A method, system, device and medium for completing missing data of ship speed

Abstract

The present invention belongs to the field of shipping technology, and specifically discloses a method, system, device and medium for completing missing data of ship speed. The method of the present invention comprehensively considers the spatiality of the ship during navigation, and combines the dynamic information of the ship to introduce the ship's heading change into the Kriging interpolation, aiming to accurately complete the missing speed data according to the actual navigation status of the ship. In terms of completing the missing speed values, the method of the present invention goes beyond the traditional simple operation scope that is limited to the same type of data (i.e., the speed data itself). It has a deep insight into the spatial characteristics of the ship's navigation process, and closely combines this key element with the longitude and latitude position of the ship, the non-missing ship speed and the heading change, and jointly acts on the accurate completion of the missing speed data. The method of the present invention can achieve more accurate and comprehensive recovery and completion of the missing speed data of the ship.

Description

Technical Field

[0001] The present invention belongs to the field of shipping technology and relates to a method, system, equipment and medium for completing missing data of ship speed. Background Art

[0002] In the shipping field, the integrity and accuracy of navigation data are directly related to the safety of ship navigation, the optimization of navigation efficiency and the advancement of intelligent shipping management. Ship speed plays a vital role in ensuring navigation safety, optimizing navigation routes and deepening navigation data analysis. However, in the actual navigation process of ships, due to various uncontrollable factors such as equipment failure, signal interference, rapidly changing weather and complex fluctuations in sea conditions, the lack of key information such as ship speed in navigation records often occurs, which not only seriously damages the integrity and continuity of the ship's trajectory, but also greatly weakens the accuracy and credibility of navigation data analysis, posing a potential threat to the safe navigation and efficient management of ships. Therefore, how to effectively respond to this challenge and ensure the integrity and accuracy of navigation data has become an important issue that needs to be urgently solved in the current shipping field.

[0003] Many scholars have explored the strategies for completing missing ship data from a variety of perspectives. Some scholars tend to use traditional data completion methods, such as direct deletion, linear interpolation, and spline interpolation, to analyze and process missing ship data. Another group of scholars introduced missing data filling methods based on intelligent ship databases, making full use of the rich historical information in the ship database, and accurately completing missing data through advanced algorithms and technical means.

[0004] Although the above-mentioned data completion methods have shown their own unique advantages in different models and practical application scenarios, they also generally have some significant defects that cannot be ignored. Specifically, although these data completion methods can fill the data gaps to some extent, they often ignore the dynamic characteristics of ship navigation and the profound impact of external environmental factors on navigation speed. More importantly, they fail to take into account the spatiality of the ship's position to more accurately complete the missing speed data. This limitation often leads to a large deviation between the completion results and the actual situation, which is difficult to meet the stringent requirements of high-precision navigation data applications. In addition, these methods lack adaptability to data missing situations under complex navigation conditions. In actual navigation environments, the pattern and frequency of data missing may vary greatly depending on the navigation stage, the geographical characteristics of the sea area, and the changeable weather conditions. However, due to its inherent limitations, the traditional fixed strategy completion method cannot flexibly adapt to these variable data missing situations, which greatly limits its extensiveness and effectiveness in practical applications. Summary of the invention

[0005] The purpose of the present invention is to propose a method for completing missing data of ship speed. The method comprehensively considers the spatiality of the ship's navigation process and combines the dynamic information of the ship (longitude and latitude, speed and heading). The change of ship heading is introduced into Kriging interpolation, aiming to accurately complete the missing speed data according to the actual navigation status of the ship.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme:

[0007] A method for completing missing ship speed data comprises the following steps:

[0008] Step 1. Obtain the AIS data information of the ship in the sea channel, including the longitude, latitude, speed and heading angle of the ship, and check and find the missing ship speed data in the AIS data information;

[0009] Step 2. Extract the ship dynamic information at the time points before and after the speed missing data points, that is, the longitude, latitude, speed and heading angle of the data points that are not missing; at the same time, extract the ship longitude, latitude and heading angle of the speed missing data points;

