Drag reduction prediction model design method based on longitudinal heterogeneous ship formation and related device

Abstract

The invention relates to a longitudinal heterogeneous ship formation-based drag reduction prediction model design method and a related device, and belongs to the technical field of ship engineering and fluid mechanics, and the method comprises the steps: setting longitudinal heterogeneous ship formation numerical simulation parameters in preset simulation software; designing numerical simulation scenes of different formation configurations by adopting a control variable method, determining ship resistance coefficients of longitudinal heterogeneous ship formations under each formation configuration, and analyzing action rules of bow flow field velocity and ship traveling waves on shear resistance and differential pressure resistance; based on the action rule, a mixed graph neural network prediction model is adopted to predict formation resistance of a to-be-formed ship group, and the mixed graph neural network prediction model comprises a collaborative architecture coded through a GATv2 dynamic attention mechanism and GINE geometric features. According to the method, the resistance reduction effect of the longitudinal heterogeneous ship formation can be predicted and optimized, and a scientific basis and an intelligent decision tool are provided for ship formation configuration.

Description

Technical Field

[0001] The present invention relates to the technical field of ship engineering and fluid mechanics, and in particular to a drag reduction prediction model design method based on a longitudinal heterogeneous ship formation and related devices. Background Art

[0002] With the acceleration of globalization, shipping has become a core pillar of international trade. Over 90% of global commercial transport flows take place by sea, making its position in cargo transportation irreplaceable. However, the rapid development of the shipping industry has also been accompanied by significant environmental challenges.

[0003] According to data from the International Maritime Organization (IMO), in 2018, the shipping industry accounted for approximately 2.9% of global carbon dioxide emissions, while sulfur oxides and nitrogen oxides accounted for 11% and 15%, respectively. Excessive carbon dioxide emissions not only exacerbate global warming but also lead to more frequent extreme weather events and rising sea levels, threatening the balance of ecosystems. Sulfur oxides and nitrogen oxides, on the other hand, cause acid rain and air pollution, posing a serious threat to soil, vegetation, and human health. Therefore, reducing ship resistance and thereby achieving energy conservation and emission reduction in shipping has become a key issue in the development of green transportation. Existing technologies typically achieve energy conservation and emission reduction in shipping through fuel substitution and exhaust gas aftertreatment, but these approaches cannot fundamentally reduce shipping's energy consumption. Summary of the Invention

[0004] Therefore, it is necessary to provide a drag reduction prediction model design method and related devices based on longitudinal heterogeneous ship formations, so as to solve the problem of high energy consumption and emissions of existing shipping ships with a new idea that does not require modification.

[0005] In order to address the above challenges and achieve the goal of reducing drag and energy consumption, in a first aspect, the present invention provides a method for designing a drag reduction prediction model based on a longitudinal heterogeneous ship formation, comprising: Setting the numerical simulation parameters of the longitudinal heterogeneous ship formation in the preset simulation software; Using the control variable method to design numerical simulation scenarios for different formation configurations, the ship resistance coefficients of longitudinally heterogeneous ship formations were determined for each formation configuration, including the number of ships in the formation, speed, spacing, and ship type. Based on the ship resistance coefficient of longitudinal heterogeneous ship formations under different formation configurations, the effects of bow flow field velocity and ship waves on shear resistance and pressure resistance are analyzed; Based on the action law, a hybrid graph neural network prediction model is used to predict the formation resistance of the group of ships to be formed, wherein the hybrid graph neural network prediction model includes a collaborative architecture through the GATv2 dynamic attention mechanism and the GINE geometric feature encoding.

[0006] In one possible implementation, setting numerical simulation parameters for a longitudinal heterogeneous ship formation in a preset simulation software includes: In the STAR-CCM+ simulation environment, eight monitoring points arranged in a row are set in front of the ship, with each monitoring point spaced 0.5 m apart, to collect data on the bow flow field. Forty monitoring points are set behind the last ship in the formation, with each monitoring point spaced 0.5 m apart, to collect data on changes in ship-borne waves.

[0007] In one possible implementation, a control variable method is used to design numerical simulation scenarios for different formation configurations, including: Under the condition that the ship speed and the ship spacing are the same, different numbers of ships are configured for the longitudinal heterogeneous ship formation model; Under the condition that the number of ships and the distance between ships are the same, different ship speeds are configured for the longitudinal heterogeneous ship formation model; When the number of ships and the speed of ships are the same, different ship spacings are configured for the longitudinal heterogeneous ship formation model.

