Design method of bow for river-sea direct ship

Abstract

The application discloses a bow design method suitable for a river-sea direct ship, and the method comprises the following steps: defining and constructing a key feature control line through a characteristic parameter, and establishing a parametric model of the bow; selecting a parent ship type, and obtaining initial values of the parent ship corresponding to the characteristic parameter; constructing a change range of the characteristic parameter relative to the initial values of the parent ship; generating N sets of characteristic parameter schemes by using a Latin hypercube design method; inputting the N sets of characteristic parameter schemes into the parametric model to generate N kinds of ship body geometric model schemes; calculating total resistance and maximum ram pressure of the bow of the N kinds of ship body geometric model schemes; respectively establishing a total resistance performance prediction model and a maximum ram pressure prediction model of the bow by means of a neural network method; constructing an improved differential evolution algorithm, taking the total resistance of the ship and the maximum ram pressure of the bow as the target, and performing optimization design on the bow of the river-sea direct ship, and obtaining a design result. The bow designed by the application can improve the resistance performance of the ship while considering the ramming performance.

Description

Technical Field

[0001] This invention relates to the field of ship design technology, and in particular to a bow design method applicable to river-sea direct vessels. Background Technology

[0002] A vessel should possess excellent performance during navigation to enable it to navigate in adverse weather or extreme sea conditions. Furthermore, the goal of vessel design is often to maximize speed—that is, to achieve a vessel design with optimal speed while using minimal engine power, thereby improving fuel efficiency. Engine power usage is closely related to vessel resistance.

[0003] As is well known, the shape of a ship's bow has a significant impact on its drag during navigation. Currently, the mainstream bow shapes are generally defined as five types: 1. Straight bow: The bowpost is straight and basically perpendicular to the baseline; commonly found on barges and special-purpose vessels. 2. Raked bow: The bowpost is straight and slightly curved forward. This type is less prone to wave impact, and in the event of a collision, the portion of the bow below the waterline is less likely to be damaged. 3. Clipper bow: Above the design waterline, the bow has a concave curve and a large forecastle deck, which is beneficial for the placement of anchors and mooring equipment; the bow is also less prone to wave impact. 4. Ice-resistant bow: Below the design waterline, the bowpost is inclined, forming a 30° angle with the baseline; commonly found on icebreakers. 5. Bulbous bow: This design features a bulbous protrusion at the front of the bow below the waterline. Its function is to reduce wave-making drag and form drag, and it is widely used in ocean-going vessels. Furthermore, bow designs have undergone several modifications in the past. In 2006, the Norwegian Ulstein Group introduced the X-shaped bow to improve fuel efficiency and safety at sea. The furthest point of the X-shaped bow is located at the very stern (facing the waterline), providing a continuous, sharp bow shape for the hull. This sharp design at the foremost point helps the vessel navigate waves and improves overall stability.

[0004] Against the backdrop of dwindling fossil fuel resources and increasing greenhouse gas emissions, inland waterway transportation offers significant environmental and economic advantages, leading to its growing share in the transportation industry and widespread attention to direct river-sea transport. However, few ship bows are currently designed specifically for the characteristics of direct river-sea vessels. Considering the nature of such vessels, using inland waterway bows directly could expose the bow to significant wave impact during the journey from river to sea, reducing comfort and potentially causing accidents in extreme cases. Conversely, using ocean-going vessel bows would result in poor fuel economy and high carbon emissions on inland waterways.

[0005] Therefore, in order to further improve the fuel economy, reduce carbon emissions, and enhance ship safety of river-sea direct vessels, developing bow shapes suitable for river-sea direct vessels is a key technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] The main objective of this invention is to propose a bow design method suitable for river-sea direct vessels. The bow designed by this method can improve the ship's resistance performance while taking into account the impact resistance of the river-sea direct vessel during navigation, thereby reducing fuel consumption and pollution emissions. At the same time, it reduces the bow impact phenomenon of river-sea direct vessels when entering the sea from the river, thus improving the safety of the vessel.