[0010] Step 3. Calculate the spatial distance, observation value difference and heading angle difference based on the extracted ship dynamic information;

[0011] Step 4. The calculated spatial distance, observation value difference and heading angle difference are used as input values ​​and introduced into the Kriging interpolation calculation to complete the speed data of the extracted missing speed data points;

[0012] Specifically, firstly, the first covariance matrix between the non-missing data points is calculated based on the difference in observed values ​​and the heading angle difference, and at the same time, the second covariance matrix between the missing data points and the non-missing data points is calculated based on the spatial distance;

[0013] A Kriging interpolation matrix is ​​constructed based on the first covariance matrix and the second covariance matrix, and the correlation between spatial data and the heading change are considered by introducing the ship heading change into the Kriging interpolation.

[0014] The Kriging interpolation matrix is ​​solved to calculate the speed weight coefficient of each non-missing data point, and then the speed value of the speed missing data point is calculated by weighted summation.

[0015] In addition, based on the ship speed missing data completion method, the present invention also proposes a corresponding ship speed missing data completion system, which adopts the following technical solutions:

[0016] A ship speed missing data completion system includes the following modules:

[0017] AIS data information acquisition module, used to obtain AIS data information of ships in the sea channel, including the longitude, latitude, speed and heading angle of the ship, and check and find the missing ship speed data in the AIS data information;

[0018] The extraction module is used to extract the ship dynamic information at the time points before and after the speed missing data point, that is, the longitude, latitude, speed and heading angle of the non-missing data point; at the same time, the longitude, latitude and heading angle of the ship at the speed missing data point are extracted;

[0019] The pre-processing module is used to calculate the spatial distance, observation value difference and heading angle difference based on the ship dynamic information;

[0020] and a speed completion module, which is used to introduce the calculated spatial distance, observation value difference and heading angle difference as input values ​​into the Kriging interpolation calculation to complete the speed data of the extracted speed missing data points;

[0021] Specifically, firstly, the first covariance matrix between the non-missing data points is calculated based on the difference in observed values ​​and the heading angle difference, and at the same time, the second covariance matrix between the missing data points and the non-missing data points is calculated based on the spatial distance;

[0022] A Kriging interpolation matrix is ​​constructed based on the first covariance matrix and the second covariance matrix, and the correlation between spatial data and the heading change are considered by introducing the ship heading change into the Kriging interpolation.

[0023] The Kriging interpolation matrix is ​​solved to calculate the speed weight coefficient of each non-missing data point, and then the speed value of the speed missing data point is calculated by weighted summation.

[0024] In addition, based on the method for completing missing data of ship speed, the present invention also proposes a computer device, which includes a memory and one or more processors. An executable code is stored in the memory. When the processor executes the executable code, it is used to implement the method for completing missing data of ship speed.

[0025] In addition, based on the method for completing missing data of ship speed, the present invention also proposes a computer-readable storage medium on which a program is stored. When the program is executed by a processor, it is used to implement the method for completing missing data of ship speed.

[0026] The present invention has the following advantages:

[0027] As described above, the present invention relates to a method for completing missing data of ship speed, which comprehensively considers the spatiality of the ship during navigation, and combines the dynamic information of the ship (i.e., longitude and latitude, speed and heading), introduces the ship's heading change (heading angle difference) into the Kriging interpolation, thereby comprehensively calculating the missing value of the speed from multiple perspectives (distance, speed difference and heading angle difference), and calculates the missing speed value by improving the Kriging interpolation method. The method of the present invention comprehensively considers the spatiality of the ship during navigation, and combines the dynamic information of the ship, introduces the ship's heading change into the Kriging interpolation, and thus can accurately complete the missing speed data according to the actual navigation status of the ship. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Flow chart of the method for completing missing data of ship speed in an embodiment of the present invention.

[0029] Figure 2 It is a schematic diagram of missing speed values ​​before being processed by the method of the present invention.

