A method for early warning of ship anchoring instability by integrating multi-source heterogeneous data

Abstract

This invention relates to the field of ship monitoring technology, specifically to a ship anchoring instability early warning method that integrates multi-source heterogeneous data. The method includes the following steps: obtaining the static holding force extreme value based on the seabed state parameters in the nautical chart seabed data and the ship length and displacement scalars in the Automatic Identification System (AIS); and generating comprehensive environmental thrust parameters based on the wind speed and current vectors in the ship's environmental monitoring instrument. This invention calculates the lateral sway angle using heading parameters, speed parameters, and the initial anchoring reference bearing, decomposes the speed parameters to obtain lateral velocity parameters, and combines this with the displacement scalar to obtain lateral kinetic energy parameters. This ensures that the final instability early warning result is simultaneously subject to multiple constraints, including excessive force, spatial dispersion, sway direction, and kinetic energy, thereby improving the lead time and stability of anchoring instability identification.

Description

Technical Field

[0001] This invention relates to the field of ship monitoring technology, and in particular to a method for early warning of ship anchoring instability by integrating multi-source heterogeneous data. Background Technology

[0002] Ship monitoring technology mainly involves the continuous perception, data collection, status identification, and risk assessment of a ship's navigation status, anchoring status, operational status, and surrounding environment.

[0003] Existing ship monitoring technologies typically focus on continuous sensing, data collection, status identification, and risk assessment related to the ship's navigation, anchoring, operational, and surrounding environmental conditions. In practice, they often rely on information such as position, speed, heading, trajectory deviation, area boundaries, and manual thresholds for status classification. Risk assessment tends to focus on surface behavioral changes, making it difficult to uniformly map the impact of seabed sediment on anchor holding power, the ship's displacement on stress response, and the influence of wind and current on hull traction to a single risk criterion. In anchoring scenarios, short-term ship displacement may be normal yaw caused by tidal currents or a precursor to anchor dragging when the anchor chain tension exceeds the seabed holding power. Judging solely based on whether the positioning coordinates exceed empirical ranges can easily misjudge normal turning as a risk or miss abnormal stresses before the ship has made a significant deviation. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a ship anchoring instability early warning method that integrates multi-source heterogeneous data.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a ship anchoring instability early warning method that integrates multi-source heterogeneous data, comprising the following steps:

[0006] The static gripping force extreme value is obtained based on the seabed state parameters in the nautical chart seabed data and the ship length and displacement scalars in the Automatic Identification System (AIS); the comprehensive environmental thrust parameters are generated based on the wind speed and current velocity vectors in the ship's environmental monitoring instrument.

[0007] Based on the ship's anchoring position coordinates and the positioning coordinates within the ship's automatic identification system, combined with the comprehensive environmental thrust parameters, the anchor chain transmitted tension scalar is calculated; the positioning coordinates corresponding to when the anchor chain transmitted tension scalar is greater than the static holding force extreme value are selected to generate an over-limit force judgment vector;

[0008] Based on the over-limit force determination vector, extract the coordinates of the corresponding point set within a specified time range, calculate the discrete matrix of the point set coordinates in the two-dimensional coordinate system, and solve the eigenvalues ​​and eigenvectors corresponding to the discrete matrix; based on the eigenvectors, extract the principal axis parameter and secondary axis parameter of the sway, and construct the sway envelope confidence ellipse.

[0009] Based on the heading and speed parameters in the Automatic Identification System (AIS), the lateral yaw angle of the ship is calculated. The lateral yaw angle is used to orthogonally decompose the speed parameters to obtain the lateral velocity parameters. The lateral kinetic energy parameters are obtained by combining the displacement scalar with the lateral velocity parameters. Based on the yaw envelope confidence ellipse, an instability warning result is generated.

[0010] Preferably, the steps for obtaining the integrated environmental thrust parameters are as follows:

[0011] The bottom sediment state parameters in the nautical chart bottom sediment data are called up, and the ship length scalar and displacement scalar in the automatic identification system are read. The displacement scalar is obtained by multiplying the draft parameter, ship beam parameter, ship length scalar and preset ship block coefficient by volume and converting it with water density. The bottom sediment state parameters and ship length scalar are matched according to the row and column index of the bottom sediment coefficient data table, and the bottom sediment coefficient recorded at the matching position is extracted. The bottom sediment coefficient is multiplied with the displacement scalar to obtain the static gripping force extreme value.

[0012] The wind speed vector and flow velocity vector in the ship's environmental monitoring instrument are called up, the direction angle and magnitude components of the wind speed vector are split, the air density parameter and the surface wind-receiving area parameter are called up to calculate the load component under the action of wind direction, the direction angle and magnitude components of the flow velocity vector are split up, the seawater density parameter and the underwater flow-receiving area parameter are called up to calculate the load component under the action of flow direction, and the wind load vector and water flow load vector are generated.

[0013] The lateral and longitudinal components of the wind load vector in the two-dimensional coordinate system are extracted, and the lateral and longitudinal components of the water flow load vector in the two-dimensional coordinate system are extracted. The components in the same coordinate axis direction are synthesized, and the direction angle and magnitude components of the synthesized components are retained to generate comprehensive environmental thrust parameters.

[0014] Preferably, the step of obtaining the anchor chain transmitted tension scalar is as follows:

[0015] The ship's anchoring position coordinates are called, and the positioning coordinates in the ship's automatic identification system are called. The ship's anchoring position coordinates are used as the anchor chain fixed end position, and the positioning coordinates are used as the hull towing end position. The lateral coordinate difference, longitudinal coordinate difference, and spatial interval change between the ship's anchoring position coordinates and the positioning coordinates are extracted point by point. The lowest point of suspension, the bottoming point, and the towing end connection point are marked according to the positional relationship between the two ends of the anchor chain to obtain the anchor chain suspension shape.

