A method, system, device and storage medium for predicting a ship's speed over ground

Abstract

The application discloses a kind of ship speed on ground prediction method, system, device and storage medium.The method is by obtaining ship pool experiment data, historical ship navigation data and ship navigation environment data;Then, according to the ship pool experiment data, construct ship still water speed prediction model;And according to the historical ship navigation data and the ship navigation environment data, construct ship speed on ground prediction model;Then, collect ship navigation data and navigation environment data;Again the ship navigation data and the navigation environment data are input into the ship speed on ground prediction model after training is completed, and the predicted ship speed on ground is obtained.The method realizes the prediction of ship speed on ground under complex environmental factors, provides data support for subsequent navigation, collision avoidance and other operations, and can be used for post-experimental analysis.The application can be widely applied in speed prediction technical field.

Description

Technical Field

[0001] This invention relates to the field of speed prediction technology, and in particular to a method, system, device, and storage medium for predicting the speed of a ship relative to the ground. Background Technology

[0002] In recent years, with the rapid development of sensor and computer communication technologies, the accuracy of ship navigation information and marine hydrological environment information has become increasingly higher, and it can be stored in storage media for a long time. This information can provide a data foundation for improving ship monitoring efficiency and preventing ship accidents.

[0003] In the shipping industry, it is usually necessary to estimate the ETA of a ship to ensure timely delivery of goods. According to the ETA calculation formula, it is necessary to predict the ship's speed above ground under a specific weather and water temperature environment. However, for the marine environment, the coupling effect of environmental factors is complex, and its impact on the ship's speed above ground is difficult to quantify.

[0004] In summary, the problems existing in the relevant technologies urgently need to be solved. Summary of the Invention

[0005] The purpose of this invention is to at least partially solve one of the technical problems existing in the prior art.

[0006] Therefore, one object of the present invention is to provide a method, system, device and storage medium for predicting the speed of a ship relative to the ground.

[0007] To achieve the above-mentioned technical objectives, the technical solutions adopted in the embodiments of the present invention include:

[0008] On one hand, embodiments of the present invention provide a method for predicting the speed of a ship relative to ground, comprising the following steps:

[0009] Acquire ship tank test data, historical ship navigation data, and ship navigation environment data;

[0010] Based on the ship tank test data, a ship still water speed prediction model is constructed.

[0011] Based on the historical ship navigation data and the ship navigation environment data, a ship ground speed prediction model is constructed.

[0012] Collect ship navigation data and navigation environment data;

[0013] The ship's navigation data and the navigation environment data are input into the ship's ground speed prediction model after training to obtain the predicted ship's ground speed.

[0014] Furthermore, the ship tank test data includes the ship's main engine power, ship's draft, and ship's still water speed; the historical ship navigation data includes the historical ship's main engine power, historical ship's draft, historical ship's land speed, and historical ship's water speed.

[0015] Furthermore, the ship navigation environment data includes significant wave height, mean wave period, relative wave angle, wind speed, relative airflow angle, current speed, and relative current angle.

[0016] Furthermore, the step of constructing a ship still water speed prediction model based on the ship tank test data includes:

[0017] Based on the experimental data from the ship's water tank, a partial differential equation was constructed to describe the balance between the ship's main engine power, the ship's still water speed, and the ship's draft.

[0018] Solve the solution set within the data boundary of the ship tank experiment based on the partial differential equation.

[0019] Based on the solution set of the partial differential equation, the functional relationship between the ship's still water speed and the ship's main engine power and draft is obtained.

[0020] The ship's still water speed prediction model is constructed based on the aforementioned functional relationship.

[0021] Furthermore, the step of constructing a ship's ground speed prediction model based on the historical ship navigation data and ship navigation environment data includes:

[0022] Based on the historical ship navigation data, the difference between the ship's speed in the water and its still water speed is obtained using the ship's still water speed prediction model.

[0023] Based on the historical ship navigation data, the ship navigation environment data, and the difference, a ship ground speed prediction model is constructed.

[0024] Furthermore, the partial differential equation is solved using a white-box model based on a physical mechanism neural network, through the following steps:

[0025] Based on the partial differential equations, a physical mechanism neural network model is constructed.

[0026] The physical mechanism neural network model was trained based on the experimental data from the ship's water tank.

[0027] The model training task is completed when the physical mechanism neural network model satisfies the partial differential equation constraint target termination condition.

