Monitoring data acquisition and analysis method

Abstract

The invention relates to the technical field of data acquisition and analysis, and discloses a monitoring data acquisition and analysis method, which comprises the following steps: acquiring monitoring data of multiple data sources of a ship, a longitude set A, a latitude set B, a course set C, a speed set D and a distance set E of the ship to a destination, and preprocessing the acquired monitoring data; according to the method, by collecting monitoring data of multiple data sources of the ship, including longitude, latitude, course, navigational speed and the distance between the ship and the destination, comprehensive monitoring of the ship navigation state is achieved, multiple dimensions of ship navigation are comprehensively considered, evaluation of the ship navigation state is more comprehensive and accurate, and the ship navigation state is more accurate. Through accurate prediction of the ship navigation trajectory, powerful support can be provided for navigation safety of the ship, potential risks in the navigation process are reduced, and meanwhile, the accurate prediction result is beneficial to optimization of the navigation route and speed of the ship, improvement of the navigation efficiency and reduction of the operation cost.

Description

Technical Field

[0001] The present invention relates to the technical field of data collection and analysis, and in particular to a monitoring data collection and analysis method. Background Art

[0002] Monitoring data refers to real-time or historical information about a specific environment, system or object collected through various monitoring devices and technical means. This data is usually used in multiple fields such as security monitoring, system performance monitoring, and environmental monitoring. It aims to provide a comprehensive understanding of the target status and help relevant personnel identify problems, make decisions and take actions in a timely manner.

[0003] With the continuous growth of global trade and the increasing busyness of maritime traffic, the prediction of ship navigation trajectories has become particularly important. Accurate navigation trajectory prediction not only helps to improve the safety and efficiency of maritime traffic, but also provides strong support for ship scheduling, route planning, fuel consumption optimization and emergency response. However, the prediction of ship navigation trajectories faces many challenges, such as massive data processing from multiple monitoring data sources, complex and changeable navigation environment, and nonlinear characteristics of ship dynamic behavior. At the same time, the existing technology lacks abnormal monitoring and alarm mechanisms for monitoring data. When the data is abnormal, it is often impossible to discover the abnormal situation in the first time, and the troubleshooting work needs to be done manually, resulting in a lag in the troubleshooting function. Therefore, a monitoring data collection and analysis method is proposed to solve the above problems. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a monitoring data collection and analysis method, which has the advantages of wide applicability, universal data source collection, and convenience for abnormal analysis of monitoring data, and solves the problems mentioned in the above background technology.

[0005] In order to achieve the above-mentioned purposes of wide application, universal data source collection, and convenience for abnormal analysis of monitoring data, the present invention provides the following technical solutions: a monitoring data collection and analysis method, comprising the following steps: S1: Collect monitoring data from multiple data sources of ships, including longitude set A, latitude set B, heading set C, speed set D, and distance set E from the ship to the destination; Configure the corresponding adapter for each data collection type, which includes: longitude set A, latitude set B, heading set C, speed set D and distance set E from the ship to the destination; S2: pre-processing the collected monitoring data to obtain a longitude set A3, a latitude set B3, a heading set C3, a speed set D3, and a distance set E3 between the ship and the destination; S3: Set corresponding weights for the longitude set A3, latitude set B3, heading set C3, speed set D3 and distance set E3 from the ship to the destination, and calculate them as a comprehensive reference index S. The sequence data S' of the comprehensive reference index S is calculated, S' = [S1, S2, ..., S T ], T is the sequence length, S T is the comprehensive reference index S at the Tth moment; S4: Construct a TBENet network learning model, input the sequence data of sequence data S' into the TBENet network learning model, and improve the prediction ability of the TBENet network learning model; S5: Test the prediction ability of the TBENet network learning model, optimize the parameters of the TBENet network learning model using the AdamW optimizer, and train the TBENet network learning model; S6: The Bayesian optimization method is used to adjust the hyperparameters of the TBENet network learning model to obtain the optimal parameters of the TBENet network learning model. The optimal parameters are introduced into the TBENet network learning model to obtain the optimized TBENet network learning model, which makes the prediction accuracy of the ship's navigation trajectory higher.