[0007] The technical solution adopted in this invention is:

[0008] A bow design method applicable to river-sea direct vessels includes the following steps:

[0009] S1. Define and construct key feature control lines through feature parameters Pi to establish a parameterized model of the bow; the key feature control lines include: bow mid-longitudinal section line, upper bow mid-longitudinal section line, bow horizontal line, bow maximum concave line, bow widest line, bow waterline, and bow deck line; i = 1, 2, ..., M, where M is the number of feature parameters;

[0010] S2. Select the parent ship type and measure and obtain the initial value vi of the parent ship corresponding to the characteristic parameter Pi;

[0011] S3. Construct the range of variation of characteristic parameter Pi relative to the initial value vi of the parent ship during the design process;

[0012] S4. Using the Latin hypercube design method according to equation (1), generate N sets of characteristic parameter schemes within the range of variation constructed in S3;

[0013] P i j =LHS(LBi,UBi) (1)

[0014] In the formula, i = 1, 2, ..., M, j = 1, 2, ..., N, LHS represents the Latin hypercube function, UBi is the upper limit of Pi, LBi is the lower limit of Pi, and P i j This represents the value of the characteristic parameter Pi in the j-th group of characteristic parameter schemes;

[0015] S5. Input the N sets of feature parameter schemes generated in S4 into the parameterized model in S1 to generate N kinds of ship hull geometric model schemes.

[0016] S6. Calculate the total resistance Fj for each of the N ship hull geometric model schemes obtained in S5, j = 1, 2, ..., N;

[0017] S7. Calculate the maximum bow impact pressure Yj for the N hull geometry model schemes obtained in S5, j = 1, 2, ..., N;

[0018] S8. Based on the total resistance and maximum bow impact pressure data collected in S6 and S7, establish a total resistance performance prediction model Mf and a maximum bow impact pressure prediction model My respectively using neural network methods.

[0019] S9. Construct an improved differential evolution algorithm to optimize the bow design of a river-sea direct vessel with the goal of reducing the total resistance of the ship and the maximum impact pressure on the bow, and obtain the design results.

[0020] In the above scheme, the characteristic parameters for constructing the longitudinal section of the bow include: the position of the bow bottom rise (P1), the position of the foremost edge of the bow (P2), the bow bottom rise angle (P3), the bow inclination angle (P4), the straight distance of the bow bottom rise (P5), and the straight distance of the bow inclination (P6).

[0021] In the above scheme, the characteristic parameters for constructing the upper bow longitudinal section include: the inclination angle of the starting position of the upper bow longitudinal section (P7), the straight-line distance of the starting position of the upper bow longitudinal section (P8), the ending position of the upper bow longitudinal section (P9), the inclination angle of the ending position of the upper bow longitudinal section (P10), and the straight-line distance of the ending position of the upper bow longitudinal section (P11).

[0022] In the above scheme, the characteristic parameters for constructing the bow flat bottom line include: bow flat bottom line contraction position (P12), bow flat bottom line contraction position straight line distance (P13), bow flat bottom line end position (P14), bow flat bottom line end position tilt angle (P15), and bow flat bottom line end position straight line distance (P16).

[0023] In the above scheme, the characteristic parameters for constructing the maximum concave line of the bow include: the starting position of the maximum concave line of the bow (P17), the inclination angle of the starting position of the maximum concave line of the bow (P18), the straight-line distance of the starting position of the maximum concave line of the bow (P19), the ending position of the maximum concave line of the bow (P20), the inclination angle of the ending position of the maximum concave line of the bow (P21), and the straight-line distance of the ending position of the maximum concave line of the bow (P22).

[0024] In the above scheme, the characteristic parameters for constructing the bow widest line include: the starting position of the bow widest line (P23), the straight-line distance from the starting position of the bow widest line (P24), the ending position of the bow widest line (P25), the straight-line distance from the ending position of the bow widest line (P26), the inclination angle of the middle position of the bow widest line (P27), and the straight-line distance of the middle position of the bow widest line (P28).