[0008] In one possible implementation, the effects of bow flow velocity and ship waves on shear resistance and pressure resistance are analyzed based on the ship resistance coefficients of longitudinally heterogeneous ship formations under different formation configurations, including: During the STAR-CCM+ simulation process, the changes in the bow flow field velocity and the ship wave height over time are statistically analyzed through monitoring points, and the average values ​​of the changes in the bow flow field velocity and the ship wave height over time are calculated and displayed in the form of a bar graph. At the same time, the wave height, wavelength and period of the ship wave are statistically analyzed, and the monitoring point data are displayed in a line graph, which is then interpolated.

[0009] In one possible implementation, analyzing the effects of bow flow velocity and ship waves on shear resistance and pressure resistance includes: During the STAR-CCM+ simulation process, the changes in the bow flow field velocity and ship wave height over time are statistically analyzed through monitoring points, and their average values ​​are calculated and displayed in the form of a bar graph. At the same time, the ship wave height, wavelength and period are statistically analyzed, and the monitoring point data are displayed in a line graph, and the curve is smoothed through interpolation.

[0010] In a possible implementation, based on the action rule, a hybrid graph neural network prediction model is used to predict the formation resistance of the group of ships to be formed, including: The hybrid graph neural network prediction model includes a parallel architecture of GATv2 layer, PNA layer, and GINE layer, which is used for multi-level feature extraction and analysis of model prediction accuracy.

[0011] In one possible implementation, analyzing the prediction accuracy of the model includes: Three data enhancement measures are taken to address the problem of scarce CFD simulation data, and the prediction accuracy of different formation sizes is quantitatively analyzed through heat maps.

[0012] In a second aspect, the present invention further provides an electronic device comprising a memory and a processor, wherein: Memory, used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the method for designing a drag reduction prediction model based on a longitudinal heterogeneous ship formation in any of the above embodiments.

[0013] In a third aspect, the present invention also provides a computer-readable storage medium for storing computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the method for designing a drag reduction prediction model based on a longitudinal heterogeneous ship formation of any of the above-mentioned embodiments.

[0014] The beneficial effect of the present invention is that by combining numerical simulation and machine learning methods, it can efficiently predict and optimize the drag reduction effect of longitudinal heterogeneous ship formations, significantly improve the navigation efficiency of the formation, and provide a scientific basis for the configuration of ship formations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings used in the description of the embodiments. The drawings are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A schematic flow chart of a method for designing a drag reduction prediction model based on a longitudinal heterogeneous ship formation provided by an embodiment of the present invention; Figure 2 A schematic diagram of a longitudinal heterogeneous ship formation model provided by an embodiment of the present invention; Figure 3 A schematic diagram of a hull simulation grid provided by an embodiment of the present invention; Figure 4 A diagram showing the ship wave and flow field detection point settings provided by an embodiment of the present invention; Figure 5 A diagram of the wave patterns of ships with different numbers of formations provided by an embodiment of the present invention; Figure 6 A total resistance diagram of different numbers of ship formations provided by an embodiment of the present invention; Figure 7 Flow field and resistance analysis diagram of ship formation at different speeds provided by an embodiment of the present invention; Figure 8Six heterogeneous ship formation navigation posture diagrams provided by the embodiment of the present invention; Figure 9 The hybrid GNN structure provided by the embodiment of the present invention; Figure 10 Loss curve of the hybrid GNN provided by the embodiment of the present invention; Figure 11 Ship formation error distribution provided by an embodiment of the present invention; Figure 12 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, and are not used to limit the scope of the present invention.

[0018] The terms "first," "second," and so on, used in the embodiments of the present invention are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, technical features designated as "first" or "second" may explicitly or implicitly include at least one such feature.

[0019] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0020] A specific embodiment of the present invention, as Figure 1 As shown, a drag reduction prediction model design method based on longitudinal heterogeneous ship formation is disclosed, including: S101, setting numerical simulation parameters of the longitudinal heterogeneous ship formation in a preset simulation software.

[0021] In the embodiment of the present invention, a longitudinal heterogeneous ship formation model is constructed using SolidWorks. Taking three ships as an example, the first two are KCS ships and the rear one is a SERIES60 ship. Figure 2 In a formation, the first ship is the lead ship (ld), and the remaining ships are followers (fw1, fw2, ...). The KCS length is denoted as Lk, the SERIES60 length is denoted as Ls, and the formation spacing is denoted as D.