[0016] By calling the positioning coordinates and the comprehensive environmental thrust parameters, the positioning coordinates are mapped to the traction end connection point in the anchor chain suspension configuration. The comprehensive environmental thrust parameters are decomposed into components acting along the anchor chain traction direction. Combining the directional change relationship from the bottoming point to the traction end connection point, the horizontal force at the bottoming point is calculated point by point to obtain the anchor chain transmitted tension scalar.

[0017] Preferably, the step of obtaining the over-limit force determination vector is as follows:

[0018] The static gripping force extreme value is invoked, and the anchor chain transmitted tension scalar corresponding to each positioning coordinate is compared with the static gripping force extreme value one by one. Positioning coordinates with anchor chain transmitted tension scalars greater than the static gripping force extreme value are retained, and positioning coordinates with anchor chain transmitted tension scalars less than or equal to the static gripping force extreme value are removed. The excess coordinate set is formed according to the order of the retained positioning coordinates. Then, each positioning coordinate in the excess coordinate set is converted into the excess mark value of the corresponding position to generate the excess force judgment vector.

[0019] Preferably, the steps for obtaining the coordinates of the point set are as follows:

[0020] Read the over-limit marker values ​​corresponding to each position in the over-limit force determination vector, filter the positioning coordinates corresponding to the over-limit marker values ​​according to the specified time range, remove the positioning coordinates outside the specified time range, and arrange the filtered positioning coordinates according to the acquisition time sequence to obtain the point set coordinates.

[0021] Preferably, the step of obtaining the eccentricity envelope confidence ellipse is as follows:

[0022] Extract the horizontal and vertical coordinate values ​​of the point set in the two-dimensional coordinate system respectively, calculate the discrete offset of the horizontal coordinate value, the discrete offset of the vertical coordinate value, and the combined offset of the horizontal and vertical coordinates, and fill the matrix position according to the order of the two-dimensional coordinate axes to obtain the discrete matrix;

[0023] Solve for the eigenvalues ​​and eigenvectors corresponding to the discrete matrix, sort the eigenvalues ​​by numerical value, select the eigenvector corresponding to the largest eigenvalue as the principal axis parameter of the oscillation, select the eigenvector corresponding to the smallest eigenvalue as the secondary axis parameter of the oscillation, construct the boundary line enclosing the coordinates of the point set based on the principal axis parameter of the oscillation, the secondary axis parameter of the oscillation, and the outer edge distribution position of the point set coordinates, and generate the oscillation envelope confidence ellipse.

[0024] Preferably, the step of obtaining the lateral kinetic energy parameter is as follows:

[0025] Call the heading and speed parameters in the Automatic Identification System (AIS) to extract the heading angle corresponding to the heading parameter, read the anchor reference angle corresponding to the initial anchor reference bearing, convert the heading angle relative to the anchor reference angle by angle difference, mark the lateral yaw direction of the ship according to the positive and negative directions of the angle difference, record the degree of lateral yaw of the ship according to the absolute magnitude of the angle difference, and obtain the lateral yaw angle.

[0026] The speed parameters are called and decomposed into velocity components along the anchor reference direction and velocity components perpendicular to the anchor reference direction according to the lateral sway angle. The velocity component perpendicular to the anchor reference direction is extracted as the lateral velocity parameter. The lateral velocity parameter is squared and the displacement scalar is called to perform kinetic energy conversion on the squared lateral velocity parameter to obtain the lateral kinetic energy parameter.

[0027] Preferably, the steps for obtaining the instability early warning result are as follows:

[0028] The secondary axis length parameter and center coordinates within the oscillation envelope confidence ellipse are called, and the difference in secondary axis length parameter and the offset of center coordinates at adjacent time points are extracted according to the acquisition time sequence.

[0029] The secondary axis change rate is calculated based on the time interval between adjacent moments, and the center coordinate offset is projected onto the direction corresponding to the principal axis parameter of the yaw to obtain the center slip rate.

[0030] The secondary axis rate of change is compared item by item with a preset rate of change threshold. The lateral kinetic energy parameter corresponding to the secondary axis rate of change being greater than the preset rate of change threshold is retained. The lateral kinetic energy parameter is then compared with a preset lateral kinetic energy tolerance threshold.

[0031] Simultaneously, the center slip rate is compared item by item with the preset slip threshold;

[0032] Output is generated when the lateral kinetic energy parameter is greater than the preset lateral kinetic energy tolerance threshold, or when the center slip rate is greater than the slip threshold, to generate an instability warning result.

[0033] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0034] In this invention, the static holding force extreme value is obtained based on the seabed state parameters, ship length scalar, and displacement scalar from the seabed surface data on the nautical chart. Then, a comprehensive environmental thrust parameter is generated using wind speed vector, current velocity vector, air density parameter, surface windward area parameter, seawater density parameter, and underwater current-receiving area parameter. This shifts the assessment of anchoring safety from a single positional state identification to a joint measurement of seabed holding force constraints, hull size constraints, and environmental load constraints. Furthermore, the anchor chain transmitted tension scalar is calculated using the ship's anchoring position coordinates, positioning coordinates, and comprehensive environmental thrust parameter. The anchor chain transmitted tension scalar is then compared point-by-point with the static holding force extreme value. This allows the determination of whether the ship has a tendency to drag anchor to be converted into a continuous screening result of force exceeding limits, reducing reliance on ship position deviations alone. False alarms generated during risk assessment based on distance are addressed. Further, by extracting the coordinates of a point set within a specified time range from the over-limit force determination vector, calculating the discrete matrix, and obtaining eigenvalues ​​and eigenvectors, the principal axis parameter, secondary axis parameter, and confidence ellipse of the sway can be identified from the distribution of over-limit points, enabling risk assessment to describe the spatial diffusion direction and sway range. Then, by calculating the lateral sway angle using heading parameters, speed parameters, and the initial anchorage reference bearing, the speed parameters are decomposed to obtain lateral velocity parameters, and combined with the displacement scalar to obtain lateral kinetic energy parameters. This ensures that the final instability warning result is simultaneously constrained by multiple factors including over-limit force, spatial discretization, sway direction, and kinetic energy, thereby improving the lead time and stability of anchorage instability identification. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the steps of the present invention;

[0036] Figure 2 This is a graph showing the relationship between the scalar value of the anchor chain transmitted tension and the extreme value of the static holding force.