[0025] In the above scheme, the characteristic parameters for constructing the bow waterline include: the starting position of the bow waterline (P29), the inclination angle of the starting position of the bow waterline (P30), the straight-line distance of the starting position of the bow waterline (P31), the ending position of the bow waterline (P32), the inclination angle of the ending position of the bow waterline (P33), and the straight-line distance of the ending position of the bow waterline (P34).

[0026] In the above scheme, the characteristic parameters for constructing the bow deck line include: the starting position of the bow deck line (P35), the inclination angle of the starting position of the bow deck line (P36), the straight-line distance of the starting position of the bow deck line (P37), the ending position of the bow deck line (P38), the inclination angle of the ending position of the bow deck line (P39), and the straight-line distance of the ending position of the bow deck line (P40).

[0027] In the above scheme, step S8, establishing the total resistance performance prediction model Mf and the maximum bow impact pressure prediction model My, specifically includes the following steps:

[0028] S8.1. Standardize the collected resistance and maximum bow impact pressure data into input matrices, namely MatrixF and MatrixY;

[0029]

[0030] S8.2 Normalize MatrixF and MatrixY using the following formula;

[0031]

[0032] In the formula: MatrixF k Let MatrixY be the element in the k-th column of MatrixF. k Let MatrixY be the element in the k-th column, where k = 1, 2, ..., M+1, and MatrixF be the element in the k-th column. k ′ and MatrixY k ′ This is the result after normalization;

[0033] S8.3, MatrixF k ′ and MatrixY k ′ They were divided into training and testing sets respectively;

[0034] S8.4 Design a neural network structure, including an input layer, a hidden layer, and an output layer; set the number of neurons in the input layer to nm, the number of neurons in the hidden layer to num, and the number of neurons in the output layer to 1; use ReLU as the activation function and set the loss function to the mean squared error between the neural network output and the actual calculation result;

[0035] S8.5, MatrixF k ′ The obtained training set is used as training data to train the neural network model, and the trained neural network model is tested using the test set to obtain the total drag performance prediction model Mf.

[0036] S8.6, MatrixY k ′ The obtained training set is used as training data to train the neural network model, and the trained neural network model is tested using the test set to obtain the maximum impact pressure prediction model My for the bow.

[0037] In the above scheme, step S9 involves constructing an improved differential evolution algorithm to optimize the bow design of a river-sea direct-route vessel, with the goal of reducing the total resistance of the ship and the maximum impact pressure on the bow. The specific steps are as follows:

[0038] S9.1 Set the maximum number of iterations of the algorithm gmax, and generate I initial optimized design schemes for the bow of the river-sea direct-access vessel by randomizing the variation range of the characteristic parameters defined in S3 according to equation (2);

[0039]

[0040] In the formula: i = 1, 2, ..., M, m = 1, 2, ..., I, For the m-th initial river-sea direct-access vessel bow optimization design scheme, r m,g The value is a random value between 0 and 1, and g is the iteration number identifier, which is recorded as 1 in this step;

[0041] S9.2. According to Equation (3), perform chaotic mapping on the I initial optimized design schemes for the bow of the randomized river-sea direct-access vessel;

[0042]

[0043] In the formula: For the m-th optimized design scheme of the bow section of a river-sea direct-access vessel after chaotic mapping, r m,g A random value between 0 and 1;

[0044] S9.3. Randomly divide the I optimized design schemes for the bow of the river-sea direct-access vessel after chaotic mapping into two groups, and label them as I1 and I2 respectively.

[0045] S9.4. The I / 2 optimized design schemes for the bow of the river-sea direct vessel after chaotic mapping marked as I1 are used to generate a further optimized design scheme for the bow of the river-sea direct vessel using Equation (4).

[0046]

[0047] In the formula: n1 = 1, 2, ..., I / 2, For the n1th generated further optimized design scheme of the bow of the river-sea direct vessel, r1, r2 and r3 are random integers from 1 to 1 / 2; F is the scaling factor with a value of 0.5;

[0048] S9.5. The I / 2 river-sea direct-ship bow optimization design schemes marked as I2 after chaotic mapping are used to generate a further river-sea direct-ship bow optimization design scheme using Equation (5).