[0006] Preferably, the specific steps of step S2 are: S2.1: Use Lagrange interpolation method to supplement the monitoring data longitude set A, latitude set B, heading set C, speed set D and distance set E from the ship to the destination, and obtain longitude set A1, latitude set B1, heading set C1, speed set D1 and distance set E1 from the ship to the destination, specifically: S2.1.1: For n+1 known data points (X0, Y0), (X1, Y1), ..., (X n , Y n ), Lagrange interpolation polynomial L Y (X) The expression is: Among them, X is the independent variable, Y i For X i The value of Y∈{A, B, C, D, E}, L i (X) is the basis polynomial; S2.1.2: Basic polynomial L i (X) Specifically: Among them, X i With X j is the independent variable with known data points, satisfying L i (X i )=1 and satisfy Li (X i )=0; S2.1.3: The calculated Lagrangian interpolation polynomial expression L A (X)、L B (X)、L C (X)、L D (X) and L E (X) The longitude set A1, latitude set B1, heading set C1, speed set D1 and distance set E1 from the ship to the destination are calculated and expressed as: Among them, Y1∈{A1, B1, C1, D1, E1}, Y new is the interpolation value, Y new ∈{A new , B new , C new , D new 、E new}; S2.2: Min-Max normalize the longitude set A1, latitude set B1, heading set C1, speed set D1 and the distance set E1 from the ship to the destination, scale the data to [0,1], and obtain the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 from the ship to the destination. The expression is: Among them, Y2∈{A2, B2, C2, D2, E2}; S2.3: Calculate the standard deviation of the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 between the ship and the destination, and standardize the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 between the ship and the destination to obtain the standardized longitude set A3, latitude set B3, heading set C3, speed set D3 and the distance set E3 between the ship and the destination, expressed as: Among them, Y` σ is the standard deviation, Y i is the i-th data in each set, Y3∈{A3, B3, C3, D3, E3}, It is the average value of A3, B3, C3, D3, and E3.

[0007] Preferably, the step S3 is specifically: in, is the average value in the longitude set A3, W A is the relative weight of the average value in the longitude set A3, is the average value in the latitude set B3, W B is the relative weight of the average value in the latitude set B3, is the average value in the heading set C3, W C is the relative weight of the average value in the heading set C3, is the average value in the speed set D3, W D is the relative weight of the average value in the speed set D3, is the average value of the distance set E3 between the ship and the destination, W E It is the weight relative to the average value in the distance set E3 between the ship and the destination.