[0028] Furthermore, the ship-to-ground speed prediction model is trained through the following steps: data preprocessing of the collected ship navigation data and navigation environment data;

[0029] The difference between the ship's speed in water and its speed in still water is obtained through a physical mechanism neural network model;

[0030] The difference is used as new feature data and input into the ground velocity prediction model in parallel.

[0031] The ground speed prediction model is trained based on the ship navigation data, the navigation environment data, and the difference.

[0032] The model training task is completed when the ground velocity prediction model meets the target termination condition.

[0033] On the other hand, embodiments of the present invention propose a system for predicting the speed of a ship relative to the ground, comprising:

[0034] The first module is used to acquire ship tank test data, historical ship navigation data, and navigation environment data.

[0035] The second module is used to construct a prediction model for the ship's still water speed based on the experimental data from the ship's water tank.

[0036] The third module is used to construct a ship ground speed prediction model based on the historical ship navigation data and the ship navigation environment data.

[0037] The fourth module is used to collect ship navigation data and navigation environment data;

[0038] The fifth module is used to input the ship navigation data and the navigation environment data into the ship-to-ground speed prediction model after training, so as to obtain the predicted ship-to-ground speed.

[0039] On the other hand, embodiments of the present invention provide a device for predicting the speed of a ship relative to the ground, comprising:

[0040] At least one processor;

[0041] At least one memory for storing at least one program;

[0042] When the at least one program is executed by the at least one processor, the at least one processor implements the method for predicting the ship's speed relative to the ground.

[0043] On the other hand, embodiments of the present invention provide a storage medium storing processor-executable instructions, which, when executed by a processor, are used to implement the method for predicting the ship's speed relative to the ground.

[0044] This invention discloses a method for predicting the speed of a ship relative to the ground, which has the following beneficial effects:

[0045] This embodiment acquires ship tank experimental data, historical ship navigation data, and ship navigation environment data. Then, based on the ship tank experimental data, a ship still water speed prediction model is constructed; and based on the historical ship navigation data and the ship navigation environment data, a ship ground speed prediction model is constructed. Next, ship navigation data and navigation environment data are collected; then, the ship navigation data and navigation environment data are input into the trained ship ground speed prediction model to obtain the predicted ship ground speed. This method constructs a white-box model of ship still water speed and a black-box model of ship ground speed, and then integrates the white-box and black-box models to obtain the ship ground speed prediction model. This enables the prediction of ship ground speed under complex environmental factors, providing data support for subsequent navigation, collision avoidance, and other operations, and can also be used for post-experimental analysis. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following description is provided with accompanying drawings of the relevant technical solutions in the embodiments of the present invention or the prior art. It should be understood that the accompanying drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0047] Figure 1 This is a schematic diagram illustrating the implementation environment of a method for predicting ship speed above ground provided in this application embodiment;

[0048] Figure 2 A flowchart illustrating a method for predicting the speed of a ship relative to the ground, provided in an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of a ship ground speed prediction system provided in an embodiment of the present invention;

[0050] Figure 4 This is a schematic diagram of a ship ground speed prediction device provided in an embodiment of the present invention. Detailed Implementation

[0051] This section will describe in detail specific embodiments of the present invention. Preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the drawings is to supplement the textual description with graphics, so that people can intuitively and vividly understand each technical feature and overall technical solution of the present invention, but they should not be construed as limiting the scope of protection of the present invention.

[0052] In the description of the embodiments of the present invention, "several" means one or more, "multiple" means two or more, "greater than," "less than," "exceeding," etc. are understood to exclude the number itself, while "above," "below," "within," etc. are understood to include the number itself. "At least one" refers to one or more, and "at least one of the following" and similar expressions refer to any combination of these items, including any combination of single or multiple items. If "first," "second," etc., are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance or implicitly indicating the number of indicated technical features or the order of the indicated technical features.

[0053] It should be noted that the terms "setting," "installing," and "connecting" in the embodiments of this invention should be interpreted broadly. Those skilled in the art can reasonably determine the specific meaning of the above terms in the embodiments of this invention in conjunction with the specific content of the technical solution. For example, the term "connection" can be a mechanical connection, an electrical connection, or a connection that allows for mutual communication; it can be a direct connection or an indirect connection through an intermediate medium.