[0030] Figure 3 It is a schematic diagram of missing speed values ​​after being processed by the method of the present invention. DETAILED DESCRIPTION

[0031] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0032] Example 1

[0033] Traditional ship speed data completion methods mainly rely on simple interpolation techniques, which are often limited to operations within the same type of data (such as speed data), usually by calculating the average of the adjacent speeds before and after the missing speed value to fill in. However, this method significantly ignores the inherent dynamic characteristics of the ship during navigation and the complex impact of external environmental factors on its speed. More importantly, traditional methods fail to fully consider the spatial attributes of the ship's position, which is crucial for accurately completing the missing speed data. The present invention proposes a method for completing missing ship speed data, which comprehensively considers the spatiality of the ship's navigation process and combines the dynamic information of the ship, introduces the ship's heading changes into the Kriging interpolation, and aims to achieve accurate completion of the missing speed data based on the actual navigation status of the ship. In terms of completing the missing speed values, the present invention goes beyond the traditional simple operation scope limited to the same type of data (i.e., the speed data itself). It has a deep insight into the spatiality of the ship's navigation process (i.e., spatial distance), and closely combines this key factor with the longitude and latitude position of the ship, the speed of the non-missing ship, and the course change, which together act on the accurate completion process of the missing speed data, so that the method of the present invention can achieve more accurate and comprehensive recovery and completion of the missing speed data of the ship.

[0034] like Figure 1 As shown, the method for completing missing data of ship speed in this embodiment includes the following steps:

[0035] Step 1. Obtain the AIS data information of the ship in the sea channel, including the longitude, latitude, speed and heading angle of the ship, and check and find the missing ship speed data in the AIS data information.

[0036] Step 2. Extract the ship dynamic information before and after the speed missing data point, that is, the longitude, latitude, speed and heading angle of the non-missing data point; at the same time, extract the ship longitude, latitude and heading angle of the speed missing data point.

[0037] The dynamic information of the ship is expressed as .

[0038] in, , Indicates the longitude and latitude of the ship, Indicates the speed of the ship, Indicates the heading angle of the ship.

[0039] Step 3. Calculate the spatial distance, observation value difference and heading angle difference based on the extracted ship dynamic information (i.e. the extracted longitude, latitude, speed and heading angle information of the ship).

[0040] In Kriging interpolation, calculating spatial distance is a key step to quantify the correlation between spatial data. For each pair of ship data points, the position of the ship is composed of longitude and latitude.

[0041] Therefore, this embodiment uses the Haversine formula to calculate the spatial distance between two points. The Haversine formula is a mathematical formula used to calculate the distance between two coordinate points on a sphere, and is particularly suitable for spherical distance calculation on the earth. The Haversine formula is based on spherical trigonometry and uses the arc length between two points on a sphere to calculate the distance between them.

[0042] The spatial distance between two points can be calculated using the following formula:

[0043] ;

[0044] ;

[0045] .

[0046] in, Indicates the longitude and latitude of the first point; Represents the longitude and latitude of the second point; The longitude difference is ; Indicates latitude difference ; represents the radius of the earth (taken as 6371 kilometers); Represents the spatial distance between two points; Represents the inverse tangent function.

[0047] There is usually a spatial correlation between the change in speed due to geographic location, that is, the speed change trends of adjacent locations are more similar. The present invention uses spatial distance to better capture this spatial correlation and improve interpolation accuracy. For speed completion, spatial distance can provide certain geographic location dependency information, so that interpolation not only depends on the speed value itself, but also takes into account the spatial distribution characteristics of different locations, thereby obtaining a more accurate speed estimate.

[0048] In addition, since the present invention is to complete the missing value of the ship speed, the speed of the ship is taken as the observed value to calculate the difference between the ship speeds. The calculation formula is as follows: .

[0049] in, Expressed as the difference between ship speeds, , Indicates the speed of the ship .

[0050] The difference in observed values ​​can better reflect the dynamic changes of speed, avoid unreasonable stationary assumptions in the interpolation process, and help to infer more reasonable speed values ​​based on existing data when the speed is missing.