[0008] Preferably, the specific steps of step S4 are: S4.1: After the sequence data S' enters the first Bi-GRU layer, it is processed to obtain the sequence data H1. The activation function ReLU is used to process the sequence data H1 into the sequence data H'1, and the sequence data H'1 is transmitted to the first fully connected layer and the first Dropout layer. The expression is: S4.2: After receiving the sequence data H'1, the first fully connected layer and the first Dropout layer process the sequence data H'1 into sequence data H2, and the first Dropout layer processes the sequence data H2 into sequence data H'2, and transmits the sequence data H'2 to the second Bi-GRU layer. The expression is: S4.3: After receiving the sequence data H'2, the second Bi-GRU layer processes the sequence data H3, uses the activation function ReLU to process the sequence data H3 into sequence data H'3, and transmits the sequence data H'3 to the second fully connected layer and the second Dropout layer. The expression is: S4.4: After the second fully connected layer and the second Dropout layer receive the sequence data H'3, the second fully connected layer processes the sequence data H'3 into sequence data H4, the second Dropout layer processes the sequence data H4 into H'4, and transmits the sequence data H'4 to the third Bi-GRU layer. The expression is: S4.5: After receiving the sequence data H'4, the third Bi-GRU layer processes the sequence data H5, uses the activation function ReLU to process the sequence data H5 into sequence data H'5, and transmits the sequence data H'5 to the third fully connected layer and the third Dropout layer. The expression is: S4.6: After the third fully connected layer and the third Dropout layer receive the sequence data H'5, the second fully connected layer processes the sequence data H'5 into sequence data H6, and the third Dropout layer processes the sequence data H6 into H'6, and transmits the sequence data H'6 to the Bi-LSTM layer. The expression is: S4.7: After receiving the sequence data H'6, the Bi-LSTM layer processes the sequence data H'6 into sequence data H7, and transmits the sequence data H7 to the fourth fully connected layer. The expression is: S4.8: After receiving the sequence data H7, the fourth fully connected layer calculates and processes the sequence data H7 through the Sigmoid function to obtain the sequence data M, which is expressed as: Preferably, the step S4 is specifically: S4.1: The processed sequence data M is evaluated for quality using binary cross entropy, expressed as: Among them, N is the total number of samples, Q t is the true label, indicating whether the ship's trajectory has shifted at time t, M t is the predicted value, L is the loss amount; S4.2: Use the AdamW optimizer to optimize the parameters of the TBENet network learning model. After multiple iterations of optimization, the optimal TBENet network learning model is obtained. The specific optimization of the AdamW optimizer is as follows: Among them, θ z is the parameter of the zth step, η z is the learning rate, P z is the first-order moment estimate of the gradient, U zis the second-order moment estimate of the gradient, ε is a constant, to prevent division by zero, λ is the weight attenuation function; S4.3: The specific steps for training the TBENet network learning model are: S4.3.1: Initialize TBENet network learning model parameters; S4.3.2: For each epoch, include the following steps: Forward propagation: Calculate the predicted value M t and the loss amount L; Back propagation: calculate gradients and update TBENet network learning model parameters; Evaluate the performance of the TBENet network learning model and record the optimal TBENet network learning model parameters.

[0009] Preferably, the step S6 specifically includes: S6.1: Select initial parameter space; S6.2: Train the surrogate model using the parameters in the initial parameter space to approximate the unknown objective function; S6.3: Based on the current proxy model, use the acquisition function to determine the next sampling point; S6.4: Evaluate the true objective function at the selected sampling points, add the newly obtained sampling points to the existing sampling point set to obtain an updated sampling point set, and retrain the proxy model using the updated sampling point set; S6.5: Repeat steps S6.3-S6.4 until the stopping condition is met, and a trained proxy model is obtained. The parameters in the trained proxy model are used as the optimal parameters of the TBENet network learning model.

[0010] Compared with the prior art, the present invention provides a monitoring data collection and analysis method, which has the following beneficial effects: This monitoring data collection and analysis method realizes comprehensive monitoring of the ship's navigation status by collecting monitoring data from multiple data sources of the ship, including longitude, latitude, heading, speed and the distance of the ship from the destination. It comprehensively considers multiple dimensions of ship navigation, making the assessment of the ship's navigation status more comprehensive and accurate. Through accurate prediction of the ship's navigation trajectory, it can provide strong support for the ship's navigation safety and reduce potential risks during the navigation process. At the same time, accurate prediction results can also help optimize the ship's navigation route and speed, improve navigation efficiency and reduce operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic diagram of a monitoring data collection and analysis method proposed by the present invention; Figure 2This is a schematic diagram of the internal flow of the TBENet network learning model of a monitoring data collection and analysis method proposed in the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] See also Figure 1-2 , a monitoring data collection and analysis method, comprising the following steps: S1: Collect monitoring data from multiple data sources of ships, including longitude set A, latitude set B, heading set C, speed set D, and distance set E from the ship to the destination; S2: pre-processing the collected monitoring data to obtain a longitude set A3, a latitude set B3, a heading set C3, a speed set D3, and a distance set E3 between the ship and the destination; By collecting monitoring data from multiple data sources of ships, including longitude, latitude, heading, speed and the distance of the ship from the destination, comprehensive monitoring of the ship's navigation status is achieved. The data preprocessing step further ensures the accuracy and reliability of the data, providing a solid foundation for subsequent analysis and prediction.