[0054] In the description of the embodiments of the present invention, the references to "one embodiment / implementation," "another embodiment / implementation," or "some embodiments / implementations," "in the above embodiments / implementations," etc., refer to specific features, structures, materials, or characteristics described in connection with embodiments or examples that are included in at least two embodiments or implementations of this disclosure. In this disclosure, the illustrative expressions of the above terms do not necessarily refer to the same illustrative embodiment or implementation. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or implementations.

[0055] It should be noted that the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0056] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0057] SOG (Speed ​​Over Ground) refers to the speed at which a ship travels over the seabed under the combined influence of wind and current; it is called "ground speed." Actual speed is the vector sum of the water velocity and the current velocity.

[0058] STW (Speed ​​Toward Water) is the speed of a ship relative to the water. It is independent of the current speed but related to the wind speed; the ship's speed is reflected in its speed relative to the water under the influence of wind.

[0059] ETA (Estimated Time of Arrival) refers to the estimated time of arrival at the port.

[0060] It refers to the estimated time of arrival (at the pilot station or anchorage), not the time of arrival at the berth.

[0061] PINNs (Physics-informed Neural Networks) are a class of neural networks used to solve supervised learning tasks while respecting any given physical laws described by general nonlinear partial differential equations.

[0062] XGBoost (an improved gradient boosting algorithm) is an improvement on the gradient boosting algorithm. It uses Newton's method to solve the extreme value of the loss function, expands the loss function to the second order using Taylor expansion, and adds a regularization term to the loss function.

[0063] There are many solutions for ship speed prediction, mainly divided into methods based on physics or empirical formulas and methods based on data-driven models, but both have certain drawbacks. Firstly, with the increasing size of ships, their tonnage has gradually exceeded the limitations of physical models or empirical formulas, resulting in larger errors in the obtained results. With the advent of the big data era, data-driven models have been developed to predict ship speed using relevant environmental data. However, because many machine learning algorithms require that the test and training datasets be independent and identically distributed, the prediction results of a new dataset exhibit significant uncertainty.

[0064] To address this, this application proposes a method, system, device, and storage medium for predicting ship speed above ground. The method involves acquiring ship tank experimental data, historical ship navigation data, and ship navigation environment data; constructing a ship still water speed prediction model based on the tank experimental data; then constructing a ship speed above ground prediction model; subsequently, integrating the still water speed prediction model and the ship speed above ground prediction model to obtain a ship speed above ground prediction model; subsequently, collecting ship navigation data and navigation environment data; and inputting the ship navigation data and the navigation environment data into the trained ship speed above ground prediction model to obtain the predicted ship speed above ground. This method constructs a white-box model of ship speed above water and a black-box model of ship speed above ground, and then integrates the white-box and black-box models to obtain a target gray-box prediction model, thereby enabling the prediction of ship speed above ground under complex environmental factors. This provides data support for subsequent navigation, collision avoidance, and other operations, and can also be used for post-experimental analysis.

[0065] Figure 1This is a schematic diagram illustrating the implementation environment of a method for predicting ship speed relative to ground provided in this application. (Refer to...) Figure 1 The main hardware and software components of this implementation environment include an operating terminal 101 and a server 102, which are communicatively connected. The method for predicting the ship's ground speed can be configured to execute independently on the operating terminal 101, independently on the server 102, or based on the interaction between the two. The appropriate choice can be made according to the actual application, and this embodiment does not impose specific limitations on this. Furthermore, the operating terminal 101 and the server 102 can be nodes in a blockchain, and this embodiment does not impose specific limitations on this either.

[0066] Specifically, the operating terminal 101 in this application may include, but is not limited to, any one or more of the following: smartwatch, smartphone, computer, personal digital assistant (PDA), smart voice interaction device, smart home appliance, or in-vehicle terminal. The server 102 may be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The operating terminal 101 and the server 102 can establish a communication connection via a wireless network or wired network. This wireless network or wired network uses standard communication technologies and / or protocols. The network can be the Internet or any other network, including but not limited to any combination of Local Area Network (LAN), Metropolitan Area Network (MAN), Wide Area Network (WAN), mobile, wired or wireless network, private network, or virtual private network.

[0067] Figure 2 This is a flowchart illustrating a method for predicting a ship's speed relative to ground, as provided in an embodiment of this application. The method can be executed by at least one of an operating terminal or a server. Figure 2 The method for predicting the ship's ground speed, configured and executed on the operating terminal, is used as an example for illustration. (Refer to...) Figure 2 The method for predicting the ship's speed relative to the ground includes, but is not limited to, steps 110 to 160.