[0022] In the embodiment of the present invention, the commercial CFD software STAR-CCM+ is used for numerical simulation calculations. Due to the geometric symmetry of the hull shape, numerical simulation is only performed on half of the ship. In the calculation domain of the towing test pool, the center of gravity of the ship is set as the origin. In the X-axis direction, the distance from the velocity inlet is 18m, and the distance from the pressure outlet is -18m; in the Y-axis direction, the distance from the ship to the test pool boundary is 18m; in the Z-axis direction, the distance from the bottom of the test pool is -18m, and the distance from the top is 9m. The hydrostatic pressure corresponds to the calm water surface at the downstream boundary of the test pool. In STAR-CCM+, a mesh is created for the ship and the test pool based on the automatic mesh, such as Figure 3 As shown. Surface reconstruction, automatic surface repair, cut volume mesh generator and prism layer mesh generator were used. The target surface size was set to 50%, the number of prism layers was set to 6, and the volume growth rate was slow. In order to capture Kelvin waves and ensure a sharp transition between the two fluids, the water surface was densified. In STAR-CCM+, the resistance can be decomposed into shear resistance and pressure resistance. In order to further analyze the reasons for the reduction in formation sailing resistance, it is necessary to detect the impact of the bow flow field velocity on the ship in the formation. Therefore, 8 monitoring points arranged in a row were set in front of the ship, with each monitoring point spaced 0.5m apart. The flow field data of the ship's navigation was statistically analyzed, as shown in the figure. Figure 4 As shown in the figure, it is also necessary to consider the impact of the ship waves of the leading ship on the following ship. Therefore, 40 consecutive monitoring points are set behind the last ship in the formation, with an interval of 0.5m between each monitoring point, to count the changes in the ship waves of the formation.

[0023] To ensure the accuracy of the numerical calculation, a Froude number of 0.2599 was selected to perform a numerical simulation of the resistance of the KCS ship's self-propulsion. The numerical simulation results were compared with the experimental results provided at the 2010 Gothenburg seminar. The expression for the resistance coefficient is:

[0024] Where cd is the drag coefficient, the average total resistance Fd is 42.083 N, the water density ρ, the ship model speed v is 2.196 m / s, and the wetted surface area A of the ship model is 9.55275 m2.

[0025] The predicted resistance coefficient calculated using the above equation is 0.003654. The experimental resistance coefficient provided at the 2010 Gothenburg workshop is 0.003711, a deviation of 1.53%. Increasing the mesh accuracy will reduce the deviation in the predicted resistance. Note that the simulation only models half of the ship; to match the experimental data, the resistance coefficient needs to be multiplied by two.

[0026] S102, using a control variable method to design numerical simulation scenarios for different formation configurations, and determining the ship resistance coefficient of the longitudinal heterogeneous ship formation under each formation configuration, wherein the formation configuration includes the number, speed, spacing, and ship type of the formation ships.

[0027] In the embodiment of the present invention, Figure 5 As shown in the figure, KCS ship formation models with different numbers are constructed. Under the conditions of formation spacing of 1 times the ship length and sailing speed of 2.12m / s, the situations of single ship, 2-ship formation, 3-ship formation and 4-ship formation are simulated respectively. The changes of bow flow field velocity and ship wave are statistically analyzed through monitoring points, and their influence on the formation resistance characteristics is analyzed. Figure 6 The comparison of total resistance for KCS formations with different numbers of ships, at a formation spacing of 1 ship length and a sailing speed of 2.12 m / s, shows that the total resistance of the lead ship in the formation is similar to that of a single ship, while the total resistance of the following ships is lower than that of the lead ship. Therefore, the total resistance of the formation is lower than that of a single ship with the same number of ships.

[0028] In this embodiment of the present invention, a single ship and a three-ship formation are constructed under the same KCS ship type and a formation spacing of 1 times the ship length. The simulation is performed at sailing speeds of 1.275m / s, 2.12m / s, and 2.975m / s, respectively. A single ship is represented by "SG", and the three speeds are marked as low speed (LS), medium speed (MS), and high speed (HS). Figure 7 (a) shows the bow flow velocity of a single ship and a three-ship formation. Figure 7 (c) shows the corresponding shear resistance of the ships. As the formation sailing speed increases, the speed difference between the bow flow field of the lead ship and the following ship gradually increases, and the shear resistance changes in a similar trend. Figure 7 (d) shows the pressure resistance of a single ship and a three-ship formation. At high speeds, the increase in pressure resistance is significantly greater than the shear resistance, and the pressure resistance of the two following ships decreases significantly. Figure 7 The ship waves of the ship formation in (b) show that the ship waves have larger wave heights and longer wavelengths at high speeds, and carry more energy. Combined with the significantly decreased pressure resistance of the two following ships, this shows that high-energy ship waves can reduce the pressure resistance.