[0051] Introducing speed observation differences can effectively capture the changes in speed between different time points or locations, especially when speed is missing, and can provide supplementary information about the speed change trend. By calculating the differences in observations, the interpolation model can identify the law of speed changes and reduce the errors caused by missing data.

[0052] In addition, the heading angle difference provides directional information, while Kriging interpolation itself is usually used for interpolation of spatial data, ignoring directionality. The present invention introduces the heading angle difference, not only interpolating the missing data in space, but also considering the direction of speed change. Since speed has not only magnitude, but also direction. Direct interpolation of speed may ignore directional changes, but by introducing the heading angle difference, the rationality and consistency of direction and speed can be ensured.

[0053] The normalized calculation formula for the heading angle difference is as follows: .

[0054] If the calculated , then it is calculated by the following formula: .in, Indicates the heading angle difference; , Indicates the heading angle of the ship; mod means remainder.

[0055] During navigation, the change of heading angle usually has a certain impact on the speed, especially when turning or making a turn, the speed often changes. Therefore, by introducing the heading angle difference, the present invention can help the model better understand the physical and dynamic laws behind the speed change, especially those involving turning, acceleration or deceleration.

[0056] The impact of navigation angle difference on speed completion is very important, especially when the heading changes significantly, the change in speed is often closely related to the heading change. Introducing the heading angle difference helps capture the law of speed changes caused by heading changes, ensuring that speed completion not only considers the position and speed differences, but also the impact of the navigation direction.

[0057] Since spatial distance can help consider the correlation between locations, the difference in observed values ​​reflects the trend of speed changes, and the difference in navigation angles further reveals the speed changes caused by changes in heading. Therefore, through the combination of these three factors, Kriging interpolation can not only extract information from geographic location and speed data, but also better simulate the law of speed changes, especially considering the dynamic characteristics of heading changes, providing more accurate completion results for the missing speed in the later stage.

[0058] Step 4. The calculated spatial distance, observation value difference, and heading angle difference are used as input values ​​and introduced into the Kriging interpolation calculation to complete the speed data of the extracted missing speed data points.

[0059] Specifically, the first covariance matrix between the non-missing data points is calculated based on the observation value difference and the heading angle difference, and the second covariance matrix between the missing data points and the non-missing data points is calculated based on the spatial distance.

[0060] The Kriging interpolation matrix is ​​built based on the first covariance matrix and the second covariance matrix. By introducing the ship heading change into the Kriging interpolation, the correlation between spatial data and heading change are taken into account.

[0061] The Kriging interpolation matrix is ​​solved to calculate the speed weight coefficient of each non-missing data point, and then the speed value of the speed missing data point is calculated by weighted summation.

[0062] Kriging interpolation calculates the spatial distance between data. , considering the spatial relationship between data and introducing the change of ship's heading into Kriging interpolation, enabling it to be dynamically adjusted according to the ship's heading angle. The present invention not only considers the distance but also takes into account the influence of the heading angle, ensuring that the actual motion law of the ship is considered during the interpolation process.

[0063] Traditional Kriging interpolation methods usually rely only on the relationship of spatial positions, while the present invention infers data values by calculating the distances between sample points. However, the motion of a ship is affected not only by spatial positions but also by dynamic factors such as heading angle and ship turning. Especially during ship navigation, speed changes are often closely related to the heading.

[0064] To accurately predict the speed of a ship at different positions, especially in the case of significant changes in the ship's heading during navigation, the single spatial distance metric of traditional Kriging interpolation is insufficient. After introducing the heading angle, the present invention can adjust the weights of different data points according to the change of the ship's heading, thereby performing dynamic optimization. This method can ensure that when the ship is in the same direction or the headings are relatively consistent, the mutual influence of relevant data points is stronger. When the heading changes, the interpolation process will automatically adjust, weakening the role of the spatial distance and enhancing the influence of the heading factor.

