the survey line modal components. The survey line modal components include mode number, center frequency, modal energy, position range along the survey line, and modal reconstruction value.

[0098] After obtaining the survey line modal components, regional mode extraction is performed on the survey line modal components based on scale data. The center scale is determined according to the center frequency and uniform sampling interval of the survey line modal components. The center frequency is the center frequency in the sampling sequence along the survey line, and the center scale is the length scale along the survey line corresponding to the center frequency. Survey line modal components with a center scale greater than or equal to the regional mode screening scale are determined as regional modal components. Survey line modal components with a center scale greater than or equal to the regional mode screening scale participate in subsequent segment reconstruction processing as regional modal components.

[0099] When performing segment reconstruction processing on regional modal components based on protected areas, the regional background reconstruction value formed by the superposition of regional modal components is retained within the protected area; outside the protected area, segment reconstruction processing is performed based on the regional background reconstruction value at the boundary of adjacent protected areas; for non-protected sections between adjacent protected areas, the reconstruction start point and reconstruction end point of the non-protected section are determined based on the regional background reconstruction value at the boundary of adjacent protected areas at both ends of the non-protected section.

[0100] Based on the distance ratio along the survey line, determine the regional background reconstruction value of each survey point in the non-protected section between the reconstruction start point and the reconstruction end point, so that the regional background reconstruction values ​​at both ends of the non-protected section are consistent with the regional background reconstruction values ​​at the boundaries of the adjacent protected areas.

[0101] For the unprotected sections at the beginning or end of the survey line, the reconstructed regional background value at the boundary of the adjacent protected area is used as the end constraint value. The difference between the end constraint value and the reconstructed regional background value formed by superimposing the regional modal components at the endpoint of the unprotected section near the boundary of the adjacent protected area is determined as the translation amount. The reconstructed regional background value formed by superimposing the regional modal components within the unprotected section is added to the translation amount to obtain the reconstructed regional background value for each measuring point within the unprotected section. After completing the segment reconstruction processing for all protected, unprotected, and unprotected sections, regional background data is obtained, which includes the measuring point locations and the reconstructed regional background values.

[0102] S104. Based on regional background data, local residuals are extracted from the shipborne gravity data within the protected area to obtain local residual data. Adaptive variational mode decomposition is then performed on the local residual data to obtain local modal components. Finally, the survey line modal components and local modal components are combined to obtain modal data.

[0103] In practice, within the protected area, the difference between the shipborne gravity observation value at each measuring point and the reconstructed regional background value corresponding to the same measuring point in the regional background data is calculated to obtain the local residual value. The local residual data is composed of the local residual values ​​corresponding to multiple measuring points within the protected area. The local residual data is used to represent the local changes remaining after deducting the regional background data within the protected area.

[0104] When performing adaptive variational mode decomposition on local residual data, the range of local candidate modes and the range of local penalty factor candidates are determined based on the local residual scale in the scale data. The upper limit of the range of local candidate modes is less than or equal to the upper limit of the range of candidate modes, and the lower limit of the range of local candidate modes is greater than or equal to one. When the local residual scale decreases, the upper limit of the range of local candidate modes increases, and the range of local penalty factor candidates is determined according to the local residual scale.

[0105] Local candidate variational mode decomposition is performed on parameter combinations within the range of local candidate modes and the range of local penalty factor candidates to obtain local candidate decomposition results; the local candidate decomposition results include multiple local candidate mode components; local center frequency interval judgment, local reconstruction error judgment, and local residual energy judgment are performed based on the local candidate decomposition results.

[0106] Multiple local candidate modal components are sorted from low to high according to their center frequencies. The local center frequency interval is the minimum value among the center frequency differences of adjacent local candidate modal components after sorting. The local reconstruction error is the average of the absolute values ​​of the differences between the reconstructed value after superimposing multiple local candidate modal components at each measurement point and the local residual value. The local residual energy is the sum of squares of the differences between the local residual value at each measurement point and the reconstructed value after superimposing multiple local candidate modal components.

[0107] The local center frequency interval threshold is determined based on the uniform sampling interval and the range of the number of local candidate modes. The local reconstruction error threshold is determined based on the fluctuation range of adjacent local residual values ​​in the local residual data. The overall energy of the local residual data is the sum of squares of the differences between multiple local residual values ​​and the average local residual value. The allowable residual ratio is consistent with the allowable residual ratio in S103. The allowable residual ratio is used to limit the proportion of the remaining unreconstructed energy in the local candidate decomposition results to the overall energy of the local residual data. The local residual energy threshold is determined based on the product of the overall energy of the local residual data and the allowable residual ratio.

[0108] When the number of local candidate modes and the local candidate penalty factor correspond to the local center frequency interval are greater than or equal to the local center frequency interval threshold, the local reconstruction error is less than or equal to the local reconstruction error threshold, and the local residual energy is less than or equal to the local residual energy threshold, the number of local candidate modes and the local candidate penalty factor are used as local optional decomposition parameters. When there are multiple sets of local optional decomposition parameters, the set of local optional decomposition parameters with the smallest local reconstruction error is determined as the local adaptive variational mode decomposition parameters.

[0109] When no local alternative decomposition parameters exist, the local candidate decomposition result with the smallest local residual energy is selected from the local candidate decomposition results corresponding to the number of local candidate modes and the local candidate penalty factor. When there are multiple local candidate decomposition results with the smallest local residual energy, the local candidate decomposition result with the largest local center frequency interval is selected from the multiple local candidate decomposition results with the smallest local residual energy, and the number of local candidate modes and the local candidate penalty factor corresponding to the selected local candidate decomposition result are determined as the local adaptive variational mode decomposition parameters.

[0110] Local adaptive variational mode decomposition parameters are used to perform adaptive variational mode decomposition on the local residual data to obtain local modal components. The local modal components include local mode number, local center frequency, local center scale, local modal energy, corresponding protected area range, and local modal reconstruction value.