[0014] S3: Set corresponding weights for the longitude set A3, latitude set B3, heading set C3, speed set D3 and distance set E3 from the ship to the destination, and calculate them as a comprehensive reference index S. The sequence data S' of the comprehensive reference index S is calculated, S' = [S1, S2, ..., S T ], T is the sequence length, S T is the comprehensive reference index S at the Tth moment; The key parameters such as longitude, latitude, heading, speed and distance to the destination are set with corresponding weight thresholds and calculated into a comprehensive reference index S. This step comprehensively considers multiple dimensions of ship navigation, making the assessment of the ship's navigation status more comprehensive and accurate.

[0015] S4: Construct a TBENet network learning model, input the sequence data of sequence data S' into the TBENet network learning model, and improve the prediction ability of the TBENet network learning model; The TBENet model combines the advantages of BiLSTM and BiGRU, and adopts a unique three-layer architecture that can effectively capture long-term dependencies and short-term changes in time series data. This combination not only ensures comprehensive data processing, but also maintains the integrity of information over a long period of time, thereby improving the accuracy and stability of predictions. At the same time, the trajectory curve predicted by the TBENet method is very consistent with the actual situation, and the prediction accuracy is better than other evaluation models.

[0016] S5: Test the prediction ability of the TBENet network learning model, optimize the parameters of the TBENet network learning model using the AdamW optimizer, and train the TBENet network learning model; The AdamW optimizer is used to optimize the parameters of the TBENet network learning model, which not only accelerates the training process of the model, but also effectively prevents the occurrence of overfitting and further improves the generalization ability of the model.

[0017] S6: The Bayesian optimization method is used to adjust the hyperparameters of the TBENet network learning model to obtain the optimal parameters of the TBENet network learning model. The optimal parameters are introduced into the TBENet network learning model to obtain the optimized TBENet network learning model, which makes the prediction accuracy of the ship's navigation trajectory higher.

[0018] The Bayesian optimization method was used to fine-tune the hyperparameters of the TBENet network learning model and obtain the optimal parameter configuration of the model. This step enables the model to achieve higher prediction accuracy while maintaining efficient operation, providing a strong guarantee for the accurate prediction of the ship's navigation trajectory.

[0019] The specific steps of step S2 are: S2.1: Use Lagrange interpolation method to supplement the monitoring data longitude set A, latitude set B, heading set C, speed set D and distance set E from the ship to the destination, and obtain longitude set A1, latitude set B1, heading set C1, speed set D1 and distance set E1 from the ship to the destination, specifically: Lagrange interpolation method approximates known data points by constructing polynomial functions, so that more accurate data values ​​can be calculated. In the long-term data collection process, errors may gradually accumulate, causing the data to deviate from the true value. Lagrange interpolation method can reduce this error accumulation to a certain extent, making the data closer to the actual situation. Lagrange interpolation method can ensure the accuracy and continuity of navigation data, so that the navigation route can be better planned and potential risks can be avoided.