[0065] Introducing the change of heading into Kriging interpolation and dynamically adjusting the influence of spatial distance and heading angle provides a more accurate and flexible solution for missing speed filling. The present invention not only considers the spatial relationship but also can adapt to the motion law of the ship under different headings. When facing missing speed data, it can more accurately fill in the speed of the ship. Through this method, the interpolation result can better reflect the actual motion trajectory of the ship. Especially in areas with large heading changes, it can effectively reduce the errors brought by traditional interpolation methods, thereby improving the accuracy and reliability of missing speed data filling.

[0066] The detailed process of step 4 is elaborated below.

[0067] Step 4.1. Construct the first covariance matrix between non-missing data points.

[0068] First, calculate the non-missing data points and between and , and then calculate the semi-variance according to the following formula , and construct the corresponding first covariance matrix . Each element in the first covariance matrix is represented by , that is .

[0069] .

[0070] Among them, i and j represent the numbers of non-missing data points, and n represents the number of non-missing data points; ; represents the variance, represents the difference between the speeds of the ships, Indicates the heading angle difference.

[0071] Differences between ship speeds The calculation formula is as follows:

[0072] .

[0073] in, , represents the speed of the ship's non-missing data points i and j, .

[0074] Heading angle difference The calculation formula is as follows: .

[0075] in, , Indicates the heading angle of the ship i and j at the non-missing speed point; if the calculated , then rewrite it by the following formula Values: ; Among them, mod means remainder.

[0076] The semivariance function is designed to calculate the semivariance of the non-missing ship data (longitude, latitude, speed, heading angle).

[0077] Since the position, speed and heading angle of the ship change during the voyage, the cos( ) weight factor, that is, the speed of the ship will change relative to the heading angle, and the difference in sailing angles at different times will also be different, so as to accurately complete the missing data of the ship. The reasons for this are as follows:

[0078] 1. Reflecting directional correlation:

[0079] When moving, the cosine function can reflect the influence of their relative directions in space. If the heading angle difference θ of the ships is small, then their moving directions are relatively close, and cosine ( ) can reflect the higher correlation between them. On the contrary, when θ increases (that is, the difference in heading angles between ships increases), their correlation weakens and the cosine value gradually approaches zero.

[0080] 2. Properties of the cosine function:

[0081] Symmetry: The cosine function has symmetry, that is, cos( ) = cos(Δ(-θ)), which shows that the correlation is the same whether the relative orientation of the ships is clockwise or counterclockwise. This is very important for modeling heading angle differences, because the relative orientation between ships does not depend on how they are rotated, only on their angular difference.

[0082] Smooth variation: The cosine function varies smoothly with the angle difference θ. When the angle approaches zero, cos( ) is close to 1, indicating that the directions of the ships are close and the correlation is strong; when the angle is close to 90°, the cosine value is close to 0, indicating that the directions of the ships are vertical and the correlation is weak. This smooth change is consistent with the physical phenomenon and the actual situation of ship movement.

[0083] 3. Consistent with point-to-point distance calculation:

[0084] The cosine function is used to describe the angle between points and their relative position. Especially when considering navigation, the relative angle difference between ships directly affects their speed and trajectory. If the two ships are heading in the same direction (θ=0°), their relative speed is close to the maximum value, which coincides with cos(0°)=1. If the angle difference between the two ships reaches 90° (i.e. completely vertical), their relative speed decreases and the value of the cosine function is 0, which is also consistent with the actual situation.

[0085] 4. In navigation, changes in heading angle and direction directly affect the relative position and motion between ships. Relative motion is usually described by the angular difference between ships, and the cosine function can well describe the effect of changes in navigation angle.

[0086] Step 4.2. Construct the second covariance matrix between the missing data points and the non-missing data points.

[0087] The semivariance value tends to gradually show a trend of stabilization with the increase of distance and the change of heading angle. In order to accurately describe and fit the difference characteristics in spatial data, the present invention selects a Gaussian model, which can derive a functional expression of the semivariance, thereby achieving effective fitting and description of the difference of spatial data.