[0068] Step 110: Obtain ship tank test data, historical ship navigation data, and ship navigation environment data.

[0069] In this step, historical ship navigation data includes data such as ship main engine power, ship draft, ship speed over land, and ship speed over water. Specifically, in this embodiment, there are no restrictions on the channels for obtaining historical ship navigation data. This historical ship navigation data can be obtained directly through a data transmission interface or obtained from other electronic devices and computer systems via remote communication transmission.

[0070] Step 120: Based on the experimental data from the ship's water tank, construct a prediction model for the ship's still water speed.

[0071] Step 130: Construct a ship ground speed prediction model based on the historical ship navigation data and the ship navigation environment data.

[0072] In this embodiment, the ship's ground speed prediction model can be established using two typical modeling architectures: serial modeling and parallel modeling. Serial modeling requires configuring two or more models in a series, including at least one white-box model and one black-box model. Here, the serial approach architecture configures two models consecutively. The first is a speed-power white-box model, i.e., a ship's still-water speed prediction model, which derives the ship's expected still-water speed based on measured propulsion power and draft. The data is then input into the black-box model, a regression model used to predict the ship's ground speed. The difference between the serial and parallel architectures is minimal. Parallel modeling requires modeling a speed-power white-box model, training a black-box model by reducing the residual between the white-box model's output and the expected output, and then integrating the white-box and black-box models in parallel.

[0073] Step 140: Collect ship navigation data and navigation environment data.

[0074] In this step, the ship navigation data includes the ship's ground speed data (SOG), and the navigation environment data includes significant wave height, mean wave period, relative wave angle, wind speed, relative airflow angle, current speed, and relative current angle. Specifically, in this embodiment, there are no restrictions on the acquisition channels for the ship navigation data and navigation environment data. The data can be obtained directly through a data transmission interface or obtained remotely from other electronic devices and computer systems.

[0075] Step 150: Input the ship navigation data and the navigation environment data into the ship-to-ground speed prediction model after training to obtain the predicted ship-to-ground speed.

[0076] This embodiment acquires historical ship navigation data; based on this data, it constructs a ship still water speed prediction model; then, it constructs a ship ground speed prediction model; next, it integrates the still water speed prediction model and the ground speed prediction model to obtain a ship ground speed prediction model; subsequently, it collects ship navigation data and navigation environment data; and inputs these data into the trained ship ground speed prediction model to obtain the predicted ship ground speed. This method constructs a white-box model of ship water speed and a black-box model of ship ground speed, using serial-parallel techniques to obtain the ship ground speed prediction model. This enables the prediction of ship ground speed under complex environmental factors, providing data support for subsequent navigation, collision avoidance, and other operations, and can also be used for post-experimental analysis.

[0077] As a further optional implementation, the ship tank test data includes the ship's main engine power, ship's draft, and ship's still water speed.

[0078] As a further optional implementation, the historical ship navigation data includes the ship's main engine power, ship's draft, ship's land speed, and ship's water speed.

[0079] As a further optional implementation, the step of constructing a ship still water speed prediction model based on the ship tank test data includes:

[0080] Based on the experimental data from the ship's water tank, the partial differential equations describing the balance relationship between the ship's main engine power p, the ship's still water speed v, and the ship's draft d were obtained by fitting and refining the speed-power curve under typical operating conditions:

[0081]

[0082] q(v)=β1-β2v-β3v 2 .

[0083] Wherein, α, β1, β2, and β3 are undetermined coefficients determined by fitting the speed-power curve and the formula.

[0084] Based on the partial differential equation, the solution set within the boundary of the ship tank experiment data is solved using an artificial neural network;

[0085] Specifically, in this embodiment, a pool experiment is conducted to obtain the ship's still water speed-power curve under typical draft, deriving a partial differential equation describing the balance between the ship's main engine power, still water speed, and draft. Next, the partial differential equation is solved forward using PINNs to obtain the solution set. Then, a reverse solution is performed using an over-the-loop neural network to derive the ship's still water speed function, which can be used to characterize the relationship between the ship's main engine power, draft, and still water speed.

[0086] As an optional implementation, the navigation environment data includes significant wave height, mean wave period, relative wave angle, wind speed, relative airflow angle, current speed, and relative current angle.