[0037] Figure 3 The point set coordinates and the confidence ellipse diagram of the oscillation envelope are given. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0039] Please see Figure 1-3 This invention provides a technical solution for a ship anchoring instability early warning method that integrates multi-source heterogeneous data, comprising the following steps:

[0040] The static gripping force extreme value is obtained based on the seabed state parameters in the nautical chart seabed data and the ship length and displacement scalars in the Automatic Identification System (AIS); the comprehensive environmental thrust parameters are generated based on the wind speed and current velocity vectors in the ship's environmental monitoring instrument.

[0041] Based on the ship's anchoring position coordinates and the positioning coordinates in the ship's automatic identification system, combined with the comprehensive environmental thrust parameters, the anchor chain transmitted tension scalar is calculated; the positioning coordinates corresponding to the anchor chain transmitted tension scalar being greater than the static holding force extreme value are selected to generate an over-limit force judgment vector;

[0042] Based on the over-limit force judgment vector, extract the coordinates of the corresponding point set within the specified time range, calculate the discrete matrix of the point set coordinates in the two-dimensional coordinate system, and solve the eigenvalues ​​and eigenvectors corresponding to the discrete matrix; based on the eigenvectors, extract the principal axis parameter and secondary axis parameter of the sway, and construct the sway envelope confidence ellipse.

[0043] Based on the heading and speed parameters in the Automatic Identification System (AIS), the ship's lateral yaw angle is calculated. The lateral yaw angle is then used to orthogonally decompose the speed parameters to obtain the lateral velocity parameters. Combined with the displacement scalar and the lateral velocity parameters, the lateral kinetic energy parameters are obtained. Finally, based on the yaw envelope confidence ellipse, an instability warning result is generated.

[0044] The steps for obtaining the comprehensive environmental thrust parameters are as follows:

[0045] The bottom sediment state parameters in the nautical chart bottom sediment data are called up, and the ship length scalar and displacement scalar in the automatic identification system are read. The displacement scalar is obtained by multiplying the draft parameters, ship beam parameters, ship length scalar and preset ship block coefficient by volume and converting it with water density. The bottom sediment state parameters and ship length scalar are matched according to the row and column index of the bottom sediment coefficient data table, and the bottom sediment coefficient recorded at the matching position is extracted. The bottom sediment coefficient is multiplied with the displacement scalar to obtain the static gripping force extreme value.

[0046] The wind speed vector and flow velocity vector in the ship's environmental monitoring instrument are called up, the direction angle and magnitude components of the wind speed vector are split, the air density parameter and the surface wind-receiving area parameter are called up to calculate the load component under the action of wind direction, the direction angle and magnitude components of the flow velocity vector are split up, the seawater density parameter and the underwater flow-receiving area parameter are called up to calculate the load component under the action of flow direction, and the wind load vector and water flow load vector are generated.

[0047] The transverse and longitudinal components of the wind load vector and the water flow load vector in the two-dimensional coordinate system are extracted. The components in the same coordinate axis direction are synthesized, and the direction angle and magnitude components of the synthesized components are retained to generate comprehensive environmental thrust parameters.

[0048] Specifically, based on the seabed state parameters within the nautical chart seabed data, a seabed coefficient data table needs to be constructed first. This table is a two-dimensional lookup table, where the row index corresponds to different seabed state parameters, such as "sand," "silt," "clay," or "rock," and the column index corresponds to a preset discretized range of ship length, such as "less than 100 meters," "100 to 150 meters," "150 to 200 meters," and "greater than 200 meters." The values ​​filled in the table are dimensionless seabed coefficients obtained through extensive anchoring experiment data and hydrodynamic simulation analysis. Next, the ship length data obtained from the nautical chart seabed data will be... The current anchorage seabed condition parameters are matched with the row index of the seabed coefficient data table. Simultaneously, the ship length scalar read from the Automatic Identification System (AIS) is compared with the column index of the data table to determine its length range. Through row and column cross-location, a unique seabed coefficient value is extracted. For example, if the seabed is "clay" and the ship length is 165 meters, the "clay" row and the "150 to 200 meters" column are matched, yielding a seabed coefficient of 0.75. Subsequently, this seabed coefficient is calculated with the displacement scalar read from the AIS to obtain the final static holding power extreme value. The calculation follows the following physical relationships: ,in, It is the calculated extreme value of static gripping force, in Newtons (N). It is a dimensionless substrate coefficient obtained from the data table. The displacement measurement is obtained from the Automatic Identification System (AIS) and needs to be converted from tons to kilograms. The local gravitational acceleration is typically taken as a standard value of 9.8 m / s². For example, for a displacement of 60,000 tons, its mass... The displacement is 60,000,000 kg; the displacement scalar is obtained by analyzing the draft and beam parameters obtained from the Automatic Identification System (AIS), and then multiplying the draft, beam, and length scalars with the preset block coefficient and combining them with the water density conversion; the calculated static gripping force is 0.75 multiplied by 60,000,000 kg and then multiplied by 9.8 m / s², resulting in 441,000,000 Newtons.