[0029] In this embodiment of the present invention, a three-ship formation was constructed under the same KCS ship type and a sailing speed of 2.12 m / s. Simulations were conducted with formation spacings of 0.5, 1, and 1.5 ship lengths, respectively. Formation spacing was categorized as close spacing (ND), medium spacing (MD), and far spacing (FD). Compared to the formation speed condition, the effect of formation spacing on drag reduction was more limited. In particular, as the formation spacing increased from medium to far spacing, the total resistance of each ship in the formation remained virtually unchanged. The proportions of shear resistance and pressure differential resistance were largely unaffected by changes in formation spacing, similar to the resistance distribution observed at a sailing speed of 2.12 m / s and a formation spacing of 1 ship length.

[0030] To verify that resistance can be reduced across different types of ships, this embodiment of the present invention constructed fleet models with varying numbers of SERIES60 ships. Simulations were conducted for single, two-ship, three-ship, and four-ship formations, with a formation spacing of 0.82 times the KCS ship length and a sailing speed of 2.12 m / s. The proportion of shear resistance in the SERIES60 fleet decreased from 62% to approximately 47%, with pressure differential resistance dominating. The resistance reduction rate also increased from 3% to approximately 10%. This change may be due to the smaller size of the SERIES60 ships (the SERIES60 weighs 485.6 kg, while the KCS ships weigh 823.0451 kg).

[0031] In the embodiment of the present invention, KCS ships and SERIES60 ships are used to construct a heterogeneous ship formation model, such as Figure 8 As shown in the figure, six heterogeneous ship formation models were constructed, including KCS-SERIES60-KCS (KSK), KCS-SERIES60-SERIES60 (KSS), KCS-KCS-SERIES60 (KKS), SERIES60-SERIES60-KCS (SSK), SERIES60-KCS-SERIES60 (SKS), and SERIES60-KCS-KCS (SKK). Simulations were conducted at a speed of 2.12 m / s and a formation spacing of 0.82 times the KCS ship length to analyze the effect of ship type on resistance characteristics. The results show that when a SERIES60 ship (small ship) follows a KCS ship (large ship), its bow flow velocity is lower, resulting in a corresponding reduction in shear resistance. High-energy ship waves help reduce pressure differential resistance, but the reduction is relatively limited. This explains why placing small ships behind large ships in a formation effectively reduces overall resistance. When there are more large ships in the formation, placing small ships in the middle can further reduce the formation's total resistance.

[0032] S103, based on the ship resistance coefficient of longitudinal heterogeneous ship formations under different formation configurations, analyze the effects of bow flow field velocity and ship waves on shear resistance and pressure difference resistance.

[0033] In an embodiment of the present invention, through the post-processing of STAR-CCM+, the shear resistance, pressure difference resistance, total resistance, bow flow field velocity and ship wave height data of the formation ships during the simulation process are output, and the following data processing is performed based on pycharm: the average value of the monitoring points is calculated and displayed in the form of a bar graph, the wave height, wavelength and period of the ship wave are statistically analyzed, the monitoring point data are displayed in a line graph, and the curve is smoothed by interpolation processing.

[0034] S104, predicting the formation resistance of the group of ships to be formed based on the action rule using a hybrid graph neural network prediction model, wherein the hybrid graph neural network prediction model includes a collaborative architecture of GATv2 dynamic attention mechanism and GINE geometric feature encoding.

[0035] In the embodiment of the present invention, a data-driven method is used to systematically study the coupled effects of ship type, spacing and speed on resistance characteristics in a ship formation, and a hybrid GNN prediction model is constructed. Given the high cost of acquiring STAR-CCM+ simulation data and the dynamic changes in the size of the formation, the use of a graph neural network architecture has two advantages: first, the graph structure of GNN naturally adapts to the modeling requirements of a variable number of ship nodes; second, its message passing mechanism can effectively capture the fluid interactions between ships. Figure 9 As shown, the model uses a three-way parallel architecture to achieve multi-level feature extraction: GATv2 layer: identifies key ship features through a dynamic attention mechanism and provides weighted neighbor information for subsequent aggregation; PNA layer: As the core of the architecture, it stably handles formations of different sizes through a multi-scale aggregator and integrates the attention weights of GATv2 and the geometric features of GINE; GINE layer: A message passing mechanism based on edge features that accurately encodes the fluid interference effects of the relative positions between ships.