[0087] As a further optional implementation, after the step of collecting ship navigation data and navigation environment data, the prediction method further includes the following steps:

[0088] Data preprocessing is performed on the collected ship navigation data and navigation environment data;

[0089] The ship's ground speed is extracted from the ship's navigation data, and the difference between the ship's still water speed and the ship's ground speed is obtained. This difference is used as input to the ship's ground speed prediction model.

[0090] Specifically, the collected ship navigation data includes values ​​for different navigation states such as anchored, drifting, moored, and underway. The data collected during navigation needs to be filtered out for subsequent preprocessing. Simultaneously, since ships inevitably accelerate, decelerate, and change speed while at sea, this maneuvering data, as well as obvious outliers, missing values, and duplicates, will be filtered out through preprocessing. After extracting the ship's navigation data, the difference between the ship's still water speed and its ground speed is calculated, which will then be used as input for the ship's ground speed prediction model.

[0091] As a further optional implementation, the ship's ground speed prediction model is trained through the following steps:

[0092] Data preprocessing is performed on the collected ship navigation data and navigation environment data;

[0093] The difference between the ship's speed relative to water and its speed in still water is obtained through a physical mechanism neural network model; by

[0094] The difference is used as new feature data and input into the ground velocity prediction model in parallel.

[0095] The ground speed prediction model is trained based on the ship navigation data, the navigation environment data, and the difference.

[0096] The model training task is completed when the ground velocity prediction model meets the target termination condition.

[0097] According to some embodiments of the present invention, the modeling process of the gray-box prediction model integrating the PINNs-based white-box model and the XGBoost-based black-box model is as follows:

[0098] The gray-box prediction model is built through two typical modeling sequences. Modeling the process using a serial approach requires configuring two or more models within a series, including at least one white-box model and one black-box model. Here, the serial approach architecture involves consecutively configuring two models. The first is a speed-power white-box model, which derives the ship's expected speed in still water based on measured propulsion power and draft. The data is then fed into the black-box model. The black-box model is a regression model used to predict the ship's speed relative to land.

[0099] In this embodiment, after the white-box model obtains a training dataset through random sampling, the training dataset can be input into the initialized artificial neural network model for training. Specifically, after inputting the data from the training dataset into the initialized ship-to-ground speed prediction model, the model outputs the recognition result, i.e., the prediction result. The accuracy of the recognition model can be evaluated based on whether the prediction result satisfies the aforementioned partial differential equation. Based on the training loss value, the backpropagation algorithm is used to update the model parameters. After several iterations, a well-trained ship-to-ground speed prediction model can be obtained. Specifically, the number of iterations can be preset, or training can be considered complete when the test set reaches the required accuracy.

[0100] In this embodiment, after the black-box model obtains a training dataset through random sampling, the training dataset can be input into the initialized ship-to-ground speed prediction model for training. Specifically, after inputting the data from the training dataset into the initialized ship-to-ground speed prediction model, the model outputs the recognition result, i.e., the prediction result. The accuracy of the recognition model's prediction can be evaluated based on the prediction result and the aforementioned ground speed measurement values. Parallel tree optimization is used to update the model's parameters. Specifically, the number of iterations can be preset, or training is considered complete when the test set reaches the required accuracy.

[0101] Reference Figure 3 An embodiment of the present invention proposes a ship ground speed prediction system, comprising:

[0102] The first module 301 is used to acquire ship tank test data, historical ship navigation data and ship navigation environment data;

[0103] The second module 302 is used to construct a ship still water speed prediction model based on the ship tank test data.

[0104] The third module 303 is used to construct a ship ground speed prediction model based on the historical ship navigation data and the ship navigation environment data.

[0105] Module 304 is used to collect ship navigation data and navigation environment data;

[0106] The fifth module 305 is used to input the ship's navigation data and the navigation environment data into the trained ship-to-ground speed prediction model to obtain the predicted ship-to-ground speed.

[0107] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0108] Reference Figure 4 This invention provides a device for predicting the speed of a ship relative to the ground, comprising:

[0109] At least one processor 401;

[0110] At least one memory 402 is used to store at least one program;

[0111] When the at least one program is executed by the at least one processor 301, the at least one processor 401 performs... Figure 2 The method for predicting ship speed relative to ground is shown.

[0112] The content of the above method embodiments is applicable to the device embodiments. The specific functions implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0113] This invention also provides a storage medium storing processor-executable instructions, which, when executed by a processor, are used to implement... Figure 2 The method for predicting the speed of a ship relative to the ground is shown.