[0111] When combining the survey line modal components and local modal components, the positional correspondence between the survey line modal components and local modal components is established according to the position range along the survey line. The modal source identifier, center scale, modal energy, and modal reconstruction value of the survey line modal components and local modal components are retained to obtain modal data. The modal source identifier includes the survey line modal source identifier and the local modal source identifier. The modal data includes the survey line modal components, local modal components, and the positional correspondence between the survey line modal components and local modal components.

[0112] S105, based on shipborne change data, baseline change data, protected area and modal data, modal data is processed for modal attribution to obtain attribution data.

[0113] In specific implementation, modal attribution processing is used to configure a modal attribution result for each modal component in the modal data. The modal components include the survey line modal components and local modal components in the modal data. Modal attribution processing is performed on the survey line modal components and local modal components in the modal data respectively to obtain attribution data. The attribution data includes modal number, modal source identifier, modal location range, modal center scale, modal energy, and modal attribution result. The modal attribution result includes regional background attribution, local signal attribution, random error attribution, navigation error attribution, and transition error attribution.

[0114] Before performing modal attribution processing on the modal data, the modal reconstruction values ​​of each modal component in the modal data are extracted according to the length of the protected area determination window to obtain the modal change direction, modal change amplitude, and modal change continuity length. The determination method of modal change direction, modal change amplitude, and modal change continuity length is consistent with the determination method of shipborne change direction, shipborne change amplitude, and shipborne change continuity length.

[0115] In the modal attribution process, the difference between the modal change amplitude and the reference change amplitude is obtained by normalizing the absolute value of the difference between the modal change amplitude and the reference change amplitude within the same measurement line location range. The scale used for normalization is consistent with the allowable deviation of the change amplitude in S102.

[0116] For survey line modal components, when the center scale of the survey line modal component is greater than or equal to the regional modal screening scale, and the continuous length of the survey line modal component within the protected area whose modal change direction is consistent with the baseline change direction is greater than or equal to the minimum continuous length of the protected area, the survey line modal component is assigned to the regional background.

[0117] For local modal components, when a local modal component is located within a protected area, the continuous length of the local modal component whose modal change direction is consistent with the shipborne change direction is greater than or equal to the minimum continuous length of the protected area, the continuous length of the local modal component whose modal change direction is consistent with the reference change direction is greater than or equal to the minimum continuous length of the protected area, and the center scale of the local modal component is greater than or equal to the local residual scale, the local modal component is assigned to a local signal.

[0118] For modal components, when the central scale of the modal component is smaller than the local residual scale, and the ratio between the number of times the modal component switches between rising and falling directions between adjacent measuring points is greater than or equal to the alternation ratio threshold, the modal component is assigned to random error. The number of times the modal component switches between rising and falling directions is not counted for stationary directions. The alternation ratio threshold is determined based on the switching ratio of rising and falling directions of each modal component within the same measuring line. Specifically, the average value and standard deviation of the switching ratio of rising and falling directions of each modal component within the same measuring line are used to determine the alternation ratio threshold. The alternation ratio threshold is used to determine whether the switching ratio of rising and falling directions of the modal component between adjacent measuring points meets the random error assignment condition.

[0119] In some embodiments, the heading change location, speed change location, and measurement interruption location are determined based on the acquisition time and measurement point location in the shipborne gravity data. The heading change location, speed change location, and measurement interruption location are used for navigation error attribution processing and are not used to change the positional correspondence between the shipborne gravity data and the reference data. The ratio between the distance along the measurement line between adjacent measurement point locations and the time interval between adjacent acquisition times is determined as the speed along the measurement line. When the absolute value of the difference between adjacent speeds along the measurement line is greater than the sum of the average difference of speeds along the measurement line and the standard deviation of the speed difference within the same measurement line, the measurement point location shared by adjacent speeds along the measurement line is determined as the speed change location.

[0120] The heading change position is determined by the angle formed by three consecutive measuring points along the survey line; when the angle along the survey line is greater than the sum of the average angle along the survey line and the standard deviation of the angle along the survey line within the same survey line, the intermediate measuring point is determined as the heading change position; the measurement interruption position is determined by the time interval between adjacent acquisition times; when the time interval between adjacent acquisition times is greater than the sum of the average acquisition time interval and the standard deviation of the acquisition time interval within the same survey line, the measuring point position corresponding to the later acquisition time is determined as the measurement interruption position.

[0121] For modal components, when the modal energy of a modal component in the vicinity of any of the following locations—heading change location, speed change location, and measurement interruption location—is greater than or equal to the total modal energy of the modal component within the modal location range, and the modal change direction is inconsistent with the reference change direction within the same range along the survey line, the modal component is assigned to the navigation error. The vicinity range is the range along the survey line centered on any of the following locations—heading change location, speed change location, and measurement interruption location—with the local residual scale as half its length. The length ratio between the length of the vicinity range and the length of the modal location range is determined as the energy concentration ratio threshold.

[0122] When the modal energy proportion is greater than or equal to the energy concentration proportion threshold, it means that the concentration of modal energy of the modal component in the vicinity of any of the following locations: heading change location, speed change location, and measurement interruption location is not less than the energy proportion formed by uniform distribution along the length of the survey line.

[0123] For modal components, when the modal energy of a modal component in the vicinity of any of the following locations—heading change location, speed change location, and measurement interruption location—accounts for a proportion greater than or equal to the total modal energy of the modal component within the modal location range, and the difference between the modal change amplitude and the reference change amplitude within the same measurement line location range is greater than the protection difference threshold, the modal component is assigned to the navigation error.

[0124] For modal components, when the modal energy of a modal component near the boundary of the protected area accounts for the total modal energy of the modal component within the modal position range, which is greater than or equal to the boundary energy ratio threshold, and the absolute value of the difference in modal reconstruction values ​​between adjacent measuring points on both sides of the protected area boundary is greater than the boundary continuity threshold, the modal component is assigned to the transition error. The area near the boundary of the protected area is the range along the measuring line centered on the boundary of the protected area and with the local residual scale as half the length. The length ratio between the length along the measuring line near the boundary of the protected area and the length of the modal position range is determined as the boundary energy ratio threshold, and the allowable deviation of the change amplitude in S102 is determined as the boundary continuity threshold.

