A method for obtaining RAO curve of damaged ships based on CFD reduced-order model

By combining the down-order model of fine and rough grid strategies, using machine learning to fit the damaged ship RAO curve, the problem of excessive calculation time in the existing technology is solved, and efficient RAO curve calculation is achieved.

CN114707400BActive Publication Date: 2025-08-29DALIAN UNIV OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210216167.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-03-07
Publication Date
2025-08-29
Estimated Expiration
2042-03-07

AI Technical Summary

Technical Problem

The existing CFD method takes too long to calculate when solving damaged ship RAO curves, especially in the simulation of ship motion responses in waves, which leads to low computational efficiency.

Method used

A down-order model combining fine grid strategy and rough grid strategy is adopted. By combining more rough data and less precise data, the RAO curve of the damaged ship is obtained, and the data fit is used to improve the calculation speed.

Benefits of technology

On the premise of ensuring the calculation accuracy, the calculation speed of the RAO curve of the damaged ship is greatly improved, the dependence on data is reduced, the calculation efficiency is increased by 27.47%, and the error is small.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114707400B_ABST
    Figure CN114707400B_ABST
Patent Text Reader

Abstract

This paper provides a method for obtaining the RAO curve of a damaged ship using a reduced-order CFD model. This method utilizes a reduced-order method that combines numerical simulation data from a fine grid strategy with data from a coarse grid strategy. By combining more coarse data with less precise data, the RAO curve of the damaged ship is derived, ensuring that the final result is similar to that obtained using high-fidelity calculations, thereby improving calculation speed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of fluid dynamics, and in particular to a method for obtaining a RAO curve of a damaged ship based on a CFD reduced-order model. Background Art

[0002] Existing techniques for solving the motion of damaged ships using CFD methods, such as obtaining the RAO curve of a damaged ship, typically select multiple wave cycles at regular intervals within a selected wave cycle range. High-fidelity CFD methods are then used to simulate the motion of the damaged ship in each wave cycle. The resulting motion response data is then dimensionlessly processed to produce the RAO curve. To ensure the accuracy of the numerical simulation, a finer mesh and smaller time step are generally required, especially for calculating the ship's motion response in waves. To ensure the quality of wave generation, the mesh and time step requirements are also higher, which increases the computational time. Summary of the Invention

[0003] In response to the technical problems raised above, a method for obtaining the RAO curve of a damaged ship using a reduced-order model based on CFD is provided. This method utilizes a reduced-order method that combines numerical simulation data using a fine grid strategy with numerical simulation data using a coarse grid strategy. By combining more coarse data with less precise data, the RAO curve of the damaged ship is obtained, ensuring that the final result is close to that obtained by high-fidelity calculations, thereby improving the calculation speed. The technical means adopted by the present invention are as follows:

[0004] A method for obtaining the RAO curve of a damaged ship based on a reduced-order CFD model includes the following steps:

[0005] Step 1: Create a model of the damaged ship to be calculated;

[0006] Step 2: Confirm the wave period range [a, b] of the RAO curve of the damaged ship to be calculated;

[0007] Step 3: Select n wave periods within the selected wave period range [a, b] for numerical simulation, denoted as x = {x1, x2…x n}, and select c wave periods in x, denoted as x s ={x s1 ,x s2 …x sc};

[0008] Step 4: Use the fine grid strategy of CFD software to calculate the damage of the damaged ship at the wave period x. s The corresponding motion response is dimensionless and recorded as y hs ={y hs1 ,y hs2 …y hscThen, the coarse grid strategy is used to calculate the motion response of the damaged ship when the wave period is x, and the dimensionless value is recorded as y l ={y l1 ,y l2 …y ln}, and put y l Medium x s The corresponding response values ​​are recorded as y ls ={y ls1 ,y ls2 …y lsc};

[0009] Step 5: hs with y ls Difference to get y ds ={y ds1 ,y ds2 …y dsc};

[0010] Step 6: Replace {(x,y l )} and {(x s ,y ds )} fitting to obtain Y l With Y d ;

[0011] Step 7: Y l With Y d Add them together to get Y, which is the RAO curve of the damaged ship to be determined.

[0012] Furthermore, in step 2, the wave period and other related wave data of the RAO curve of the damaged ship to be calculated are scaled based on the scaling size of the model.

[0013] Furthermore, after step 7, the following steps are further provided:

[0014] Step 8: Compare the RAO curve calculated by the reduced-order model with the RAO curve directly fitted by high-fidelity numerical simulation to make an error judgment.

[0015] The present invention has the following advantages:

[0016] 1. The present invention can significantly improve the speed of obtaining the RAO curve of a damaged ship through numerical simulation while ensuring calculation accuracy;

[0017] 2. This method has low dependence on data, that is, when the same rough data is used, the use of different precise data has little impact on the final result, and can obtain results close to high-precision simulation, thereby improving the calculation speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 This is a diagram of a damaged ship model constructed in an embodiment of the present invention.

[0020] Figure 2 Schematic diagram of the calculation process after CFD software processing in an embodiment of the present invention.

[0021] Figure 3 Schematic diagram comparing the reduced-order model and high-fidelity simulation in an embodiment of the present invention.

[0022] Figure 4 Schematic diagram of the error between the reduced-order model and high-precision simulation and experiment in an embodiment of the present invention.

[0023] Figure 5 It is a schematic diagram of the overall process of the present invention. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings 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 making creative efforts shall fall within the scope of protection of the present invention.