[0020] S2.1.1: For n+1 known data points (X0, Y0), (X1, Y1), ..., (Xn , Y n ), Lagrange interpolation polynomial L Y (X) The expression is: Among them, X is the independent variable, Y i For X i The value of Y∈{A, B, C, D, E}, L i (X) is the basis polynomial; S2.1.2: Basic polynomial L i (X) Specifically: Among them, X i With X j is the independent variable with known data points, satisfying L i (X i )=1 and satisfy L i (X i )=0; S2.1.3: The calculated Lagrangian interpolation polynomial expression L A (X)、L B (X)、L C (X)、L D (X) and L E (X) The longitude set A1, latitude set B1, heading set C1, speed set D1 and distance set E1 from the ship to the destination are calculated and expressed as: Among them, Y1∈{A1, B1, C1, D1, E1}, Y new is the interpolation value, Y new ∈{A new , B new , C new , D new 、E new}; The Lagrange interpolation polynomial expression of longitude A is: Among them, A i is the longitude in X i The value at Li （A） (X) is the basic polynomial of longitude A; The Lagrange interpolation polynomial expression for latitude B is: Among them, B i is the latitude in X i The value at Li（B） (X) is the basis polynomial of latitude B; The Lagrange interpolation polynomial expression of the heading C is: Among them, C i The heading is in X i The value at Li （C） (X) is the basic polynomial of the heading C; The Lagrange interpolation polynomial expression of the speed D is: Among them, D i is the speed at X i The value at Li （D） (X) is the basic polynomial of the speed D; The Lagrange interpolation polynomial expression of the distance E between the ship and the destination is: Among them, E i The distance between the ship and the destination is X i The value at Li （E） (X) is the basic polynomial of the distance E between the ship and the destination; S2.2: Min-Max normalize the longitude set A1, latitude set B1, heading set C1, speed set D1 and the distance set E1 from the ship to the destination, scale the data to [0,1], and obtain the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 from the ship to the destination. The expression is: Among them, Y2∈{A2, B2, C2, D2, E2}; S2.3: Calculate the standard deviation of the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 between the ship and the destination, and standardize the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 between the ship and the destination to obtain the standardized longitude set A3, latitude set B3, heading set C3, speed set D3 and the distance set E3 between the ship and the destination, expressed as: Among them, Y` σ is the standard deviation, Y i is the i-th data in each set, Y3∈{A3, B3, C3, D3, E3}, It is the average value of A3, B3, C3, D3, and E3.

[0021] Step S3 is specifically as follows: in, is the average value in the longitude set A3, W A is the relative weight of the average value in the longitude set A3, is the average value in the latitude set B3, W B is the relative weight of the average value in the latitude set B3, is the average value in the heading set C3, W C is the relative weight of the average value in the heading set C3, is the average value in the speed set D3, W D is the relative weight of the average value in the speed set D3, is the average value of the distance set E3 between the ship and the destination, W E It is the weight relative to the average value in the distance set E3 between the ship and the destination.

[0022] The structural design of the TBENet model takes advantage of the complementary advantages of different network units and enhances the data fitting ability. In the model, each network unit is followed by a fully connected layer to enhance nonlinear processing and adaptability. In addition, each component in the network can have a unique structure tailored for different functions, including enhancing long-term and short-term features. This design enables the TBENet model to better adapt to complex ship navigation trajectory data and improve the reliability of prediction. The specific steps of step S4 are: S4.1: After the sequence data S' enters the first Bi-GRU layer, it is processed to obtain the sequence data H1. The activation function ReLU is used to process the sequence data H1 into the sequence data H'1, and the sequence data H'1 is transmitted to the first fully connected layer and the first Dropout layer. The expression is: S4.2: After receiving the sequence data H'1, the first fully connected layer and the first Dropout layer process the sequence data H'1 into sequence data H2, and the first Dropout layer processes the sequence data H2 into sequence data H'2, and transmits the sequence data H'2 to the second Bi-GRU layer. The expression is: S4.3: After receiving the sequence data H'2, the second Bi-GRU layer processes the sequence data H3, uses the activation function ReLU to process the sequence data H3 into sequence data H'3, and transmits the sequence data H'3 to the second fully connected layer and the second Dropout layer. The expression is: S4.4: After the second fully connected layer and the second Dropout layer receive the sequence data H'3, the second fully connected layer processes the sequence data H'3 into sequence data H4, the second Dropout layer processes the sequence data H4 into H'4, and transmits the sequence data H'4 to the third Bi-GRU layer. The expression is: S4.5: After receiving the sequence data H'4, the third Bi-GRU layer processes the sequence data H5, uses the activation function ReLU to process the sequence data H5 into sequence data H'5, and transmits the sequence data H'5 to the third fully connected layer and the third Dropout layer. The expression is: S4.6: After the third fully connected layer and the third Dropout layer receive the sequence data H'5, the second fully connected layer processes the sequence data H'5 into sequence data H6, and the third Dropout layer processes the sequence data H6 into H'6, and transmits the sequence data H'6 to the Bi-LSTM layer. The expression is: S4.7: After receiving the sequence data H'6, the Bi-LSTM layer processes the sequence data H'6 into sequence data H7, and transmits the sequence data H7 to the fourth fully connected layer. The expression is: S4.8: After receiving the sequence data H7, the fourth fully connected layer calculates and processes the sequence data H7 through the Sigmoid function to obtain the sequence data M, which is expressed as: The Sigmoid function maps the output value to the interval (0,1), indicating the probability of the deviation.