[0125] When the modal energy proportion near the boundary of the protected area is greater than or equal to the boundary energy proportion threshold, it indicates that the concentration of modal energy of the modal components near the boundary of the protected area is not less than the energy proportion formed by uniform distribution along the length of the survey line.

[0126] When the same local modal component simultaneously satisfies the judgment conditions for two or more of the following modal assignment results: local signal assignment, navigation error assignment, transition error assignment, and random error assignment, the modal assignment result of the same local modal component shall be determined in the order of local signal assignment, navigation error assignment, transition error assignment, and random error assignment. When the same survey line modal component simultaneously satisfies the judgment conditions for two or more of the following modal assignment results: regional background assignment, navigation error assignment, transition error assignment, and random error assignment, the modal assignment result of the same survey line modal component shall be determined in the order of regional background assignment, navigation error assignment, transition error assignment, and random error assignment.

[0127] When the same local modal component does not meet the judgment conditions for local signal assignment, navigation error assignment, transition error assignment, and random error assignment, if the central scale of the same local modal component is smaller than the local residual scale, then the same local modal component is assigned to random error; if the central scale of the same local modal component is greater than or equal to the local residual scale, then the same local modal component is assigned to transition error.

[0128] When the same survey line mode component does not meet the judgment conditions for regional background assignment, navigation error assignment, transition error assignment, and random error assignment, if the central scale of the same survey line mode component is smaller than the local residual scale, then the same survey line mode component is assigned to random error assignment; if the central scale of the same survey line mode component is greater than or equal to the local residual scale, then the same survey line mode component is assigned to transition error assignment.

[0129] S106, based on the attribution data and modal data, performs hierarchical reconstruction processing to obtain ocean gravity data after error processing.

[0130] In practice, based on the attribution data, the survey line modal components belonging to the regional background are extracted from the modal data, and the survey line modal components belonging to the regional background are superimposed to obtain the regional background layer data; based on the attribution data, the local modal components belonging to the local signal are extracted from the modal data, and the local modal components belonging to the local signal are superimposed according to the corresponding protection zone range to obtain the local signal layer data.

[0131] Error suppression processing is performed on the modal components that are assigned to random error, navigation error, and transition error. The error suppression processing includes not including the modal components corresponding to random error, navigation error, and transition error as regional background layer data or local signal layer data in the superposition and reconstruction.

[0132] For the locations at the boundaries of the protected area and the locations of measuring points near the boundaries, the regional background layer data and local signal layer data of the adjacent measuring points in front of the boundary are superimposed to obtain the reconstructed value before the boundary; the regional background layer data and local signal layer data of the adjacent measuring points behind the boundary are superimposed to obtain the reconstructed value after the boundary; based on the ratio of the distance along the survey line between the measuring point location to be processed and the measuring points in front of and behind the boundary, the reconstructed values ​​in front of and behind the boundary are weighted and calculated to obtain the transitional reconstructed value of the measuring point location to be processed.

[0133] When there are no measuring points in front of or behind the boundary at the boundary of the protected area or at the measuring point location near the boundary of the protected area, the superposition value of the regional background layer data and the local signal layer data corresponding to the existing adjacent measuring point locations is determined as the transition reconstruction value of the corresponding measuring point location.

[0134] For the locations at the boundaries of the protected area and the locations near the boundaries of the protected area, the transition reconstruction value of the corresponding location is determined as the gravity value after error processing. For the locations at the locations other than the boundaries of the protected area and the locations near the boundaries of the protected area, the superposition value of the regional background layer data and the local signal layer data is determined as the gravity value after error processing. The ocean gravity data after error processing is composed of the location of the location and the gravity value after error processing.

[0135] Example 2:

[0136] Based on Example 1, this embodiment provides an optional calculation method for location correspondence, protected area identification, adaptive variational mode decomposition parameter screening, regional background segment reconstruction, mode attribution processing, and hierarchical reconstruction processing; the contents not described in this embodiment are the same as in Example 1.

[0137] S201, unify the coordinates of the measuring point positions in the shipborne gravity data and the reference positions in the reference data, and establish the distance along the measuring line according to the measuring line sequence of the shipborne gravity data to obtain the position corresponding data.

[0138] In practice, the measurement point positions in the shipborne gravity data and the reference positions in the reference data are converted to the same coordinate reference. The measurement point positions are sorted according to the measurement line order of the shipborne gravity data, and the distance along the measurement line between adjacent measurement point positions is calculated. The distance along the measurement line of the first measurement point position is recorded as zero. The distances between subsequent measurement point positions and the previous measurement point position are accumulated to obtain the distance along the measurement line corresponding to each measurement point position.

[0139] For the There are 1 measuring point locations, and the measuring point locations are denoted as _____. The shipboard gravity observation value is denoted as The distance along the survey line is denoted as The reference position is denoted as The baseline gravity value is denoted as Measurement point location relative to the reference position The distance between them is denoted as When the distance between the nearest reference position and the measuring point position is less than or equal to the corresponding distance threshold, the reference gravity value corresponding to the nearest reference position is determined as the reference gravity value corresponding to the measuring point position.

[0140] The interpolation search distance is determined based on the average distance along the survey line between the reference resolution scale and the adjacent measuring point positions. The interpolation search distance is greater than or equal to the larger of the average distances along the survey line between the reference resolution scale and the adjacent measuring point positions, and the interpolation search distance is greater than the corresponding distance threshold. When the distance between the nearest reference position and the measuring point position is greater than the corresponding distance threshold, the nearest reference position is determined in the reference data with the measuring point position as the center and the interpolation search distance as the search radius. When the number of nearest reference positions within the interpolation search distance is less than two, at least two reference positions are selected as nearest reference positions according to the distance between the reference position and the measuring point position from smallest to largest, and distance-weighted interpolation is used to determine the reference gravity value corresponding to the measuring point position. The distance-weighted interpolation uses the following formula:

[0141] ;

[0142] ;

[0143] In the formula, For the first The reference gravity value corresponding to each measuring point location. For the first The set of neighboring reference locations corresponding to each measuring point location For the first Distance weights corresponding to each neighboring reference location This is the distance correction constant. and They have the same dimensions. The location-corresponding data includes the survey line number, the location of the survey point, the distance along the survey line, the shipborne gravity observation value, and the corresponding reference gravity value.