[0049] Based on the wind speed and flow velocity vectors from the ship's environmental monitoring instrument, the wind load is first calculated. This process follows the standard drag equation in fluid dynamics, specifically by breaking down the wind speed vector into its direction angle and magnitude components. Then, relevant parameters are used to calculate the load components under wind direction. The calculation formula is as follows: ,in, It is the calculated wind load magnitude. This refers to the air density parameter, typically taken as a value under standard atmospheric conditions, approximately 1.225 kg / m³. It is the magnitude component extracted from the wind speed vector, i.e., the wind speed magnitude. This is the wind force coefficient, which is not a constant, but rather the angle between the wind direction and the ship's heading. The relevant functions, whose values ​​are obtained by querying a pre-established database of ship wind tunnel experiments or a database of computational fluid dynamics simulation results, This is the windward surface area parameter, which is the projected area of ​​the ship's above-water portion in the wind direction. It also varies with the included angle. The data, derived from the ship's design database, calculates the wind load magnitude and combines it with the wind direction angle to form a wind load vector. Then, the water flow load is calculated using the same method: separating the velocity vector by its direction angle and magnitude components, and applying the resistance formula. ,in, It is the calculated magnitude of the water flow load. This refers to the seawater density parameter, typically taken as 1025 kg / m³. It is the magnitude component, i.e., the magnitude of the flow velocity, extracted from the velocity vector. It is the hydrodynamic coefficient, the value of which depends on the relative angle between the direction of the current and the heading of the ship. This information was obtained by querying ship tank test databases or hydrodynamic performance simulation databases. This is the underwater current-receiving area parameter, that is, the orthographic projection area of ​​the underwater part of the hull in the direction of the water flow, which also varies with the included angle. After the calculation is completed, the magnitude of the water flow load and the flow direction angle are combined to generate the wind load vector and the water flow load vector.

[0050] Based on the wind load vector and water flow load vector generated in the preceding steps, a fixed two-dimensional geographic coordinate system is first established, typically with the ship's initial anchoring position as the origin. The x-axis is set to geographic east, and the y-axis to geographic north. Next, the wind load vector is decomposed within this coordinate system, analyzing it into a lateral component along the x-axis and a longitudinal component along the y-axis. This decomposition is performed using standard trigonometric functions: the lateral component of the wind load is equal to its magnitude multiplied by the sine of its direction angle (relative to north), and the longitudinal component is equal to its magnitude multiplied by the cosine of its direction angle. Then, the same decomposition operation is performed on the water flow load vector to obtain its lateral and longitudinal components along the x-axis and y-axis. After decomposition, the coaxial components from wind and water flow are... Algebraic summation, specifically, involves adding the lateral components of the wind load and the water flow load to obtain the total lateral environmental thrust component. Similarly, the longitudinal components of the two loads are added to obtain the total longitudinal environmental thrust component. Finally, based on these two combined total components, a single resultant force vector is reconstructed. The magnitude of this vector is calculated using the Pythagorean theorem, which is equal to the square root of the sum of the squares of the total lateral and total longitudinal components. The direction angle of this vector is calculated using a two-parameter arctangent function, with the total longitudinal and lateral components as inputs. This calculation ensures that the final direction angle is within the correct quadrant, thus accurately reflecting the direction of the combined environmental force. The final vector containing the combined magnitude and direction angle is the combined environmental thrust parameter.

[0051] The steps to obtain the anchor chain transmitted tension scalar are as follows:

[0052] The ship's anchoring position coordinates are called, and the positioning coordinates in the ship's automatic identification system are called. The ship's anchoring position coordinates are used as the anchor chain fixed end position, and the positioning coordinates are used as the hull towing end position. The lateral coordinate difference, longitudinal coordinate difference, and spatial interval change between the ship's anchoring position coordinates and the positioning coordinates are extracted point by point. The lowest point of suspension, the bottoming point, and the towing end connection point are marked according to the positional relationship between the two ends of the anchor chain to obtain the anchor chain suspension shape.

[0053] By calling the positioning coordinates and the comprehensive environmental thrust parameters, the positioning coordinates are mapped to the traction end connection point in the anchor chain suspension configuration. The comprehensive environmental thrust parameters are decomposed into components acting along the anchor chain traction direction. Combining the directional change relationship from the bottoming point to the traction end connection point, the horizontal force at the bottoming point is calculated point by point to obtain the scalar of the anchor chain transmitted tension.

[0054] Specifically, based on the ship's anchoring position coordinates and the positioning coordinates within the Automatic Identification System (AIS), the ship's anchoring position coordinates are first set as the fixed end position of the anchor chain on the seabed. Each point in the time-varying sequence of ship positioning coordinates is then designated as the traction connection point at the ship's hull. For each positioning coordinate, the lateral and longitudinal coordinate differences between it and the ship's anchoring position coordinates are calculated. The Pythagorean theorem is then used to calculate the projected distance (spatial interval) between the two coordinates on the horizontal plane. Next, based on the physical characteristics of the anchor chain, a catenary model is used to mathematically describe the actual spatial form of the anchor chain. The input parameters of this model include: the pre-set total length of the anchor chain and the weight per unit length of the anchor chain. Based on the horizontal and vertical distances between the two ends of the anchor chain calculated from the coordinates, the three-dimensional curve shape of the anchor chain in the water can be determined by solving the catenary equation. On this basis, the curve is analyzed, and several key geometric points are marked. Among them, the connection point of the traction end is the current positioning coordinate of the ship, and the lowest point of the suspension is the vertex of the catenary curve, that is, the point where the tangent slope is zero. If the anchor chain is partially laid on the seabed, this point coincides with the bottom contact point, which is the critical point where the anchor chain transitions from a suspended state to contact with the seabed. By iteratively calculating the positioning coordinates at each moment, a series of anchor chain spatial shapes that change over time are obtained. The collection of this series of shapes constitutes the complete anchor chain suspension shape.