[0114] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for predicting the speed of a ship relative to ground, characterized in that, Includes the following steps: Acquire ship tank test data, historical ship navigation data, and ship navigation environment data; Based on the ship tank test data, a ship still water speed prediction model is constructed, including: constructing a partial differential equation describing the balance relationship between ship main engine power, ship still water speed, and ship draft based on the ship tank test data; constructing a physical mechanism neural network model based on the partial differential equation; training the physical mechanism neural network model based on the ship tank test data; and completing the model training task in response to the physical mechanism neural network model satisfying the partial differential equation constraint target termination condition. Based on the historical ship navigation data and the ship navigation environment data, a ship ground speed prediction model is constructed, including: based on the historical ship navigation data, obtaining the difference between the ship's water speed and its still water speed through the ship still water speed prediction model; and inputting the difference as new feature data into the ground speed prediction model in parallel. The ground speed prediction model is trained based on the historical ship navigation data, the ship navigation environment data, and the difference; the model training task is completed when the ground speed prediction model meets the target termination condition. Collect ship navigation data and navigation environment data; The ship's navigation data and the navigation environment data are input into the ship's ground speed prediction model after training to obtain the predicted ship's ground speed.

2. The method for predicting ship speed above ground according to claim 1, characterized in that, The ship tank test data includes the ship's main engine power, ship's draft, and ship's still water speed; the historical ship navigation data includes the historical ship's main engine power, historical ship's draft, historical ship's land speed, and historical ship's water speed.

3. The method for predicting ship speed above ground according to claim 1, characterized in that, The ship navigation environment data includes significant wave height, average wave period, relative wave angle, wind speed, relative airflow angle, current speed, and relative current angle.

4. The method for predicting ship speed above ground according to claim 1, characterized in that, The step of constructing a ship still water speed prediction model based on the ship tank test data includes: Based on the experimental data from the ship's water tank, a partial differential equation was constructed to describe the balance between the ship's main engine power, the ship's still water speed, and the ship's draft. Solve the solution set within the data boundary of the ship tank experiment based on the partial differential equation. Based on the solution set of the partial differential equation, the functional relationship between the ship's still water speed and the ship's main engine power and draft is obtained. The ship's still water speed prediction model is constructed based on the aforementioned functional relationship.

5. The method for predicting ship speed relative to ground according to any one of claims 1, characterized in that, The ship's ground speed prediction model is trained through the following steps: Data preprocessing is performed on the collected ship navigation data and navigation environment data; The difference between the ship's speed in water and its speed in still water is obtained through a physical mechanism neural network model; The difference is used as new feature data and input into the ground velocity prediction model in parallel. The ground speed prediction model is trained based on the ship navigation data, the navigation environment data, and the difference. The model training task is completed when the ground velocity prediction model meets the target termination condition.

6. A system for predicting the speed of a ship relative to ground, characterized in that, include: The first module is used to acquire ship tank test data, historical ship navigation data, and ship navigation environment data. The second module is used to construct a ship still water speed prediction model based on the ship tank experimental data. Specifically, the second module is used to: construct partial differential equations describing the balance between ship main engine power, ship still water speed, and ship draft based on the ship tank experimental data; construct a physical mechanism neural network model based on the partial differential equations; train the physical mechanism neural network model based on the ship tank experimental data; and complete the model training task in response to the physical mechanism neural network model satisfying the partial differential equation constraint termination condition. The third module is used to construct a ship ground speed prediction model based on the historical ship navigation data and the ship navigation environment data. Specifically, the third module is used to: obtain the difference between the ship's water speed and the ship's still water speed based on the historical ship navigation data and the ship still water speed prediction model; and input the difference as new feature data into the ground speed prediction model in parallel. The ground speed prediction model is trained based on the historical ship navigation data, the ship navigation environment data, and the difference; the model training task is completed when the ground speed prediction model meets the target termination condition. The fourth module is used to collect ship navigation data and navigation environment data; The fifth module is used to input the ship navigation data and the navigation environment data into the ship-to-ground speed prediction model after training, so as to obtain the predicted ship-to-ground speed.

7. A device for predicting the speed of a ship relative to ground, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method for predicting the ship's ground speed as described in any one of claims 1-5.

8. A computer-readable storage medium storing processor-executable instructions, characterized in that, The processor-executable instructions, when executed by the processor, are used to implement the method for predicting the ship's ground speed as described in any one of claims 1-5.

Images