[0144] S202, scale analysis is performed based on location-corresponding data to obtain scale data.

[0145] In practice, a uniform sampling interval is determined based on the average distance along the survey line between adjacent measuring points in the location-corresponding data; the survey line analysis scale is determined based on the difference between the maximum and minimum distances along the survey line for the same survey line; the benchmark resolution scale is determined based on the benchmark resolution in the benchmark data; and the geometric mean of the benchmark resolution scale and the survey line analysis scale is determined as the regional background scale.

[0146] The fluctuation range of adjacent measuring points is determined based on the absolute value of the difference between adjacent shipborne gravity observations within a uniform sampling interval. The integer multiple relationship between the local residual scale and the uniform sampling interval is determined based on the ratio of the standard deviation of the shipborne gravity observations to the fluctuation range of adjacent measuring points. Specifically, the ratio of the standard deviation of the shipborne gravity observations to the fluctuation range of adjacent measuring points is rounded up to obtain the local residual scale coefficient; when the local residual scale coefficient is less than one, it is set to one; the product of the uniform sampling interval and the local residual scale coefficient is determined as the local residual scale.

[0147] As the fluctuation range between adjacent measuring points increases, the local residual scale decreases. The local residual scale is used to limit the range of variation along the measuring line corresponding to the local residual data. The length of the protected area decision window, the minimum continuous length of the protected area, the range of candidate modes, the range of candidate penalty factors, and the regional mode selection scale are determined based on the regional background scale and the local residual scale. The length along the measuring line corresponding to the protected area decision window length is less than or equal to the regional background scale; the minimum continuous length of the protected area is greater than or equal to the local residual scale; and the regional mode selection scale is less than or equal to the regional background scale. The range of candidate modes is determined based on the scale ratio between the measuring line analysis scale and the regional background scale. As the scale ratio between the measuring line analysis scale and the regional background scale increases, the upper limit of the candidate mode range increases. The range of candidate penalty factors is determined based on the uniform sampling interval and the local residual scale. As the local residual scale decreases, the upper limit of the candidate penalty factor range increases.

[0148] The scale data includes uniform sampling interval, benchmark resolution scale, survey line analysis scale, regional background scale, local residual scale, protected area determination window length, minimum continuous length of protected area, candidate mode number range, penalty factor candidate range, and regional mode screening scale.

[0149] S203, based on the length of the protected area determination window in the scale data, continuously changes are extracted from the shipborne gravity data and the baseline data, and the differences in changes between the shipborne change data and the baseline change data are identified to obtain the protected area.

[0150] In practice, before extracting continuous changes, the allowable deviation of the change amplitude is determined based on the larger of the average absolute value of the difference between adjacent shipborne gravity observations within a unified sampling interval and the average absolute value of the difference between adjacent reference gravity values ​​after the position is matched. The allowable deviation of the change amplitude is then determined as the threshold for the change direction.

[0151] Divide the length of the protected area determination window by the uniform sampling interval and take the smallest integer greater than or equal to the quotient to obtain the number of measuring point intervals corresponding to the protected area determination window length. Perform sliding window processing on the shipborne gravity data according to the number of measuring point intervals. The difference between the shipborne gravity observation value at the end of the sliding window and the shipborne gravity observation value at the beginning of the sliding window is determined as the shipborne variation difference. The absolute value of the shipborne variation difference is determined as the shipborne variation amplitude. The shipborne variation direction is determined based on the relationship between the shipborne variation difference and the variation direction threshold.

[0152] A sliding window process is applied to the baseline gravity values ​​at the corresponding positions after completion, using the same number of measurement point intervals. The difference between the baseline gravity value at the end of the sliding window and the baseline gravity value at the beginning of the sliding window is determined as the baseline variation difference. The absolute value of the baseline variation difference is determined as the baseline variation amplitude. The baseline variation direction is determined based on the relationship between the baseline variation difference and the variation direction threshold. The continuous length along the measurement line with a consistent ship load variation direction is determined as the ship load variation continuous length, and the continuous length along the measurement line with a consistent baseline variation direction is determined as the baseline variation continuous length. The ship load variation direction, ship load variation amplitude, and ship load variation continuous length constitute the ship load variation data, and the baseline variation direction, baseline variation amplitude, and baseline variation continuous length constitute the baseline variation data.

[0153] Calculate the difference between the ship's load variation amplitude and the baseline variation amplitude. The difference is calculated using the following formula:

[0154] ;

[0155] In the formula, For the first The difference in change corresponding to each sliding window For the first The ship load variation amplitude corresponding to each sliding window For the first The baseline change amplitude corresponding to each sliding window To allow for deviation in the amplitude of change, This is a correction constant for the amplitude of gravity variation. and They have the same dimensions.

[0156] The protection difference threshold is calculated using the following formula:

[0157] ;

[0158] In the formula, To protect the difference threshold, To calculate the average of multiple variation values ​​within the same survey line after location correspondence, () represents the standard deviation of multiple variation values ​​within the same survey line after location correspondence has been completed.

[0159] When the same location range along the survey line simultaneously meets the following conditions: the direction of ship-borne change is consistent with the reference direction of change, the difference value of change is less than or equal to the protection difference threshold, the continuous length of ship-borne change is greater than or equal to the minimum continuous length of the protection zone, and the continuous length of the reference change is greater than or equal to the minimum continuous length of the protection zone, the same location range along the survey line is determined as the initial protection zone. When the initial protection zone is determined, the location range along the survey line where the direction of ship-borne change is consistent with the reference direction of change, the difference value of change is minimal, the continuous length of ship-borne change is greater than or equal to the minimum continuous length of the protection zone, and the continuous length of the reference change is greater than or equal to the minimum continuous length of the protection zone is determined as the candidate protection zone, and the candidate protection zone is used as the protection zone. When the initial protection zone is determined, the initial protection zone is used as the protection zone.