[0055] Based on the retrieved positioning coordinates and the integrated environmental thrust parameters, the positioning coordinates at each moment are first mapped onto the anchor chain suspension configuration obtained in the previous step, serving as the location of the traction end connection point. Simultaneously, the integrated environmental thrust parameters acting on these positioning coordinates are obtained. This parameter is a horizontal force vector representing the combined effect of wind and current on the hull. Next, this integrated environmental thrust parameter vector is decomposed. Specifically, a unit direction vector is calculated from the ship's anchoring position coordinates to the current positioning coordinates. This vector represents the horizontal traction direction of the anchor chain. Then, the integrated environmental thrust parameter vector is projected onto this unit direction vector to obtain a vector along the anchor chain traction direction. The action component, a scalar value representing the primary tensile force required to resist environmental forces, is determined by the catenary theory. Under static or quasi-static equilibrium conditions, the horizontal component of the tension at any point on the anchor chain is constant and equal to the horizontal force at the bottoming point. Therefore, the action component along the anchor chain traction direction calculated above is considered as the horizontal tensile force transmitted from the anchor chain to the anchor at the bottoming point. By performing point-by-point calculations on the continuous positioning coordinate sequence provided by the Automatic Identification System (AIS), a scalar sequence of anchor chain transmitted tension corresponding to the time series can be obtained. This sequence reflects the dynamic process of the horizontal holding force borne by the anchor changing with ship sway and environmental forces.

[0056] The steps for obtaining the over-limit force determination vector are as follows:

[0057] Call the static gripping force extreme value, compare the anchor chain transmitted tension scalar corresponding to each positioning coordinate with the static gripping force extreme value, retain the positioning coordinates whose anchor chain transmitted tension scalar is greater than the static gripping force extreme value, and remove the positioning coordinates whose anchor chain transmitted tension scalar is less than or equal to the static gripping force extreme value. Arrange the retained positioning coordinates in the order to form an over-limit coordinate set, and then convert each positioning coordinate in the over-limit coordinate set into the over-limit mark value of the corresponding position to generate an over-limit force judgment vector.

[0058] Specifically, based on the invoked static gripping force extreme value, this value is used as a fixed evaluation benchmark. The anchor chain tension scalar corresponding to each positioning coordinate calculated in the previous step is iteratively compared item by item. Specifically, an initially empty set of out-of-limit coordinates is established, and then the sequence of anchor chain tension scalars containing timestamps is traversed. In each iteration, it is determined whether the current anchor chain tension scalar is greater than the static gripping force extreme value. For example, if the static gripping force extreme value is 450,000 Newtons, and the anchor chain tension scalar calculated at the current moment is 460,000 Newtons, then because 460,000 is greater than 450,000, it is determined to be out-of-limit force. In this case, the positioning coordinate corresponding to that moment is recorded in the out-of-limit coordinate set. Conversely, if the calculated... The calculated anchor chain tension scalar is 430,000 Newtons. Since this is not greater than the static holding force extreme value, no operation is performed. After traversing all data points, the over-limit coordinate set contains the position coordinates of the ship when all anchor holding force over-limit events occurred, arranged in chronological order. Then, a one-dimensional vector of the same length as the original positioning coordinate sequence is created, and all its elements are initialized to non-over-limit marker values, such as the number 0. Finally, the over-limit coordinate set is traversed again. Based on the timestamp information of each coordinate point, the position of the corresponding time index is found in the newly created one-dimensional vector, and the element value at that position is modified to the over-limit marker value, such as the number 1. After all marking is completed, this one-dimensional vector is the final generated over-limit force judgment vector.

[0059] The steps to obtain the coordinates of a point set are as follows:

[0060] Read the over-limit marker values ​​corresponding to each position in the over-limit force judgment vector, filter the positioning coordinates corresponding to the over-limit marker values ​​according to the specified time range, remove the positioning coordinates outside the specified time range, and arrange the filtered positioning coordinates according to the acquisition time sequence to obtain the point set coordinates.

[0061] Specifically, based on the over-limit marker values ​​corresponding to each position in the read over-limit force judgment vector, a specified time range is first set. This range is a dynamically updated time window, set according to the typical cycle of ship anchoring yaw, usually to cover the duration of several complete yaw cycles. For example, based on empirical data and hydrodynamic analysis, this time range is set to 30 minutes before the current analysis time. Next, the over-limit force judgment vector is traversed. The index of this vector corresponds one-to-one with the timestamp of the original positioning coordinates. When an over-limit marker value of a certain position is detected as a preset "1", the timestamp corresponding to that position is extracted, and it is determined whether the timestamp falls within the set 30-minute time window. For example, if the current analysis time is... At 10:30, all positioning coordinates marked as exceeding limits between 10:00 and 10:30 are filtered. If the timestamp meets the range requirement, the corresponding positioning coordinates (usually latitude and longitude pairs) are extracted from the original Automatic Identification System (AIS) data stream based on the timestamp or index, and added to an initially empty set. Exceeding limits markers with timestamps earlier than 10:00 are removed and not included in the calculation. Since the traversal process is performed in chronological order, the positioning coordinates in the new set are naturally arranged according to the acquisition time sequence. After traversing the entire exceeding force determination vector, this ordered set containing the position coordinates of all ships that have experienced exceeding force within the specified time range is the obtained point set coordinates.

[0062] The steps to obtain the skewness envelope confidence ellipse are as follows:

[0063] Extract the horizontal and vertical coordinates of the point set in the two-dimensional coordinate system, calculate the discrete offset of the horizontal coordinates, the discrete offset of the vertical coordinates, and the combined offset of the horizontal and vertical coordinates, and fill the matrix positions according to the order of the two-dimensional coordinate axes to obtain the discrete matrix;

[0064] Solve for the eigenvalues ​​and eigenvectors corresponding to the discrete matrix, sort the eigenvalues ​​by numerical value, select the eigenvector corresponding to the largest eigenvalue as the principal axis parameter of the oscillation, select the eigenvector corresponding to the smallest eigenvalue as the secondary axis parameter of the oscillation, construct the boundary line of the point set coordinates based on the principal axis parameter of the oscillation, the secondary axis parameter of the oscillation, and the distribution position of the outer edge of the point set coordinates, and generate the oscillation envelope confidence ellipse.