[0023] Step S4 is specifically as follows: S4.1: The processed sequence data M is evaluated for quality using binary cross entropy, expressed as: Among them, N is the total number of samples, Q t is the true label, which usually outputs 1 or 0, indicating whether the ship's trajectory has shifted at time t. t is the predicted value, which is expressed as the probability of the ship's trajectory deviation at the tth moment, and L is the loss amount; S4.2: Use the AdamW optimizer to optimize the parameters of the TBENet network learning model. After multiple iterations of optimization, the optimal TBENet network learning model is obtained. The specific optimization of the AdamW optimizer is as follows: Among them, θ z is the parameter of the zth step, η z is the learning rate, P z is the first-order moment estimate of the gradient, U z is the second-order moment estimate of the gradient, ε is a constant, to prevent division by zero, λ is the weight attenuation function; S4.3: The specific steps for training the TBENet network learning model are: S4.3.1: Initialize TBENet network learning model parameters; S4.3.2: For each epoch, include the following steps: Forward propagation: Calculate the predicted value M t and the loss amount L; Back propagation: calculate gradients and update TBENet network learning model parameters; Evaluate the performance of the TBENet network learning model and record the optimal TBENet network learning model parameters.

[0024] Step S6 specifically includes: S6.1: Select initial parameter space; S6.2: Train the surrogate model using the parameters in the initial parameter space to approximate the unknown objective function; S6.3: Based on the current proxy model, use the acquisition function to determine the next sampling point; S6.4: Evaluate the true objective function at the selected sampling points, add the newly obtained sampling points to the existing sampling point set to obtain an updated sampling point set, and retrain the proxy model using the updated sampling point set; S6.5: Repeat steps S6.3-S6.4 until the stopping condition is met, and a trained proxy model is obtained. The parameters in the trained proxy model are used as the optimal parameters of the TBENet network learning model.

[0025] In summary, the monitoring data collection and analysis method realizes comprehensive monitoring of the ship's navigation status by collecting monitoring data from multiple data sources of ships, including longitude, latitude, heading, speed and the distance of the ship from the destination. It comprehensively considers multiple dimensions of ship navigation, making the assessment of the ship's navigation status more comprehensive and accurate. Through accurate prediction of the ship's navigation trajectory, it can provide strong support for the ship's navigation safety and reduce potential risks during navigation. At the same time, accurate prediction results can also help optimize the ship's navigation route and speed, improve navigation efficiency and reduce operating costs.