[0088] Counting missing data points With each non-missing data point The space distance between , according to the semivariance function formula of Gaussian simulation fitting, calculate the semivariance , and construct the corresponding second covariance matrix , .

[0089] .

[0090] in, represents the semivariance obtained by fitting the Gaussian function, It represents the variance of the observed value, i.e., the velocity data; represents the decay rate of spatial autocorrelation, represents the spatial distance between the missing data point and each non-missing data point i.

[0091] It is to calculate the variance of the data of all data points, and its calculation method is relatively conventional.

[0092] The value representing the decay rate of spatial autocorrelation is obtained by fitting the Gaussian model. and The relationship between. Indicates the value of the decay speed of spatial autocorrelation. Through the autocorrelation analysis, the correlation of the data is significant within a certain range, and then Set to a specific value, indicating that the data has significant spatial correlation within this range.

[0093] For example, if the autocorrelation analysis finds that the data correlation is significant within 100 meters, but almost disappears when it exceeds 100 meters, It is set to 100 meters, indicating that the data has significant spatial correlation within this range.

[0094] Use the Haversine formula to calculate the spatial distance between two points , the formula is as follows:

[0095] ;

[0096] ;

[0097] .

[0098] in, represents the spatial distance between the missing data point and the non-missing data point i, , Indicates the longitude and latitude of missing data points; Indicates the longitude and latitude of the non-missing data points; represents the longitude difference, ; represents the latitude difference, ; represents the radius of the Earth; Represents the inverse tangent function.

[0099] The semivariance function expression obtained by the Gaussian model fitting in the present invention is to calculate the semivariance of the speed missing point. First, the longitude and latitude of the ship at the speed missing point are used to calculate the distance between it and the non-missing ship data. , and then estimate the semivariance of the velocity missing point by fitting the semivariance function of the Gaussian model .

[0100] Step 4.3. Kriging interpolation method obtains weight coefficients by solving a system of linear equations , the formula is as follows:

[0101] ;

[0102] in, is the weight coefficient of Kriging interpolation; is the Lagrange multiplier used to handle the unbiasedness constraint.

[0103] Step 4.4. Figure 2 and Figure 3 As shown, the ultimate goal of the present invention is to calculate the missing ship speed value. By introducing the ship heading change into the Kriging interpolation, the correlation between spatial data and heading variability are considered, and the speed value of the missing data point is calculated by weighted summation. The calculation formula is as follows:

[0104] .

[0105] in, Ship speed values ​​representing missing data points; Represents the velocity value for the non-missing data points.

[0106] When completing the missing speed data of a ship, the present invention breaks through the limitation of the traditional ship data completion method that only relies on the same type of data, innovatively incorporates the spatial characteristics of ship navigation, and combines the longitude and latitude position of the ship, known speed information, and heading change information to comprehensively improve the completion accuracy of the missing speed data. By introducing the ship heading change into the Kriging interpolation, the missing speed data can be accurately completed according to the actual navigation status of the ship.

[0107] Example 2

[0108] This embodiment 2 describes a system for completing missing data on ship speed, which is based on the same inventive concept as the method for completing missing data on ship speed described in the above embodiment 1.

[0109] A ship speed missing data completion system includes the following modules:

[0110] AIS data information acquisition module, used to obtain AIS data information of ships in the sea channel, including the longitude, latitude, speed and heading angle of the ship, and check and find the missing ship speed data in the AIS data information;

[0111] The extraction module is used to extract the ship dynamic information at the time points before and after the speed missing data point, that is, the longitude, latitude, speed and heading angle of the non-missing data point; at the same time, the longitude, latitude and heading angle of the ship at the speed missing data point are extracted;

[0112] The pre-processing module is used to calculate the spatial distance, observation value difference and heading angle difference based on the ship dynamic information;

[0113] and a speed completion module, which is used to introduce the calculated spatial distance, observation value difference and heading angle difference as input values ​​into the Kriging interpolation calculation to complete the speed data of the extracted speed missing data points;

[0114] Specifically, firstly, the first covariance matrix between the non-missing data points is calculated based on the difference in observed values ​​and the heading angle difference, and at the same time, the second covariance matrix between the missing data points and the non-missing data points is calculated based on the spatial distance;

[0115] A Kriging interpolation matrix is ​​constructed based on the first covariance matrix and the second covariance matrix, and the correlation between spatial data and the heading change are considered by introducing the ship heading change into the Kriging interpolation.