[0036] In this embodiment of the present invention, three data enhancement measures were implemented to address the challenge of scarce CFD simulation data: first, the complexity of individual samples was reduced by decoupling flow field features; second, supplementary data was generated using a physically constrained interpolation algorithm; and finally, a dataset containing approximately 100,000 samples was constructed (training set:test set = 5:1). Regarding feature engineering, formation spacing was dimensionless (L / L_KCS∈[0.5,1.5]) based on the KCS length, while speed was retained at its original value (v∈[1.275,2.975] m / s) to mitigate bias introduced by inconsistent Froude numbers for heterogeneous ship types.

[0037] In the embodiment of the present invention, Figure 10As shown in the figure, the model training process exhibits three significant stages: rapid convergence period (0-15 Epochs): the loss function decreases by 86%, completing basic feature extraction; fine tuning period (15-60 Epochs): local optimal search is achieved through parameter oscillation; stable convergence period (>60 Epochs): the final training loss is 0.00148, the test loss is 0.0149, and the fluctuation range is <1.5%.

[0038] In the embodiment of the present invention, the prediction accuracy of different formation sizes is quantitatively analyzed by heat map ( Figure 11 ), and found that: the prediction error of the 2-ship formation is the lowest (the heat map has the lightest tone), which is due to the completeness of the training data; the error of the 4-ship formation is relatively high (the tone is the darkest), reflecting the modeling challenges in data-sparse areas; the accuracy distribution is strongly correlated with the training sample density.

[0039] In the embodiments of the present invention, comparative experiments, as shown in Table 1, confirm that the hybrid GNN significantly outperforms the baseline model in two key metrics: prediction accuracy: MSE is reduced by 13.6% (vs. GAT) and 21.3% (vs. random forest); generalization capability: R² is increased to as high as 0.92, and it supports formation prediction of any size. While random forests have computational efficiency advantages, their fixed input dimension makes them unsuitable for application scenarios with varying formation sizes.

[0040] Table 1 Comparison of MSE and R2 indicators of each model

[0041] like Figure 12 As shown, the present invention also provides an electronic device 1200. The electronic device 1200 includes a processor 1201, a memory 1202 and a display 1203. Figure 12 Only some of the components of the electronic device 1200 are shown, but it should be understood that it is not required to implement all of the shown components, and more or fewer components may be implemented instead.

[0042] In some embodiments, the processor 1201 may be a central processing unit (CPU), a microprocessor, or other data processing chip, configured to execute program codes or process data stored in the memory 1202, such as the longitudinal heterogeneous ship formation design method of the present invention.

[0043] In some embodiments, processor 1201 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, processor 1201 may be local or remote. In some embodiments, processor 1201 may be implemented on a cloud platform. In some embodiments, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, multiple clouds, or any combination thereof.

[0044] In some embodiments, the memory 1202 may be an internal storage unit of the electronic device 1200, such as a hard disk or memory of the electronic device 1200. In other embodiments, the memory 1202 may also be an external storage device of the electronic device 1200, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 1200.

[0045] Furthermore, the memory 1202 may include both an internal storage unit of the electronic device 1200 and an external storage device. The memory 1202 is used to store application software installed on the electronic device 1200 and various data.

[0046] In some embodiments, display 1203 can be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 1203 is used to display information about electronic device 1200 and to display a visual user interface. Components 1201-1203 of electronic device 1200 communicate with each other via a system bus.

[0047] In some embodiments, when the processor 1201 executes the longitudinal heterogeneous ship formation design program in the memory 1202, the following steps may be implemented: The control variable method is used to numerically simulate the longitudinal heterogeneous ship formations with different formation configurations, and the ship resistance coefficients of the longitudinal heterogeneous ship formations under each formation configuration are determined. A dataset is constructed based on the formation configurations of longitudinally heterogeneous ship formations and the corresponding resistance coefficients of each formation configuration. The dataset is then used to train the preset ship formation design model. The trained ship formation design model is used to predict the resistance coefficients of different formation configurations of the ship group to be formed, and the formation configuration with the smallest resistance coefficient is determined as the target formation configuration.

[0048] It should be understood that, when the processor 1201 executes the longitudinal heterogeneous ship formation design program in the memory 1202 , in addition to the above functions, it can also implement other functions. For details, please refer to the description of the corresponding method embodiment above.