[0049]

[0050] In the formula: n2=I / 2+1,…,I, For the n2th generated further optimized design scheme of the bow of the river-sea direct vessel, r4, r5, r6, r7 and r8 are random integers from I / 2+1 to I, and F is a scaling factor of 0.5;

[0051] S9.6, Merging and for m = 1, 2, ..., I;

[0052] S9.7. The further optimized design scheme of the bow section of the river-sea direct vessel is refined by formula (6) to generate a refined optimized design scheme of the bow section of the river-sea direct vessel.

[0053]

[0054] In the formula, i rand is a random integer from 1 to M, and r is a random value from 0 to 1;

[0055] S9.8, will The input is obtained in Mf of S8 The total resistance performance of the generated initial and refined river-sea direct vessel bow optimization design schemes are calculated using the ship total resistance performance prediction model Mf constructed in S8.

[0056] S9.9, will The input is obtained in My in S8 Even if we use the maximum bow impact pressure prediction model My built in S8 to calculate the maximum bow impact pressure of the generated initial and refined river-sea direct-ship bow optimization design schemes;

[0057] S9.10, according to formula (7) The corresponding optimized design scheme for the bow of the ship is selected, and the selected scheme is assigned a value. The iteration number marker g is updated to iter, and iter = g + 1, meaning that the schemes with smaller total resistance and maximum bow impact pressure generated in this iteration are retained for the next iteration calculation;

[0058]

[0059] In the formula, This represents the initial optimized design scheme for the bow of the river-sea direct-access vessel in the next iteration calculation process;

[0060] S9.11, Determine if iter is greater than gmax. If less, jump to S9.3 and set g in S9.3 to the value of iter for the next iteration. If greater, then... The scheme with the lowest total resistance and the lowest maximum impact pressure on the ship's bow is determined as the optimal design.

[0061] The beneficial effects of this invention are:

[0062] The parameterized bow model constructed in this application can achieve significant deformation of the bow shape, enabling different bow design options. This method allows for the rapid acquisition of various bow designs, expanding the design space and enabling more design schemes. Furthermore, the total resistance performance prediction model Mf and the maximum bow impact pressure prediction model My established in S8 can be used for rapid prediction of resistance and bow impact characteristics in other design schemes, improving design efficiency during preliminary design. Finally, the improved differential evolution algorithm in S9 further enhances the algorithm's ability to optimize the bow design of river-sea direct-route vessels, aiming to reduce total resistance and maximum bow impact pressure. Verification shows that the composite bow design using this invention improves the vessel's resistance performance while also considering the impact performance of river-sea direct-route vessels during navigation, thereby reducing fuel consumption and pollution emissions; simultaneously, it reduces bow impact when river-sea direct-route vessels enter the sea, improving vessel safety. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a schematic diagram of the key feature control lines of the bow of a ship;

[0065] Figure 2 These are schematic diagrams of the characteristic parameters corresponding to the key feature control lines; among them, (2-1) is a schematic diagram of the characteristic parameters of the longitudinal section line in the bow; (2-2) is a schematic diagram of the characteristic parameters of the longitudinal section line in the upper part of the bow; (2-3) is a schematic diagram of the characteristic parameters of the bow flat bottom line; (2-4) is a schematic diagram of the characteristic parameters of the bow maximum concave line; (2-5) is a schematic diagram of the characteristic parameters of the bow widest line; (2-6) is a schematic diagram of the characteristic parameters of the bow waterline; and (2-7) is a schematic diagram of the characteristic parameters of the bow deck line.

[0066] Figure 3 These are schematic diagrams of the CAD geometric models of two bow designs generated during processes S4 and S5 in this embodiment of the invention. Among them, (3-1) is a schematic diagram of the constructed bulbous bow design, and (3-2) is a schematic diagram of the upright bow design.

[0067] Figure 4 These are partial test results of the accuracy of the trained neural network models Mf and My in embodiment S8 of the present invention. Among them, (4-1) is a partial test result of the total resistance performance prediction model Mf, and (4-2) is a partial test result of the maximum bow impact pressure prediction model My.