[0065] Specifically, based on the coordinates of the point set, this process essentially involves performing principal component analysis on the two-dimensional point set to determine its main distribution direction and dispersion. First, each two-dimensional coordinate point in the point set is decomposed into its horizontal and vertical coordinate values, forming two independent one-dimensional arrays. Then, the average of all horizontal coordinate values ​​and the average of all vertical coordinate values ​​are calculated to determine the geometric center of the point set. Next, based on this geometric center, the covariance matrix of the point set is calculated. This is a 2x2 symmetric matrix, where the diagonal elements are the variances along each coordinate axis, and the off-diagonal elements are the covariances between the two coordinate axes. Specifically, the discrete offset (i.e., variance) of the horizontal coordinate values ​​is the average of the squares of the differences between each horizontal coordinate value and its average value. The discrete offset (i.e., variance) of the vertical coordinate values ​​is calculated similarly. The covariance of the horizontal and vertical coordinates is the average of the products of the horizontal and vertical coordinate deviations of each point. These calculation results are then filled into the matrix positions according to the order of the two-dimensional coordinate axes to obtain the discrete matrix, the specific form of which is: ,in, The resulting discrete matrix, It is the discrete offset (variance) of the horizontal coordinate value. It is the discrete offset (variance) of the vertical coordinate value. and It is the covariance of the horizontal and vertical coordinates, and the two values ​​are equal.

[0066] Based on the discrete matrix obtained from the preceding steps, eigenvalue decomposition is performed on it. This process is accomplished by solving the characteristic equation of the 2x2 matrix, yielding two real eigenvalues. ) and a set of corresponding orthogonal eigenvectors ( Sort the two feature values ​​by their numerical value, for example, for example... Then, select the larger eigenvalue. The corresponding eigenvector As a principal axis parameter of the yaw motion, this vector indicates the direction with the largest variance in the point set coordinate data, i.e., the most dominant direction of the ship's yaw motion. A smaller eigenvalue is selected. The corresponding eigenvector As a secondary axis parameter of the oscillation, this vector indicates the direction perpendicular to the principal axis and with the minimum data variance. Next, based on these parameters and the outer edge distribution of the point set coordinates, a boundary line is constructed to enclose the point set. This boundary line is an ellipse, centered at the geometric center of the previously calculated point set coordinates. The major axis of the ellipse coincides with the direction of the oscillation principal axis parameter, and the minor axis coincides with the direction of the secondary axis parameter. The semi-major and semi-minor axis lengths of the ellipse are proportional to the square root of the eigenvalues. The scaling factor determines the confidence level. For example, to construct a 95% confidence ellipse, this scaling factor is typically taken from the square root of the critical value (approximately 5.991) of a chi-square distribution with 2 degrees of freedom at a 95% confidence level, which is approximately 2.448. Therefore, the semi-major axis length is... The semi-minor axis length is An ellipse is uniquely determined by five parameters: the center of the ellipse, the direction of the principal and secondary axes, and the length of the principal and secondary axes. This ellipse is the final generated eccentric envelope confidence ellipse.

[0067] The steps for obtaining the lateral kinetic energy parameters are as follows:

[0068] Call the heading and speed parameters in the Automatic Identification System (AIS) to extract the heading angle corresponding to the heading parameter, read the anchor reference angle corresponding to the initial anchor reference bearing, convert the heading angle relative to the anchor reference angle by angle difference, mark the lateral yaw direction of the ship according to the positive and negative directions of the angle difference, record the degree of lateral yaw of the ship according to the absolute magnitude of the angle difference, and obtain the lateral yaw angle.

[0069] The speed parameters are called and decomposed into velocity components along the anchor reference direction and velocity components perpendicular to the anchor reference direction according to the lateral sway angle. The velocity component perpendicular to the anchor reference direction is extracted as the lateral velocity parameter. The lateral velocity parameter is squared and the displacement scalar is called to perform kinetic energy conversion on the squared lateral velocity parameter to obtain the lateral kinetic energy parameter.

[0070] Specifically, based on the heading and speed parameters from the Automatic Identification System (AIS), the discrete-time series of the ship's heading angle is first extracted from the heading parameters. Simultaneously, the initial anchoring reference bearing recorded when the ship completes anchoring is read, serving as a fixed anchoring reference angle (e.g., set to 30 degrees). Then, the heading angle at each moment is converted to this anchoring reference angle using an angle difference calculation. The periodicity of the angle needs to be handled during the conversion to ensure the difference is between -180 and +180 degrees. Specifically, the heading angle is subtracted from the anchoring reference angle; if the result is greater than 180 degrees, then the result is considered a reference angle. Subtract 360 degrees from the result. If the result is less than -180 degrees, add 360 degrees to the result. For example, if the current bow angle is 45 degrees, the angle difference is 45-30=15 degrees. If the current bow angle is 350 degrees, the angle difference is 350-30=320 degrees. Since it is greater than 180 degrees, 360 degrees need to be subtracted to get -40 degrees. This converted angle difference is the lateral yaw angle. Its positive or negative sign is used to mark the direction of the ship's lateral yaw. For example, a positive value indicates yaw to the right (or clockwise), and a negative value indicates yaw to the left (or counterclockwise). The magnitude of its absolute value is used to record the degree of the ship's lateral yaw.

[0071] Based on the retrieved speed parameter, the ship's ground speed is first combined with the lateral yaw angle obtained in the previous steps to orthogonally decompose the ship's instantaneous velocity vector. This decomposition is performed in a coordinate system with the anchor reference direction as the reference axis. The speed parameter is decomposed into a velocity component along the anchor reference direction and a velocity component perpendicular to the anchor reference direction. The velocity component perpendicular to the anchor reference direction is obtained by multiplying the magnitude of the speed parameter by the sine of the lateral yaw angle, and this component is extracted as the lateral velocity parameter. Next, the lateral velocity parameter is squared. Then, the ship's displacement scalar is retrieved, and the additional mass effect generated by the ship moving the surrounding water along with it during lateral motion is considered. The kinetic energy conversion of the squared lateral velocity parameter follows the kinetic energy theorem. ,in, These are the calculated lateral kinetic energy parameters. The ship displacement measurement being used must be standardized to kilograms. This refers to the added mass, which is usually estimated as a coefficient of displacement based on the ship's hydrodynamic characteristics. For example, for lateral rolling motion, the added mass coefficient can be taken as 0.6. , The lateral velocity parameter obtained from the aforementioned steps is expressed in meters per second. The result calculated using this formula is the lateral kinetic energy parameter.