[0026] It should be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0027] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A monitoring data collection and analysis method, characterized in that: The following steps are involved: S1: Collect monitoring data from multiple data sources of ships, including longitude set A, latitude set B, heading set C, speed set D, and distance set E from the ship to the destination; S2: pre-processing the collected monitoring data to obtain a longitude set A3, a latitude set B3, a heading set C3, a speed set D3, and a distance set E3 between the ship and the destination; S3: Set corresponding weights for the longitude set A3, latitude set B3, heading set C3, speed set D3 and distance set E3 from the ship to the destination, and calculate them as a comprehensive reference index S. The sequence data S' of the comprehensive reference index S is calculated, S' = [S1, S2, ..., S T ], T is the sequence length, S T is the comprehensive reference index S at the Tth moment; S4: Construct a TBENet network learning model, input the sequence data of sequence data S' into the TBENet network learning model, and improve the prediction ability of the TBENet network learning model; S5: Test the prediction ability of the TBENet network learning model, optimize the parameters of the TBENet network learning model using the AdamW optimizer, and train the TBENet network learning model; S6: The Bayesian optimization method is used to adjust the hyperparameters of the TBENet network learning model to obtain the optimal parameters of the TBENet network learning model. The optimal parameters are introduced into the TBENet network learning model to obtain the optimized TBENet network learning model, which makes the prediction accuracy of the ship's navigation trajectory higher.

2. A monitoring data collection and analysis method according to claim 1, characterized in that: The specific steps of step S2 are: S2.1: Use Lagrange interpolation method to supplement the monitoring data longitude set A, latitude set B, heading set C, speed set D and distance set E from the ship to the destination, and obtain longitude set A1, latitude set B1, heading set C1, speed set D1 and distance set E1 from the ship to the destination, specifically: S2.1.1: For n+1 known data points (X0, Y0), (X1, Y1), ..., (X n , Y n ), Lagrange interpolation polynomial L Y (X) The expression is: Among them, X is the independent variable, Y i For X i The value of Y∈{A, B, C, D, E}, L i (X) is the basis polynomial; S2.1.2: Basic polynomial L i (X) Specifically: Among them, X i With X j is the independent variable with known data points, satisfying L i (X i )=1 and satisfy L i (X i )=0; S2.1.3: The calculated Lagrangian interpolation polynomial expression L A (X)、L B (X)、L C (X)、L D (X) and L E (X) The longitude set A1, latitude set B1, heading set C1, speed set D1 and distance set E1 from the ship to the destination are calculated and expressed as: Among them, Y1∈{A1, B1, C1, D1, E1}, Y new is the interpolation value, Y new ∈{A new , B new , C new , D new 、E new }; S2.2: Min-Max normalize the longitude set A1, latitude set B1, heading set C1, speed set D1 and the distance set E1 from the ship to the destination, scale the data to [0,1], and obtain the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 from the ship to the destination. The expression is: Among them, Y2∈{A2, B2, C2, D2, E2}; S2.3: Calculate the standard deviation of the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 between the ship and the destination, and standardize the longitude set A2, latitude set B2, heading set C2, speed set D2 and the distance set E2 between the ship and the destination to obtain the standardized longitude set A3, latitude set B3, heading set C3, speed set D3 and the distance set E3 between the ship and the destination, expressed as: Among them, Y` σ is the standard deviation, Y i is the i-th data in each set, Y3∈{A3, B3, C3, D3, E3}, It is the average value of A3, B3, C3, D3, and E3.

3. The monitoring data collection and analysis method according to claim 1, characterized in that: The step S3 is specifically as follows: in, is the average value in the longitude set A3, W A is the relative weight of the average value in the longitude set A3, is the average value in the latitude set B3, W B is the relative weight of the average value in the latitude set B3, is the average value in the heading set C3, W C is the relative weight of the average value in the heading set C3, is the average value in the speed set D3, W D is the relative weight of the average value in the speed set D3, is the average value of the distance set E3 between the ship and the destination, W E It is the weight relative to the average value in the distance set E3 between the ship and the destination.