[0068] Figure 5 This refers to the calculation iteration process in embodiment S9 of the present invention;

[0069] Figure 6 This is the bow of the river-sea direct vessel obtained through iterative calculation in this embodiment of the invention, wherein (6-1) is a side view and (6-2) is a front view;

[0070] Figure 7 These are three existing bow designs used for resistance performance verification in this embodiment of the invention. Among them, (7-1) is a conventional bulbous bow design, (7-2) is a conventional upright bow design, and (7-3) is a waterline-contracting upright bow design.

[0071] Figure 8This is a schematic diagram of the distribution of impact pressure measurement points on the bow structure during impact performance verification in an embodiment of the present invention. (8-1) is a typical bulbous bow, and (8-2) is the bow of a river-sea direct vessel in an embodiment of the present invention.

[0072] Figure 9 This is a curve showing the peak value of the slamming pressure at different measuring points in this invention as a function of the ship's bottom along the positive X-axis.

[0073] Figure 10 This is the curve showing the variation of the peak impact pressure along the positive Z-axis of the first group of pressure measuring points in this embodiment of the invention.

[0074] Figure 11 This is the curve showing the peak impact pressure of the second group of pressure measuring points in this embodiment of the invention changing along the positive Z-axis.

[0075] Figure 12 This is the curve showing the change of the peak impact pressure along the positive Z-axis of the third group of pressure measuring points in this embodiment of the invention. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0077] It should be noted that the illustrations provided in the embodiments of the present invention are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0078] In this invention, it should also be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. Furthermore, the terms "first" and "second" are used only for descriptive and distinguishing purposes and should not be construed as indicating or implying relative importance.

[0079] This invention proposes a bow design method suitable for river-sea direct-route vessels. By constructing a parametric model of the bow's three-dimensional structure, and leveraging intelligent optimization algorithms combined with simulation and neural network methods, aiming at drag performance and bow impact performance, an energy-saving, low-impact bow design suitable for river-sea direct-route vessels is discovered. The method specifically includes:

[0080] S1. Define and construct key feature control lines using feature parameters Pi to establish a parametric model of the bow. For example... Figure 1 As shown, the key feature control lines include: bow mid-longitudinal section line, upper bow mid-longitudinal section line, bow horizontal line, bow maximum concave line, bow widest line, bow waterline, and bow deck line; i = 1, 2, ..., M, where M is the number of feature parameters.

[0081] As shown in Figure (2-1), the characteristic parameters for constructing the longitudinal section of the bow include: the position of the bow bottom rise (P1), the position of the foremost edge of the bow (P2), the bow bottom rise angle (P3), the bow inclination angle (P4), the straight distance of the bow bottom rise (P5), and the straight distance of the bow inclination (P6).

[0082] As shown in Figure (2-2), the characteristic parameters for constructing the mid-longitudinal section of the upper bow include: the inclination angle of the starting position of the mid-longitudinal section of the upper bow (P7), the straight-line distance of the starting position of the mid-longitudinal section of the upper bow (P8), the ending position of the mid-longitudinal section of the upper bow (P9), the inclination angle of the ending position of the mid-longitudinal section of the upper bow (P10), and the straight-line distance of the ending position of the mid-longitudinal section of the upper bow (P11).

[0083] As shown in Figure (2-3), the characteristic parameters for constructing the bow flat bottom line include: the bow flat bottom line contraction position (P12), the straight-line distance from the bow flat bottom line contraction position (P13), the bow flat bottom line ending position (P14), the inclination angle of the bow flat bottom line ending position (P15), and the straight-line distance from the bow flat bottom line ending position (P16).

[0084] As shown in Figure (2-4), the characteristic parameters for constructing the maximum concave line of the bow include: the starting position of the maximum concave line of the bow (P17), the inclination angle of the starting position of the maximum concave line of the bow (P18), the straight-line distance of the starting position of the maximum concave line of the bow (P19), the ending position of the maximum concave line of the bow (P20), the inclination angle of the ending position of the maximum concave line of the bow (P21), and the straight-line distance of the ending position of the maximum concave line of the bow (P22).