[0160] S204. Based on scale data, adaptive variational mode decomposition parameters are selected for shipborne gravity data to obtain survey line modal components. Based on scale data, regional modal extraction is performed on the survey line modal components to obtain regional modal components.

[0161] In practice, the range of candidate modalities and the range of candidate penalty factors are read from the scale data. The number of candidate modalities is denoted as... Candidate penalty factors are denoted as For each candidate mode number and each candidate penalty factor within the candidate range, candidate variational mode decomposition is performed on the shipboard gravity data to obtain multiple candidate modal components.

[0162] For a set of candidate mode numbers and candidate penalty factors, the corresponding candidate mode components are denoted as follows: , The residual energy is calculated using the following formula:

[0163] ;

[0164] In the formula, The residual energy corresponds to the number of candidate modes and the candidate penalty factor. For the first The shipboard gravity observation values ​​corresponding to each measuring point location For the first The candidate modal component in the th... Modal reconstruction values ​​at each measurement point location This represents the number of measuring points within the same measuring line.

[0165] The average of the absolute values ​​of the differences between the reconstructed value obtained by superimposing multiple candidate modal components and the shipborne gravity observation value is determined as the modal reconstruction error. When the number of candidate modes is greater than one, the candidate modal components are sorted from low to high center frequency, and the minimum value of the center frequency difference between adjacent candidate modal components after sorting is determined as the modal center frequency interval. When the number of candidate modes is equal to one, the modal center frequency interval is not used as a screening condition for the candidate decomposition results; instead, modal reconstruction error and residual energy are judged for the candidate decomposition results.

[0166] The overall energy of shipborne gravity observations is the sum of squared differences between multiple shipborne gravity observations and their average value in the shipborne gravity data. The ratio between the local residual scale and the survey line analysis scale is determined as the allowable residual proportion; when this ratio is greater than one, the allowable residual proportion is set to one. The residual energy threshold is determined by multiplying the overall energy of the shipborne gravity observations by the allowable residual proportion. The center frequency interval threshold is determined based on the uniform sampling interval and the range of candidate modes, and the reconstruction error threshold is determined based on the fluctuation range between adjacent measurement points.

[0167] When the number of candidate modes is greater than one, and the modal center frequency interval corresponding to the number of candidate modes and the candidate penalty factor is greater than or equal to the center frequency interval threshold, the modal reconstruction error is less than or equal to the reconstruction error threshold, and the residual energy is less than or equal to the residual energy threshold, the number of candidate modes and the candidate penalty factor are used as optional decomposition parameters. When the number of candidate modes is equal to one, and the modal reconstruction error corresponding to the number of candidate modes and the candidate penalty factor is less than or equal to the reconstruction error threshold, and the residual energy is less than or equal to the residual energy threshold, the number of candidate modes and the candidate penalty factor are used as optional decomposition parameters. When multiple sets of optional decomposition parameters exist, the set of optional decomposition parameters with the smallest modal reconstruction error is determined as the adaptive variational mode decomposition parameters.

[0168] When no alternative decomposition parameters exist, the candidate decomposition result with the smallest residual energy is selected first. If there are multiple candidate decomposition results with the smallest residual energy, the candidate decomposition result with the largest modal center frequency interval is selected from among them. If none of the candidate decomposition results being compared have a modal center frequency interval, the candidate decomposition result with the smallest modal reconstruction error is selected from among the multiple candidate decomposition results with the smallest residual energy. The number of candidate modes and the candidate penalty factor corresponding to the selected candidate decomposition results are determined as adaptive variational mode decomposition parameters.

[0169] Adaptive variational mode decomposition (ADMD) parameters were used to perform ADMD on the shipborne gravity data to obtain the survey line modal components. The center scale was determined based on the center frequency and uniform sampling interval of the survey line modal components. The center scale represents the length scale along the survey line corresponding to the survey line modal component; with the same uniform sampling interval, the lower the center frequency, the larger the center scale. Survey line modal components with a center scale greater than or equal to the regional modal screening scale were identified as regional modal components.

[0170] S205, based on the protected area, the regional modal components are reconstructed in segments to obtain regional background data, and based on the regional background data, local residuals are extracted from the shipborne gravity data in the protected area to obtain local residual data.

[0171] In practice, the regional background reconstruction value formed by the superposition of regional modal components is retained within the protected area. For the non-protected section between adjacent protected areas, the regional background reconstruction values ​​at the boundaries of the adjacent protected areas at both ends of the non-protected section are used as constraints. Within the non-protected section... The regional background reconstruction value for each measuring point location is calculated using the following formula:

[0172] ;

[0173] In the formula, For the first The reconstructed background value of the region at each measuring point location The distance along the survey line is the boundary of the adjacent protected area on the starting side of the non-protected section. The distance along the survey line corresponding to the boundary of the adjacent protected area on the end side of the non-protected section. This represents the reconstructed regional background value at the boundary of the adjacent protected area on the starting side of the non-protected section. This is the reconstructed regional background value at the boundary of the adjacent protected area on the end side of the non-protected section.

[0174] For the unprotected sections at the beginning or end of the survey line, the reconstructed regional background value at the boundary of the adjacent protected area is used as the end constraint value. The difference between the end constraint value and the reconstructed regional background value formed by superimposing the regional modal components at the endpoint of the unprotected section near the boundary of the adjacent protected area is determined as the translation amount. The reconstructed regional background value formed by superimposing the regional modal components within the unprotected section is added to the translation amount to obtain the reconstructed regional background value for each measuring point within the unprotected section. The regional background data consists of the measuring point locations and the reconstructed regional background values.