[0072] The steps for obtaining the instability warning result are as follows:

[0073] The secondary axis length parameter and center coordinates within the oscillation envelope confidence ellipse are called, and the difference in secondary axis length parameter and the offset of center coordinates at adjacent time points are extracted according to the acquisition time sequence.

[0074] The secondary axis change rate is calculated based on the time interval between adjacent moments, and the center coordinate offset is projected onto the direction corresponding to the principal axis parameter of the yaw to obtain the center slip rate.

[0075] The secondary axis rate of change is compared item by item with the preset rate of change threshold. The lateral kinetic energy parameter corresponding to the secondary axis rate of change being greater than the preset rate of change threshold is retained. The lateral kinetic energy parameter is then compared with the preset lateral kinetic energy tolerance threshold.

[0076] At the same time, the center slip rate is compared with the preset slip threshold item by item;

[0077] Output is generated when the lateral kinetic energy parameter is greater than the preset lateral kinetic energy tolerance threshold, or when the center slip rate is greater than the slip threshold, to produce an instability warning result.

[0078] Specifically, based on the secondary axis length parameter and center coordinates within the invoked sway envelope confidence ellipse, the difference in the secondary axis length parameter between two adjacent moments (e.g., a 1-minute time interval) is first calculated according to the acquisition sequence. This difference is then divided by the time interval to obtain the secondary axis rate of change. Simultaneously, the vector difference in the center coordinates between adjacent moments is calculated to obtain the center coordinate offset. This offset vector is then projected onto the direction indicated by the sway principal axis parameter. The projected length is then divided by the time interval to obtain the center slip rate. Next, the calculated secondary axis rate of change is compared with a preset rate of change threshold. This threshold is obtained through statistical analysis of historical stable anchoring data of similar ship types under similar sea conditions, typically taken as the 95th percentile of the secondary axis rate of change distribution under stable conditions, for example, set to 0.1 m / min. If the secondary axis rate of change exceeds this threshold, then... At this point, the corresponding lateral kinetic energy parameter is compared with a preset lateral kinetic energy tolerance threshold. This tolerance threshold is set based on the ship's safety operation manual and simulation data of anchor dragging accidents, representing the critical kinetic energy level that may cause anchor failure. For example, for a 60,000-ton cargo ship, this threshold can be set to 5 megajoules. At the same time, the calculated center slip rate is compared with a preset slip threshold, which is set according to the recommendations of the International Maritime Organization and the captain's experience. It is usually a value slightly higher than the virtual speed that can be generated by GPS positioning error, for example, set to 0.1 knots (about 0.05 meters / second). Finally, when the combined condition of (secondary axis change rate greater than its threshold and lateral kinetic energy parameter greater than its tolerance threshold) is met, or the condition of (center slip rate greater than its slip threshold) is met, an output is generated to produce an instability warning result.

[0079] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for early warning of ship anchoring instability by integrating multi-source heterogeneous data, characterized in that, Includes the following steps: The static gripping force extreme value is obtained based on the seabed state parameters in the nautical chart seabed data and the ship length and displacement scalars in the Automatic Identification System (AIS); the comprehensive environmental thrust parameters are generated based on the wind speed and current velocity vectors in the ship's environmental monitoring instrument. Based on the ship's anchoring position coordinates and the positioning coordinates within the ship's automatic identification system, combined with the comprehensive environmental thrust parameters, the anchor chain transmitted tension scalar is calculated; the positioning coordinates corresponding to when the anchor chain transmitted tension scalar is greater than the static holding force extreme value are selected to generate an over-limit force judgment vector; Based on the over-limit force determination vector, extract the coordinates of the corresponding point set within a specified time range, calculate the discrete matrix of the point set coordinates in the two-dimensional coordinate system, and solve for the eigenvalues ​​and eigenvectors corresponding to the discrete matrix. Based on the feature vector, extract the principal axis parameter and secondary axis parameter of the oscillation, and construct the oscillation envelope confidence ellipse. Based on the heading and speed parameters in the Automatic Identification System (AIS), the lateral yaw angle of the ship is calculated. The lateral yaw angle is used to orthogonally decompose the speed parameters to obtain the lateral velocity parameters. The lateral kinetic energy parameters are obtained by combining the displacement scalar with the lateral velocity parameters. Based on the yaw envelope confidence ellipse, an instability warning result is generated.

2. The ship anchoring instability early warning method according to claim 1, which integrates multi-source heterogeneous data, is characterized in that, The steps for obtaining the comprehensive environmental thrust parameters are as follows: The bottom sediment state parameters in the nautical chart bottom sediment data are called up, and the ship length scalar and displacement scalar in the automatic identification system are read. The displacement scalar is obtained by multiplying the draft parameter, ship beam parameter, ship length scalar and preset ship block coefficient by volume and converting it with water density. The bottom sediment state parameters and ship length scalar are matched according to the row and column index of the bottom sediment coefficient data table, and the bottom sediment coefficient recorded at the matching position is extracted. The bottom sediment coefficient is multiplied with the displacement scalar to obtain the static gripping force extreme value. The wind speed vector and flow velocity vector in the ship's environmental monitoring instrument are called up, the direction angle and magnitude components of the wind speed vector are split, the air density parameter and the surface wind-receiving area parameter are called up to calculate the load component under the action of wind direction, the direction angle and magnitude components of the flow velocity vector are split up, the seawater density parameter and the underwater flow-receiving area parameter are called up to calculate the load component under the action of flow direction, and the wind load vector and water flow load vector are generated. The lateral and longitudinal components of the wind load vector in the two-dimensional coordinate system are extracted, and the lateral and longitudinal components of the water flow load vector in the two-dimensional coordinate system are extracted. The components in the same coordinate axis direction are synthesized, and the direction angle and magnitude components of the synthesized components are retained to generate comprehensive environmental thrust parameters.