[0116] The Kriging interpolation matrix is ​​solved to calculate the speed weight coefficient of each non-missing data point, and then the speed value of the speed missing data point is calculated by weighted summation.

[0117] It should be noted that, in the ship speed missing data completion system in this embodiment 2, the implementation process of the functions and effects of each functional module is detailed in the implementation process of the corresponding steps of the method in the above embodiment 1, and will not be repeated here.

[0118] Example 3

[0119] This embodiment 3 describes a computer device. The computer device includes a memory and one or more processors. An executable code is stored in the memory. When the processor executes the executable code, it is used to implement the steps of the method for completing missing ship speed data in the above embodiment 1.

[0120] In this embodiment, the computer device is any device or apparatus with data processing capability, which will not be described in detail here.

[0121] Example 4

[0122] This embodiment 4 describes a computer-readable storage medium on which a program is stored. When the program is executed by a processor, it is used to implement the steps of the method for completing missing ship speed data in the above-mentioned embodiment 1.

[0123] The computer-readable storage medium may be an internal storage unit of any device or apparatus with data processing capabilities, such as a hard disk or memory, or an external storage device of any device with data processing capabilities, such as a plug-in hard disk, a smart media card (SMC), an SD card, a flash card, etc., equipped on the device.

[0124] Of course, the above description is only a preferred embodiment of the present invention, and the present invention is not limited to the above embodiments. It should be noted that all equivalent substitutions and obvious deformation forms made by any technician familiar with the field under the guidance of this specification fall within the essential scope of this specification and should be protected by the present invention.

Claims

1. A method for completing missing data of ship speed, characterized in that: The steps include: Step 1. Obtain the AIS data information of the ship in the sea channel, including the longitude, latitude, speed and heading angle of the ship, and check and find the missing ship speed data in the AIS data information; Step 2. Extract the ship dynamic information at the time points before and after the speed missing data points, that is, the longitude, latitude, speed and heading angle of the data points that are not missing; at the same time, extract the ship longitude, latitude and heading angle of the speed missing data points; Step 3. Calculate the spatial distance, observation value difference and heading angle difference based on the extracted ship dynamic information; Step 4. The calculated spatial distance, observation value difference and heading angle difference are used as input values ​​and introduced into the Kriging interpolation calculation to complete the speed data of the extracted missing speed data points; Firstly, the first covariance matrix between the non-missing data points is calculated based on the difference in observed values ​​and the heading angle difference, and at the same time, the second covariance matrix between the missing data points and the non-missing data points is calculated based on the spatial distance and through Gaussian fitting; A Kriging interpolation matrix is ​​constructed based on the first covariance matrix and the second covariance matrix, and the correlation between spatial data and the heading change are considered by introducing the ship heading change into the Kriging interpolation. The Kriging interpolation matrix is ​​solved to calculate the speed weight coefficient of each non-missing data point, and then the speed value of the speed missing data point is calculated by weighted summation.

2. The method for completing missing ship speed data according to claim 1, characterized in that: In step 2, the dynamic information of the ship is expressed as ; in, , Indicates the longitude and latitude of the ship, Indicates the speed of the ship, Indicates the heading angle of the ship.

3. The method for completing missing ship speed data according to claim 1, characterized in that: The step 4 is specifically as follows: Step 4.

1. Construct the first covariance matrix between non-missing data points; First calculate the non-missing data points and Between and , and then calculate the semivariance according to the following formula , and construct the corresponding first covariance matrix , the first covariance matrix Each element in express; ； Among them, i and j represent the numbers of non-missing data points, and n represents the number of non-missing data points; ; represents the variance, represents the difference between the speeds of the ships, Indicates the heading angle difference; Step 4.