[0175] Within the protected area, the difference between the shipborne gravity observation value at each measuring point and the corresponding regional background reconstruction value is defined as the local residual value. Local residual data is composed of the local residual values ​​corresponding to multiple measuring points within the protected area.

[0176] S206. Adaptive variational mode decomposition is performed on the local residual data to obtain local modal components. The line modal components and local modal components are then combined to obtain modal data.

[0177] In practice, the range of local candidate modes and the range of local penalty factor candidates are determined based on the local residual scale in the scale data. For each protected area, the set of numbers corresponding to the measurement points within the protected area is denoted as follows: Local candidate variational mode decomposition was performed on the local residual data within each protected area to obtain the local candidate decomposition results.

[0178] The calculation methods for local reconstruction error, local residual energy, and local center frequency interval are consistent with those in S204. The calculation object is replaced by local residual data within each protected area, instead of shipborne gravity data. The local residual energy threshold is determined by multiplying the overall energy of the local residual data within each protected area by the allowable residual proportion, which is consistent with the allowable residual proportion in S204. The local reconstruction error threshold is determined based on the fluctuation range of adjacent local residual values ​​within each protected area. The local center frequency interval threshold is determined based on the uniform sampling interval and the range of the number of local candidate modes.

[0179] When the number of local candidate modes is greater than one, and the local center frequency interval corresponding to the number of local candidate modes and the local candidate penalty factor is greater than or equal to the local center frequency interval threshold, the local reconstruction error is less than or equal to the local reconstruction error threshold, and the local residual energy is less than or equal to the local residual energy threshold, the number of local candidate modes and the local candidate penalty factor are used as locally selectable decomposition parameters. When the number of local candidate modes is equal to one, and the local reconstruction error corresponding to the number of local candidate modes and the local candidate penalty factor is less than or equal to the local reconstruction error threshold, and the local residual energy is less than or equal to the local residual energy threshold, the number of local candidate modes and the local candidate penalty factor are used as locally selectable decomposition parameters. When multiple sets of locally selectable decomposition parameters exist, the set of locally selectable decomposition parameters with the smallest local reconstruction error is determined as the local adaptive variational mode decomposition parameters.

[0180] When no locally selectable decomposition parameters exist, the local candidate decomposition result with the smallest local residual energy is selected from the local candidate decomposition results. When there are multiple local candidate decomposition results with the smallest local residual energy, the local candidate decomposition result with the largest local center frequency interval is selected from these multiple local candidate decomposition results with the smallest local residual energy. When none of the local candidate decomposition results participating in the comparison have a local center frequency interval, the local candidate decomposition result with the smallest local reconstruction error is selected from the multiple local candidate decomposition results with the smallest local residual energy. The number of local candidate modes and the local candidate penalty factor corresponding to the selected local candidate decomposition result are determined as the local adaptive variational mode decomposition parameters.

[0181] Adaptive variational mode decomposition (ADMD) parameters were used to perform ADMD on the local residual data within each protected area to obtain local modal components. A positional correspondence between the survey line modal components and the local modal components was established according to their location along the survey line. The modal source identifiers, center scales, modal energies, and modal reconstruction values ​​of both the survey line and local modal components were retained to obtain modal data. The modal source identifiers include both the survey line modal source identifier and the local modal source identifier.

[0182] S207, based on shipborne change data, baseline change data, protected area and modal data, modal data is processed for modal attribution to obtain attribution data.

[0183] In practice, the changes of each modal component in the modal data are extracted to obtain the modal change direction, modal change amplitude, and modal change continuity length. The calculation method for the change difference between the modal change amplitude and the reference change amplitude is the same as the calculation method for the change difference value in S203, but the calculation object is replaced by the modal change amplitude instead of the shipborne change amplitude.

[0184] When the central scale of the survey line modal component is greater than or equal to the regional modal screening scale, and the continuous length of the survey line modal component within the protected area whose modal change direction is consistent with the baseline change direction is greater than or equal to the minimum continuous length of the protected area, the survey line modal component is assigned to the regional background.

[0185] When a local modal component is located within a protected area, and the continuous length of the local modal component whose modal change direction is consistent with the shipborne change direction is greater than or equal to the minimum continuous length of the protected area, the continuous length of the local modal component whose modal change direction is consistent with the reference change direction is greater than or equal to the minimum continuous length of the protected area, and the center scale of the local modal component is greater than or equal to the local residual scale, the local modal component is assigned to the local signal.

[0186] When the central scale of a modal component is smaller than the local residual scale, and the switching ratio of the modal component between adjacent measuring points in the ascending and descending directions is greater than or equal to the alternation ratio threshold, the modal component is classified as a random error. The switching ratio between the ascending and descending directions is the ratio between the number of times the switching occurs between the ascending and descending directions and the number of adjacent measuring points; the number of switching occurrences between the ascending and descending directions is not counted for stationary directions. The alternation ratio threshold is the sum of the average and standard deviation of the switching ratios of the ascending and descending directions for all modal components within the same measuring line.

[0187] The locations of course change, speed change, and measurement interruption are determined as described in Example 1. The vicinity of any one of these locations is defined as the area along the survey line centered on that location, with half the local residual scale as its length. The area near the boundary of the protected area is defined as the area along the survey line centered on the boundary of the protected area, with half the local residual scale as its length. The modal energy percentage is calculated using the following formula:

[0188] ;

[0189] In the formula, For the first The proportion of modal energy of each modal component within the target range. For the first The modal component in the ... Modal reconstruction values ​​at each measurement point location The set of measurement point location numbers within the target area. For the first The set of measurement point location numbers within the modal location range of each modal component. When the target range is the vicinity of any of the following locations: heading change location, speed change location, or measurement interruption location, the modal energy percentage is used for navigation error attribution processing; when the target range is near the boundary of the protected area, the modal energy percentage is used for transition error attribution processing. The target range is not output as independent data; it is used to calculate the modal energy percentage in the modal attribution processing.