[0085] As shown in Figure (2-5), the characteristic parameters for constructing the bow widest line include: the starting position of the bow widest line (P23), the straight-line distance from the starting position of the bow widest line (P24), the ending position of the bow widest line (P25), the straight-line distance from the ending position of the bow widest line (P26), the inclination angle of the middle position of the bow widest line (P27), and the straight-line distance of the middle position of the bow widest line (P28).

[0086] As shown in Figure (2-6), the characteristic parameters for constructing the bow waterline include: the starting position of the bow waterline (P29), the inclination angle of the starting position of the bow waterline (P30), the straight-line distance from the starting position of the bow waterline (P31), the ending position of the bow waterline (P32), the inclination angle of the ending position of the bow waterline (P33), and the straight-line distance from the ending position of the bow waterline (P34).

[0087] As shown in Figure (2-7), the characteristic parameters for constructing the bow deck line include: the starting position of the bow deck line (P35), the inclination angle of the starting position of the bow deck line (P36), the straight-line distance of the starting position of the bow deck line (P37), the ending position of the bow deck line (P38), the inclination angle of the ending position of the bow deck line (P39), and the straight-line distance of the ending position of the bow deck line (P40).

[0088] Therefore, M = 40.

[0089] S2. Select the parent ship type and measure and obtain the initial value vi of the characteristic parameter Pi corresponding to the parent ship.

[0090] S3. Construct the range of variation of characteristic parameter Pi relative to the initial value vi of the parent ship during the design process. Establish an optimization design requirement table, as shown in Table 1:

[0091] Table 1

[0092]

[0093] S4. Using the Latin hypercube design method according to equation (1), generate N sets of characteristic parameter schemes within the range of variation constructed in S3;

[0094] P i j =LHS(LBi,UBi) (1)

[0095] In the formula, i = 1, 2, ..., M, j = 1, 2, ..., N, LHS represents the Latin hypercube function, UBi is the upper limit of Pi, LBi is the lower limit of Pi, and P i j This represents the value of the characteristic parameter Pi in the j-th group of characteristic parameter schemes.

[0096] In this embodiment, N = 500.

[0097] S5. Input the N sets of feature parameter schemes generated in S4 into the parameterized model in S1 to generate N kinds of ship hull geometric model schemes.

[0098] In this embodiment, the 500 sets of feature parameter schemes generated in S4 are input into the parameterized model in S1 to generate 500 ship hull geometric model schemes, such as... Figure 3As shown, two different schemes are presented.

[0099] S6. Calculate the total resistance Fj for each of the 500 ship hull geometric model schemes obtained in S5, where j = 1, 2, ..., 500.

[0100] S7. Calculate the maximum bow impact pressure Yj for each of the 500 ship hull geometry model schemes obtained in S5, where j = 1, 2, ..., 500.

[0101] S8. Based on the total resistance and maximum bow impact pressure data collected in S6 and S7, a total resistance performance prediction model Mf and a maximum bow impact pressure prediction model My are established using neural network methods.

[0102] The specific process is as follows:

[0103] S8.1 Standardize the collected resistance and maximum bow impact pressure data from the input matrices, which are MatrixF and MatrixY, respectively;

[0104]

[0105] S8.2 Normalize MatrixF and MatrixY using the following formula;

[0106]

[0107] In the formula: MatrixF k Let MatrixY be the element in the k-th column of MatrixF. k Let be the element in the k-th column of matrix Y, where k = 1, 2, ..., 41. MatrixF k ′ and MatrixY k ′ This is the result after normalization;

[0108] S8.3, MatrixF k ′ and MatrixY k ′ The dataset is divided into training and test sets at a ratio of 70% and 30%, respectively.

[0109] S8.4 In this embodiment, the number of neurons in the input layer of the neural network is set to nm = 40, the number of neurons in the hidden layer is num = 200, and the number of neurons in the output layer is 1. ReLU is used as the activation function, and the loss function is set to the mean squared error between the neural network output and the actual calculation result.

[0110] S8.5, MatrixF k′ The obtained training set is used as training data to train the neural network model, and the trained neural network model is tested using the test set to obtain the total drag performance prediction model Mf.