[0025] This example takes the solution of the roll RAO ​​curve of the damaged ship DTMB-5415 as an example to introduce the implementation route and effect of this method.

[0026] like Figure 5 As shown, this embodiment discloses a method for obtaining the RAO curve of a damaged ship based on a CFD reduced-order model, comprising the following steps:

[0027] Step 1: Use CATIA to create a damaged ship model of DTMB-5415 for subsequent numerical simulation. Figure 1 , the main parameters are shown in Table 1;

[0028] Table 1: Main parameters of DTMB-5415 vessel

[0029]

[0030] Step 2: Confirm that the wave period range of the RAO curve of the damaged ship to be calculated is 8s to 14s. Since the model is a 1:51 scale model, after scaling the wave period, it is finally confirmed that the wave period range is 1.12s to 1.96s.

[0031] Step 3: Select the following 9 wave periods in the selected wave period range for numerical simulation, denoted as x = {1.12s, 1.26s, 1.33s, 1.40s, 1.47s, 1.54s, 1.68s, 1.82s, 1.96s}, and select 6 wave periods in x, denoted as x s ={1.12s, 1.33s, 1.47s, 1.54s, 1.68s, 1.96s};

[0032] Step 4: Use the CFD commercial software STAR-CCM+ to calculate the damage of the damaged ship with a wave period of x using a fine grid strategy. s The corresponding roll response amplitude is obtained after dimensionless processing. hs (See Figure 2 ); Then, a coarse grid strategy (selecting a larger basic grid size, a larger time step, a smaller number of internal iterations, etc.) is used to calculate the motion response of the damaged ship when the wave period is x, and after dimensionless processing, it is recorded as y l (See Figure 2 ), and put y l Medium x s The corresponding response values ​​are recorded as y ls ; For different calculation problems, the grid size used may be different, and it is impossible to use an accurate numerical limit to distinguish between fine grids and coarse grids. In this embodiment, the number of fine simulation grids and the number of coarse simulation grids are shown in Table 2.

[0033] Table 2: Number of grids in different working conditions

[0034]

[0035] Step 5: hs with y ls Difference to get y ds (See Figure 2 ), that is, y ds =y hs -y ls ;

[0036] Step 6: Replace {(x,y l )} and {(x s ,y ds )} is fitted by machine learning or some general fitting methods. In this embodiment, the PCHIP (interpolation polynomial) method is used for fitting to obtain Y l With Yd (See Figure 2 );

[0037] Step 7: Y l With Y d Add together to get Y (see Figure 2 ), that is, Y = Y l +Y d , Y is the RAO curve of the damaged ship to be determined.

[0038] Figure 3 It can be observed that the RAO curve calculated by the reduced-order model is basically consistent with the RAO curve directly fitted by the high-fidelity numerical simulation. s The MAE error of the damaged ship motion response obtained by high-precision calculation of the remaining wave periods is 0.6643.

[0039] All calculations were performed using an Intel(R) Xeon(R) GOLD 6152CPU @ 2.10 GHz computer. The physical calculation time for each wave cycle varied: a precise simulation calculation of 1 second required 88.5 CPU core hours, while a rough simulation calculation of 1 second required 5 CPU core hours. The total calculation time for the reduced-order model was 19,770 CPU core hours, and the total calculation time for the high-precision simulation was 27,257 CPU core hours, achieving a 27.47% improvement in computational efficiency.

[0040] In order to illustrate the low data dependence of this method, 8 grouping situations were made to complete the acquisition of RAO curves of damaged ships using this method when 6 accurate data were selected. The grouping situations are shown in Table 3:

[0041] Table 3: Grouping of training points

[0042]

[0043] In order to more intuitively represent the data dependence of this method, the MAE method is used to directly calculate the error between the RAO curve obtained by the reduced-order model and the experimental data, and the obtained error is compared with the error of high-precision simulation data and experimental data. Figure 4 shown.

[0044] As can be seen from the figure, the errors between the reduced-order model and the experiment in different groups are close to the errors between the high-precision simulation and the experiment, and the fluctuation of the errors between the reduced-order model and the experiment is also small as the grouping changes. This shows that the model proposed in the present invention is less sensitive to data, that is, when the rough data used are the same, the selection of different precise data has little effect on the final result, and can obtain results that are relatively close to the high-precision simulation.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for obtaining RAO curves of damaged ships based on a reduced-order CFD model, characterized in that: The steps include: Step 1: Create a model of the damaged ship to be calculated; Step 2: Confirm the wave period range [a, b] of the RAO curve of the damaged ship to be calculated; Step 3: Select n wave periods within the selected wave period range [a, b] for numerical simulation, denoted as , and in Select c wave periods from ; Step 4: Use the fine grid strategy of CFD software to calculate the damage of the damaged ship when the wave period is The corresponding motion response is dimensionless and recorded as ; Then the coarse grid strategy is used to calculate the damaged ship when the wave period is The corresponding motion response is dimensionless and recorded as , and put middle The corresponding response values ​​are recorded as ; Step 5: and Do the difference ; Step 6: and Fitting and ; Step 7: and Add together to get , This is the RAO curve of the damaged ship to be found; In the step 2, the wave data related to the wave period of the RAO curve of the damaged ship to be calculated is scaled based on the scaling size of the model; After step 7, the following steps are further performed: Step 8: Compare the RAO curve obtained in step 7 with the RAO curve obtained by direct fitting of high-fidelity numerical simulation to determine the error.

Citation Information

Patent Citations

  • DE290030A

  • A method for calculating droplet motion phase change parameters based on a steam flow field

    CN109740281A