[0190] The energy concentration ratio threshold is the ratio of the length of the adjacent range to the length of the modal location range. The boundary energy ratio threshold is the ratio of the length along the survey line near the boundary of the protected area to the length of the modal location range. The boundary continuity threshold is the allowable deviation of the variation amplitude.

[0191] When the modal energy proportion of the adjacent range of any of the following locations—heading change location, speed change location, and measurement interruption location—is greater than or equal to the energy concentration ratio threshold, and the modal change direction is inconsistent with the reference change direction within the same range along the survey line, the modal component is assigned to the navigation error.

[0192] When the modal energy proportion of the adjacent range of any of the following locations—heading change location, speed change location, and measurement interruption location—is greater than or equal to the energy concentration ratio threshold, and the difference between the modal change amplitude and the baseline change amplitude within the same measurement line location range is greater than the protection difference threshold, the modal component is assigned to the navigation error.

[0193] When the modal energy proportion corresponding to the area near the boundary of the protected area is greater than or equal to the boundary energy proportion threshold, and the absolute value of the difference in modal reconstruction values ​​between adjacent measuring points on both sides of the boundary of the protected area is greater than the boundary continuity threshold, the modal component is assigned as the transition error.

[0194] When the same local modal component simultaneously satisfies the judgment conditions for two or more of the following modal assignment results: local signal assignment, navigation error assignment, transition error assignment, and random error assignment, the modal assignment result of the same local modal component shall be determined in the order of local signal assignment, navigation error assignment, transition error assignment, and random error assignment. When the same survey line modal component simultaneously satisfies the judgment conditions for two or more of the following modal assignment results: regional background assignment, navigation error assignment, transition error assignment, and random error assignment, the modal assignment result of the same survey line modal component shall be determined in the order of regional background assignment, navigation error assignment, transition error assignment, and random error assignment.

[0195] When the same local modal component does not meet the judgment conditions for local signal assignment, navigation error assignment, transition error assignment, and random error assignment, if the central scale of the same local modal component is smaller than the local residual scale, then the same local modal component is assigned to random error; if the central scale of the same local modal component is greater than or equal to the local residual scale, then the same local modal component is assigned to transition error.

[0196] When the same survey line mode component does not meet the judgment conditions for regional background assignment, navigation error assignment, transition error assignment, and random error assignment, if the central scale of the same survey line mode component is smaller than the local residual scale, then the same survey line mode component is assigned to random error assignment; if the central scale of the same survey line mode component is greater than or equal to the local residual scale, then the same survey line mode component is assigned to transition error assignment.

[0197] The attribution data consists of mode number, mode source identifier, mode location range, mode center scale, mode energy, and mode attribution result.

[0198] S208, based on the attribution data and modal data, performs hierarchical reconstruction processing to obtain ocean gravity data after error processing.

[0199] In practice, modal components belonging to the regional background are extracted from the modal data, and these modal components are superimposed to obtain the regional background layer data. Local modal components belonging to local signals are extracted from the modal data, and these local modal components are superimposed according to the corresponding protected area to obtain the local signal layer data.

[0200] For measuring points other than those at or near the boundary of the protected area, the superimposed value of the regional background layer data and the local signal layer data is determined as the error-processed gravity value for the corresponding measuring point. For measuring points at or near the boundary of the protected area, the regional background layer data and the local signal layer data of the adjacent measuring points in front of the boundary are superimposed to obtain the pre-boundary reconstruction value; the regional background layer data and the local signal layer data of the adjacent measuring points behind the boundary are superimposed to obtain the post-boundary reconstruction value; a weighted calculation is performed based on the pre-boundary reconstruction value and the post-boundary reconstruction value to obtain the transition reconstruction value, which is then determined as the error-processed gravity value for the corresponding measuring point.

[0201] Modal components assigned to random error, navigation error, and transition error are not included in the superposition and reconstruction as regional background layer data and local signal layer data. Error-processed ocean gravity data consists of the measurement point locations and the error-processed gravity values.

[0202] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for processing ocean gravity data errors based on adaptive variational mode decomposition, characterized in that, include: Obtain shipboard gravity data and corresponding baseline data, perform positional correspondence and scale analysis on the shipboard gravity data and baseline data to obtain scale data; Based on scale data, continuous changes in shipborne gravity data and baseline data are extracted to obtain shipborne change data and baseline change data. The differences in changes between shipborne change data and baseline change data are then identified to obtain the protected area. Based on scale data, adaptive variational mode decomposition is performed on shipborne gravity data to obtain survey line mode components. Based on scale data, regional mode extraction is performed on the survey line mode components to obtain regional mode components. Based on the protected area, the regional mode components are reconstructed into segments to obtain regional background data. Based on regional background data, local residuals are extracted from shipborne gravity data within the protected area to obtain local residual data. Adaptive variational mode decomposition is then performed on the local residual data to obtain local modal components. Finally, the survey line modal components and local modal components are combined to obtain modal data. Based on shipborne variation data, baseline variation data, protected areas, and modal data, modal data is processed for modal attribution to obtain attribution data; Based on the attribution data and modal data, hierarchical reconstruction processing is performed to obtain ocean gravity data after error processing; The method for obtaining scale data by performing positional correspondence and scale analysis on shipborne gravity data and reference data includes: The positions of the measuring points in the shipborne gravity data and the reference positions in the reference data are processed to obtain position correspondence data. Based on the location-corresponding data, the shipborne gravity data and the reference data are extracted using a unified sampling interval, reference resolution scale, survey line analysis scale, and local residual scale. The regional background scale is calculated based on the baseline resolution scale and the survey line analysis scale to obtain the regional background scale. Based on the unified sampling interval, survey line analysis scale, regional background scale, and local residual scale, the protection zone determination window length, minimum continuous length of the protection zone, candidate mode number range, penalty factor candidate range, and regional mode screening scale are generated to obtain scale data; The method for identifying the differences between shipborne change data and baseline change data to obtain the protected area includes: Extract the direction of shipborne change, the amplitude of shipborne change, and the continuous length of shipborne change from the shipborne change data; extract the reference change direction, the reference change amplitude, and the reference change continuous length from the reference change data. A consistency comparison is performed between the ship's load change direction and the reference change direction to obtain the direction comparison result; The difference between the shipborne variation amplitude and the baseline variation amplitude is calculated to obtain the variation difference value; The average and standard deviation of multiple variation values ​​within the same survey line are statistically analyzed to obtain the protection difference threshold. Based on the direction comparison results, change difference value, protection difference threshold, shipborne change continuous length, benchmark change continuous length and minimum continuous length of the protection zone, the same range of the same survey line location corresponding to the shipborne change data and benchmark change data is filtered to obtain the protection zone. The method for obtaining survey line modal components by performing adaptive variational mode decomposition on shipborne gravity data based on scale data includes: Extract the range of candidate modal numbers and the range of candidate penalty factors from the scale data; Based on the range of candidate modes and the range of candidate penalty factors, candidate variational mode decomposition is performed on the shipborne gravity data to obtain the candidate decomposition results; The candidate decomposition results are evaluated for modal center frequency interval, modal reconstruction error, and residual energy to obtain adaptive variational mode decomposition parameters. Adaptive variational mode decomposition is performed on the shipborne gravity data according to the adaptive variational mode decomposition parameters to obtain the survey line mode components; The method for performing modal attribution processing on modal data to obtain attribution data includes: Modal variation data is obtained by extracting the modal variation data from the transect modal components and local modal components in the modal data. Based on the protected area, the modal change data corresponding to the modal components of the survey line and the baseline change data are continuously compared to obtain the regional background attribution results. Based on the protected area, the modal change data, shipborne change data and baseline change data corresponding to the local modal components are continuously compared to obtain the local signal attribution results. Based on modal variation data, protected area and shipborne gravity data, error attribution is determined for modal data, and error attribution results are obtained; Prioritize the regional background attribution results, local signal attribution results, and error attribution results to obtain the attribution data.