4. The monitoring data collection and analysis method according to claim 1 is characterized in that , the specific steps of step S4 are: S4.1: After the sequence data S' enters the first Bi-GRU layer, it is processed to obtain the sequence data H1. The activation function ReLU is used to process the sequence data H1 into the sequence data H'1, and the sequence data H'1 is transmitted to the first fully connected layer and the first Dropout layer. The expression is: S4.2: After receiving the sequence data H'1, the first fully connected layer and the first Dropout layer process the sequence data H'1 into sequence data H2, and the first Dropout layer processes the sequence data H2 into sequence data H'2, and transmits the sequence data H'2 to the second Bi-GRU layer. The expression is: S4.3: After receiving the sequence data H'2, the second Bi-GRU layer processes the sequence data H3, uses the activation function ReLU to process the sequence data H3 into sequence data H'3, and transmits the sequence data H'3 to the second fully connected layer and the second Dropout layer. The expression is: S4.4: After the second fully connected layer and the second Dropout layer receive the sequence data H'3, the second fully connected layer processes the sequence data H'3 into sequence data H4, the second Dropout layer processes the sequence data H4 into H'4, and transmits the sequence data H'4 to the third Bi-GRU layer. The expression is: S4.5: After receiving the sequence data H'4, the third Bi-GRU layer processes the sequence data H5, uses the activation function ReLU to process the sequence data H5 into sequence data H'5, and transmits the sequence data H'5 to the third fully connected layer and the third Dropout layer. The expression is: S4.6: After the third fully connected layer and the third Dropout layer receive the sequence data H'5, the second fully connected layer processes the sequence data H'5 into sequence data H6, and the third Dropout layer processes the sequence data H6 into H'6, and transmits the sequence data H'6 to the Bi-LSTM layer. The expression is: S4.7: After receiving the sequence data H'6, the Bi-LSTM layer processes the sequence data H'6 into sequence data H7, and transmits the sequence data H7 to the fourth fully connected layer. The expression is: S4.8: After receiving the sequence data H7, the fourth fully connected layer calculates and processes the sequence data H7 through the Sigmoid function to obtain the sequence data M, which is expressed as: 。 5. The monitoring data collection and analysis method according to claim 1, characterized in that: The step S4 is specifically as follows: S4.1: The processed sequence data M is evaluated for quality using binary cross entropy, expressed as: Among them, N is the total number of samples, Q t is the true label, indicating whether the ship's trajectory has shifted at time t, M t is the predicted value, L is the loss amount; S4.2: Use the AdamW optimizer to optimize the parameters of the TBENet network learning model. After multiple iterations of optimization, the optimal TBENet network learning model is obtained. The specific optimization of the AdamW optimizer is as follows: Among them, θ z is the parameter of the zth step, η z is the learning rate, P z is the first-order moment estimate of the gradient, U z is the second-order moment estimate of the gradient, ε is a constant, and λ is the weight decay function; S4.3: The specific steps for training the TBENet network learning model are: S4.3.1: Initialize TBENet network learning model parameters; S4.3.2: For each epoch, include the following steps: Forward propagation: Calculate the predicted value M t and the loss amount L; Back propagation: calculate gradients and update TBENet network learning model parameters; Evaluate the performance of the TBENet network learning model and record the optimal TBENet network learning model parameters.

6. The monitoring data collection and analysis method according to claim 1, characterized in that: The specific steps of step S6 are as follows: S6.1: Select initial parameter space; S6.2: Train the surrogate model using the parameters in the initial parameter space to approximate the unknown objective function; S6.3: Based on the current proxy model, use the acquisition function to determine the next sampling point; S6.4: Evaluate the true objective function at the selected sampling points, add the newly obtained sampling points to the existing sampling point set to obtain an updated sampling point set, and retrain the proxy model using the updated sampling point set; S6.5: Repeat steps S6.3-S6.4 until the stopping condition is met, and a trained proxy model is obtained. The parameters in the trained proxy model are used as the optimal parameters of the TBENet network learning model.

Images