[0111] S8.6, MatrixY k ′ The obtained training set is used as training data to train the neural network model, and the trained neural network model is tested using the test set to obtain the maximum impact pressure prediction model My for the bow.

[0112] Figures (4-1) and (4-2) show partial test results for the accuracy of Mf and My, respectively. As can be seen from the figures, the total ship resistance and maximum bow impact pressure calculated using Mf and My show good agreement with the simulation results.

[0113] S9. Construct an improved differential evolution algorithm to optimize the bow design of a river-sea direct-route vessel, aiming to reduce the total resistance and maximum bow impact pressure. The specific process is as follows:

[0114] S9.1 Set the maximum number of iterations of the algorithm gmax. In this embodiment, gmax is set to 280. I = 50 initial optimization design schemes for the bow of the river-sea direct-access ship are generated by randomizing the variation range of the characteristic parameters defined in S3 according to formula (2).

[0115]

[0116] In the formula: For the m-th initial river-sea direct-access vessel bow optimization design scheme, r m,g The value is a random value between 0 and 1, and g is the iteration number identifier, which is recorded as 1 in this step; in this embodiment, i = 1, 2, ... 40, m = 1, 2, ... 50, g = 1.

[0117] S9.2. According to formula (3), perform chaotic mapping on the 50 initial optimized design schemes for the bow of the randomized river-sea direct-access ship;

[0118]

[0119] In the formula: For the m-th optimized design scheme of the bow section of a river-sea direct-access vessel after chaotic mapping, r m,g It is a random value between 0 and 1.

[0120] S9.3. The 50 optimized bow design schemes of the river-sea direct-access ships after chaotic mapping are randomly divided into two groups and labeled as I1 and I2 respectively.

[0121] S9.4. The 25 optimized bow design schemes of the river-sea direct vessel after chaotic mapping marked as I1 are used to generate further optimized bow design schemes of the river-sea direct vessel using Equation (4).

[0122]

[0123] In the formula: For the n1th generated further optimized design scheme of the bow of the river-sea direct vessel, n1 = 1, 2, ..., 25, r1, r2 and r3 are random integers from 1 to 25; F is the scaling factor with a value of 0.5.

[0124] S9.5. The 25 optimized bow design schemes of the river-sea direct-access vessel after chaotic mapping marked as I2 are used to generate further optimized bow design schemes of the river-sea direct-access vessel using Equation (5).

[0125]

[0126] In the formula: For the n2th generated further optimized design scheme of the bow of the river-sea direct vessel, n2 = 26, 27, ..., 50, r4, r5, r6, r7 and r8 are random integers from I / 2+1 to I, and F is a scaling factor of 0.5.

[0127] S9.6, Merging and for m = 1, 2, ..., 50;

[0128] S9.7. The further optimized design scheme of the bow section of the river-sea direct vessel is refined by formula (6) to generate a refined optimized design scheme of the bow section of the river-sea direct vessel.

[0129]

[0130] In the formula, i rand is a random integer between 1 and 40, and r is a random value between 0 and 1.

[0131] S9.8, will The input is obtained in Mf of S8 The total resistance performance of the generated initial and refined river-sea direct vessel bow optimization design schemes are calculated using the ship total resistance performance prediction model Mf constructed in S8.

[0132] S9.9, will Get the input into Mp in S8 The maximum bow impact pressure prediction model My, constructed in S8, is used to calculate the maximum bow impact pressure of the generated initial and refined river-sea direct-route vessel bow optimization design schemes.

[0133] S9.10, according to formula (7) The corresponding optimized design scheme for the bow of the ship is selected, and the selected scheme is assigned a value. The iteration number g is then updated to iter, where iter = g + 1. This means that the solutions with lower total resistance and maximum bow impact pressure generated during this iteration are retained for the next iteration calculation.

[0134]

[0135] In the formula, This represents the initial optimized design scheme for the bow of the river-sea direct-access vessel in the next iteration calculation process.