[0049] Accordingly, an embodiment of the present application also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, it can implement the steps or functions in the longitudinal heterogeneous ship formation design method provided by the above-mentioned method embodiments.

[0050] Those skilled in the art will appreciate that all or part of the process steps of the above-described embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, such as a magnetic disk, an optical disk, a read-only memory, or a random access memory.

[0051] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A drag reduction prediction model design method based on longitudinal heterogeneous ship formation, characterized by: include: Setting the numerical simulation parameters of the longitudinal heterogeneous ship formation in the preset simulation software; Using the control variable method to design numerical simulation scenarios for different formation configurations, the ship resistance coefficients of longitudinally heterogeneous ship formations were determined for each formation configuration, including the number of ships in the formation, speed, spacing, and ship type. Based on the ship resistance coefficient of longitudinal heterogeneous ship formations under different formation configurations, the effects of bow flow field velocity and ship waves on shear resistance and pressure resistance are analyzed; Based on the action law, a hybrid graph neural network prediction model is used to predict the formation resistance of the group of ships to be formed, wherein the hybrid graph neural network prediction model includes a collaborative architecture through the GATv2 dynamic attention mechanism and the GINE geometric feature encoding.

2. The drag reduction prediction model design method based on longitudinal heterogeneous ship formation according to claim 1 is characterized in that: The setting of numerical simulation parameters of the longitudinal heterogeneous ship formation in the preset simulation software includes: In the STAR-CCM+ simulation environment, eight monitoring points arranged in a row are set in front of the ship, with each monitoring point spaced 0.5 m apart, to collect data on the bow flow field. Forty monitoring points are set behind the last ship in the formation, with each monitoring point spaced 0.5 m apart, to collect data on changes in ship-borne waves.

3. The drag reduction prediction model design method based on longitudinal heterogeneous ship formation according to claim 1 is characterized in that: The numerical simulation scenarios for designing different formation configurations using the control variable method include: Under the condition that the ship speed and the ship spacing are the same, different numbers of ships are configured for the longitudinal heterogeneous ship formation model; Under the condition that the number of ships and the distance between ships are the same, different ship speeds are configured for the longitudinal heterogeneous ship formation model; When the number of ships and the speed of ships are the same, different ship spacings are configured for the longitudinal heterogeneous ship formation model.

4. The drag reduction prediction model design method based on longitudinal heterogeneous ship formation according to claim 1 is characterized in that: The numerical simulation scenarios for designing different formation configurations using the control variable method specifically include: A heterogeneous ship formation model is constructed using KCS ships and SERIES60 ships. Under the conditions of a certain number of ships, the same ship speed and the same ship spacing, different ship type combinations are configured for the longitudinal heterogeneous ship formation model.

5. The drag reduction prediction model design method based on longitudinal heterogeneous ship formation according to claim 2 is characterized in that: The analysis of the effects of bow flow field velocity and ship waves on shear resistance and pressure difference resistance based on the ship resistance coefficient of longitudinal heterogeneous ship formations under different formation configurations includes: During the STAR-CCM+ simulation process, the changes in the bow flow field velocity and the ship wave height over time are statistically analyzed through monitoring points, and the average values ​​of the changes in the bow flow field velocity and the ship wave height over time are calculated and displayed in the form of a bar graph. At the same time, the wave height, wavelength and period of the ship wave are statistically analyzed, and the monitoring point data are displayed in a line graph, which is then interpolated.

6. The drag reduction prediction model design method based on longitudinal heterogeneous ship formation according to claim 2 is characterized in that: The method of predicting the formation resistance of the group of ships to be formed by using a hybrid graph neural network prediction model based on the action law includes: The hybrid graph neural network prediction model includes a parallel architecture of GATv2 layer, PNA layer, and GINE layer, which is used for multi-level feature extraction and analysis of model prediction accuracy.

7. The drag reduction prediction model design method based on longitudinal heterogeneous ship formation according to claim 6 is characterized in that: The prediction accuracy of the analytical model includes: Three data enhancement measures are taken to address the problem of scarce CFD simulation data, and the prediction accuracy of different formation sizes is quantitatively analyzed through heat maps.

8. An electronic device, characterized in that: comprising a memory and a processor, wherein, The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the method for designing a drag reduction prediction model based on a longitudinal heterogeneous ship formation as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the method for designing a drag reduction prediction model based on a longitudinal heterogeneous ship formation as described in any one of claims 1 to 7.

Images