2. The ocean gravity data error processing method based on adaptive variational mode decomposition according to claim 1, characterized in that, The method for calculating the difference between the shipborne variation amplitude and the reference variation amplitude to obtain the variation difference value includes: Extract the ship-borne variation amplitude and the reference variation amplitude within the same location range along the survey line; The absolute value of the difference between the shipborne variation amplitude and the reference variation amplitude is calculated to obtain the variation amplitude difference. A uniform sampling interval is extracted from the scale data, and the average absolute value of the difference between adjacent shipborne gravity observations and adjacent reference gravity values ​​in the corresponding reference data is calculated. The larger of the two average values ​​is taken as the allowable deviation of the variation amplitude. Using the allowable deviation of the change amplitude as the normalization scale, the difference in change amplitude is normalized to obtain the change difference value.

3. The ocean gravity data error processing method based on adaptive variational mode decomposition according to claim 1, characterized in that, The method for obtaining regional background data by segment reconstruction of regional modal components based on protected areas includes: Within the protected area, the regional modal components are superimposed and reconstructed to obtain the regional background reconstruction value within the protected area; Based on the regional background reconstruction value at the boundary of adjacent protected areas, the non-protected sections between adjacent protected areas are reconstructed proportionally along the survey line. Based on the regional background reconstruction values ​​at the boundaries of adjacent protected areas, the end constraint values ​​and translation amounts of the end unprotected sections are determined, and the end unprotected sections located at the beginning or end of the survey line are translated and reconstructed to obtain the regional background data.

4. The ocean gravity data error processing method based on adaptive variational mode decomposition according to claim 1, characterized in that, The method for performing adaptive variational mode decomposition on local residual data to obtain local mode components includes: Based on the local residual scale in the scale data, configure the range of local candidate modes and the range of local penalty factor candidates for the local residual data. Based on the range of local candidate modes and the range of local penalty factor candidates, local candidate variational mode decomposition is performed on the local residual data to obtain the local candidate decomposition results; Local candidate decomposition results are filtered by local center frequency interval, local reconstruction error and local residual energy to obtain local adaptive variational mode decomposition parameters; Adaptive variational mode decomposition is performed on the local residual data according to the local adaptive variational mode decomposition parameters to obtain local modal components.

5. The ocean gravity data error processing method based on adaptive variational mode decomposition according to claim 1, characterized in that, The method for determining the error attribution of modal data based on modal variation data, protected area data, and shipborne gravity data, and obtaining the error attribution result, includes: Based on the modal change data, random error assignment is determined for modal components whose central scale is smaller than the local residual scale and whose switching ratio between the rising and falling directions satisfies the alternating change condition, and the random error assignment results are obtained. Based on the acquisition time and measurement point location in the shipborne gravity data, the locations of course changes, speed changes, and measurement interruptions are identified. The navigation error attribution is determined for modal components with concentrated energy in the vicinity that are inconsistent with the baseline change data, and the navigation error attribution result is obtained. Based on the boundary of the protected area, the transition error attribution is determined for the modal components with concentrated energy near the boundary and discontinuous modal reconstruction values ​​on both sides of the boundary, and the transition error attribution result is obtained. The results of random error assignment, navigation error assignment, and transition error assignment are summarized to obtain the error assignment results.

6. The ocean gravity data error processing method based on adaptive variational mode decomposition according to claim 1, characterized in that, The method for obtaining error-processed ocean gravity data by performing hierarchical reconstruction based on attribution data and modal data includes: Based on the attribution data, the survey line modal components attribution to the regional background are extracted from the modal data and superimposed to obtain the regional background layer data. Based on the attribution data, the local modal components of the local signal attribution are extracted from the modal data and superimposed according to the protected area to obtain the local signal layer data; Modal components for random error assignment, navigation error assignment, and transient error assignment are reconstructed and excluded. The regional background layer data and local signal layer data are overlaid, and the location of the protected area boundary and the location of the measuring points near the protected area boundary are reconstructed to obtain the ocean gravity data after error processing.

Images