2. Construct the second covariance matrix between the missing data points and the non-missing data points; Counting missing data points With each non-missing data point The space distance between , calculate the semivariance according to the semivariance function formula of Gaussian simulation fitting , and construct the corresponding second covariance matrix , ; ； in, represents the semivariance obtained by fitting the Gaussian function, It represents the variance of the observed value, i.e., the velocity data; represents the decay rate of spatial autocorrelation, represents the spatial distance between the missing data point and each non-missing data point i; Step 4.

3. Build the Kriging interpolation matrix based on the first covariance matrix and the second covariance matrix. The Kriging interpolation method obtains the speed weight coefficient of each non-missing data point by solving the linear equation system. , the formula is as follows: ； in, is the weight coefficient of Kriging interpolation; is the Lagrange multiplier used to handle the unbiasedness constraint; Step 4.

4. By introducing the ship heading change into the Kriging interpolation, the correlation between spatial data and heading variability are taken into account, and the speed value of the missing data point is calculated by weighted summation.

4. The method for completing missing ship speed data according to claim 3, characterized in that: In step 4.1, the difference between the ship speeds The calculation formula is as follows: ； in, , represents the speed of the ship's non-missing data points i, j.

5. The method for completing missing ship speed data according to claim 3, characterized in that: In step 4.1, the heading angle difference The calculation formula is as follows: ; in, , Indicates the heading angle of the ship i and j at the non-missing speed point; if the calculated , then rewrite it by the following formula Values: ; where mod means remainder.

6. The method for completing missing ship speed data according to claim 3, characterized in that: In step 4.2, the Haversine formula is used to calculate the spatial distance between two points. The formula is expressed as follows: ； ； ； in, represents the spatial distance between the missing data point and the non-missing data point i, Indicates the longitude and latitude of missing data points; Indicates the longitude and latitude of the non-missing data points; represents the radius of the Earth; represents the inverse tangent function; represents the longitude difference, ; represents the latitude difference, .

7. The method for completing missing data of ship speed according to claim 3, characterized in that: In step 4.4, the formula for calculating the velocity value of the missing data point by weighted summation is as follows: ； in, Ship speed values ​​representing missing data points; Represents the velocity value for the non-missing data points.

8. A system for completing missing data of ship speed, characterized in that: Includes the following modules: AIS data information acquisition module, used to obtain AIS data information of ships in the sea channel, including the longitude, latitude, speed and heading angle of the ship, and check and find the missing ship speed data in the AIS data information; The extraction module is used to extract the ship dynamic information at the time points before and after the speed missing data point, that is, the longitude, latitude, speed and heading angle of the non-missing data point; at the same time, the longitude, latitude and heading angle of the ship at the speed missing data point are extracted; The pre-processing module is used to calculate the spatial distance, the difference of the observed values ​​and the heading angle difference according to the ship dynamic information; and a speed completion module, which is used to introduce the calculated spatial distance, observation value difference and heading angle difference as input values ​​into the Kriging interpolation calculation to complete the speed data of the extracted speed missing data points; Specifically, firstly, the first covariance matrix between the non-missing data points is calculated based on the difference in observed values ​​and the heading angle difference, and at the same time, the second covariance matrix between the missing data points and the non-missing data points is calculated based on the spatial distance; A Kriging interpolation matrix is ​​constructed based on the first covariance matrix and the second covariance matrix, and the correlation between spatial data and the heading change are considered by introducing the ship heading change into the Kriging interpolation. The Kriging interpolation matrix is ​​solved to calculate the speed weight coefficient of each non-missing data point, and then the speed value of the speed missing data point is calculated by weighted summation.

9. A computer device comprising a memory and one or more processors; an executable code is stored in the memory; characterized in that: When the processor executes the executable code, it is used to implement the steps of the method for completing missing ship speed data as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a program stored thereon; characterized in that: When the program is executed by a processor, it is used to implement the steps of the method for completing missing ship speed data as described in any one of claims 1 to 7 above.

Images