3. The ship anchoring instability early warning method according to claim 1, which integrates multi-source heterogeneous data, is characterized in that... The steps for obtaining the anchor chain transmitted tension scalar are as follows: The ship's anchoring position coordinates are called, and the positioning coordinates in the ship's automatic identification system are called. The ship's anchoring position coordinates are used as the anchor chain fixed end position, and the positioning coordinates are used as the hull towing end position. The lateral coordinate difference, longitudinal coordinate difference, and spatial interval change between the ship's anchoring position coordinates and the positioning coordinates are extracted point by point. The lowest point of suspension, the bottoming point, and the towing end connection point are marked according to the positional relationship between the two ends of the anchor chain to obtain the anchor chain suspension shape. By calling the positioning coordinates and the comprehensive environmental thrust parameters, the positioning coordinates are mapped to the traction end connection point in the anchor chain suspension configuration. The comprehensive environmental thrust parameters are decomposed into components acting along the anchor chain traction direction. Combining the directional change relationship from the bottoming point to the traction end connection point, the horizontal force at the bottoming point is calculated point by point to obtain the anchor chain transmitted tension scalar.

4. The ship anchoring instability early warning method according to claim 1, which integrates multi-source heterogeneous data, is characterized in that... The steps for obtaining the over-limit force determination vector are as follows: The static gripping force extreme value is invoked, and the anchor chain transmitted tension scalar corresponding to each positioning coordinate is compared with the static gripping force extreme value one by one. Positioning coordinates with anchor chain transmitted tension scalars greater than the static gripping force extreme value are retained, and positioning coordinates with anchor chain transmitted tension scalars less than or equal to the static gripping force extreme value are removed. The excess coordinate set is formed according to the order of the retained positioning coordinates. Then, each positioning coordinate in the excess coordinate set is converted into the excess mark value of the corresponding position to generate the excess force judgment vector.

5. The ship anchoring instability early warning method according to claim 1, which integrates multi-source heterogeneous data, is characterized in that... The steps for obtaining the coordinates of the point set are as follows: Read the over-limit marker values ​​corresponding to each position in the over-limit force determination vector, filter the positioning coordinates corresponding to the over-limit marker values ​​according to the specified time range, remove the positioning coordinates outside the specified time range, and arrange the filtered positioning coordinates according to the acquisition time sequence to obtain the point set coordinates.

6. The ship anchoring instability early warning method according to claim 1, which integrates multi-source heterogeneous data, is characterized in that... The steps for obtaining the skewed envelope confidence ellipse are as follows: Extract the horizontal and vertical coordinate values ​​of the point set in the two-dimensional coordinate system respectively, calculate the discrete offset of the horizontal coordinate value, the discrete offset of the vertical coordinate value, and the combined offset of the horizontal and vertical coordinates, and fill the matrix position according to the order of the two-dimensional coordinate axes to obtain the discrete matrix; Solve for the eigenvalues ​​and eigenvectors corresponding to the discrete matrix, sort the eigenvalues ​​by numerical value, select the eigenvector corresponding to the largest eigenvalue as the principal axis parameter of the oscillation, select the eigenvector corresponding to the smallest eigenvalue as the secondary axis parameter of the oscillation, construct the boundary line enclosing the coordinates of the point set based on the principal axis parameter of the oscillation, the secondary axis parameter of the oscillation, and the outer edge distribution position of the point set coordinates, and generate the oscillation envelope confidence ellipse.

7. The ship anchoring instability early warning method according to claim 1, which integrates multi-source heterogeneous data, is characterized in that... The steps for obtaining the lateral kinetic energy parameters are as follows: Call the heading and speed parameters in the Automatic Identification System (AIS) to extract the heading angle corresponding to the heading parameter, read the anchor reference angle corresponding to the initial anchor reference bearing, convert the heading angle relative to the anchor reference angle by angle difference, mark the lateral yaw direction of the ship according to the positive and negative directions of the angle difference, record the degree of lateral yaw of the ship according to the absolute magnitude of the angle difference, and obtain the lateral yaw angle. The speed parameters are called and decomposed into velocity components along the anchor reference direction and velocity components perpendicular to the anchor reference direction according to the lateral sway angle. The velocity component perpendicular to the anchor reference direction is extracted as the lateral velocity parameter. The lateral velocity parameter is squared and the displacement scalar is called to perform kinetic energy conversion on the squared lateral velocity parameter to obtain the lateral kinetic energy parameter.

8. The ship anchoring instability early warning method according to claim 1, which integrates multi-source heterogeneous data, is characterized in that... The steps for obtaining the instability early warning result are as follows: The secondary axis length parameter and center coordinates within the eccentric envelope confidence ellipse are called, and the difference in secondary axis length parameter and the offset of center coordinates at adjacent time points are extracted according to the acquisition time sequence. The secondary axis change rate is calculated based on the time interval between adjacent moments, and the center coordinate offset is projected onto the direction corresponding to the principal axis parameter of the yaw to obtain the center slip rate. The secondary axis rate of change is compared item by item with a preset rate of change threshold. The lateral kinetic energy parameter corresponding to the secondary axis rate of change being greater than the preset rate of change threshold is retained. The lateral kinetic energy parameter is then compared with a preset lateral kinetic energy tolerance threshold. Simultaneously, the center slip rate is compared item by item with the preset slip threshold; Output is generated when the lateral kinetic energy parameter is greater than the preset lateral kinetic energy tolerance threshold, or when the center slip rate is greater than the slip threshold, to generate an instability warning result.

Images