[0136] S9.11, Determine if iter is greater than gmax. If less, jump to S9.3 and set g in S9.3 to the value of iter, then perform the next iteration. If greater, then... The scheme with the lowest total resistance and the lowest maximum impact pressure on the ship's bow is determined as the optimal design.

[0137] The solution process in this embodiment is as follows: Figure 5 As shown, Figure 5 As can be seen, the algorithm gradually converges after multiple iterations, yielding the design results of the embodiment.

[0138] Finally, in this embodiment, the bow shape of the river-sea direct vessel obtained through iterative calculation is as follows: Figure 6 As shown, this can be described as a composite bow. Compared to bulbous bows and typical upright bows, the wetted surface area below the waterline is reduced, which helps to further reduce the frictional drag experienced by the ship. In addition, compared to the X-Bow, the bow is more pointed and slender, which helps to further reduce the impact of wave-breaking drag.

[0139] To verify the effectiveness of the bow design method of this invention for river-sea direct vessels, the drag performance and impact performance of the composite bow designed in this embodiment are analyzed below.

[0140] (1) Drag performance

[0141] The drag performance of the bow of the river-sea direct vessel constructed in the embodiment was analyzed using three other bow designs of the same scale as comparative designs to illustrate the effectiveness of the method of the present invention. To more clearly demonstrate the different bow designs, in Figure 7 A 3D model diagram is provided.

[0142] The total resistance of the river-sea direct vessel bow design constructed by the above three schemes and embodiments at the design speed (10.8 knots) was predicted using the CFD method. The prediction results are shown in Table 2. It can be seen that the resistance performance of the river-sea direct vessel bow design (composite bow design) constructed by the method of the present invention is better than that of the existing conventional bow design.

[0143] Table 2 Optimized ship resistance data at various speeds

[0144] serial number Total resistance N Conventional bulbous bow design 16.688 Conventional upright bow design 15.164 Vertical bow design with waterline contraction 16.231 Composite bow design 14.709

[0145] (2) Impact performance

[0146] To address the problem of determining the maximum impact pressure on the bow in this invention, Ansys / LS-DYNA was used for finite element analysis. A typical bulbous bow model was used as a comparative example. A computational model was constructed for the bow (composite bow design) of a river-sea direct-transport vessel in the example, and pressure measurement points were set up. Figure 8 As shown.

[0147] Figure 9 The comparison results of pressure peak values ​​at different measuring points for the composite bow and bulbous bow show that the pressure peak value of the bulbous bow is higher than that of the composite bow. Furthermore, due to the rapid increase in the sloping angle along the positive X-axis for both the bulbous bow and the composite bow, the vertical entry slamming pressure of the two structures will show a rapid decrease along the positive X-axis, and then tend to stabilize and decrease.

[0148] like Figure 10-12 The image shows a comparative analysis of the peak pressure values ​​at three measurement points along the positive Z-axis for both composite and bulbous bow types. Figure 10 It can be seen that at low altitudes, the peak slamming pressure at the first set of measuring points is higher for the bulbous bow than for the composite bow. However, because the composite bow exhibits a more gradual change in elevation profile than the bulbous bow, its peak slamming pressure at higher measuring points is higher. Figure 11 It can be seen that the peak impact pressure at the second set of measuring points follows the same pattern with height. Figure 12 It can be seen that the measuring points in the third group are at a more distant location (closer to the bow), and the peak impact pressure of the bulbous bow is greater than that of the composite bow.

[0149] In summary, simulation models of bow impact pressure for different bow structures were established using ANSYS / LS-DYNA software. The distribution of impact loads for bulbous bows and composite bows was analyzed. Furthermore, the variation characteristics of the peak impact pressure along the longitudinal direction of the hull bottom and the height direction of the two bow types were compared. The conclusion is that, overall, the composite bow has a better impact resistance than the bulbous bow.

[0150] It should be noted that, depending on the implementation needs, the various steps / components described in this application can be broken down into more steps / components, or two or more steps / components or parts of the operation of steps / components can be combined into new steps / components to achieve the purpose of this invention.

[0151] The order of the steps in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0152] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A bow design method applicable to river-sea direct vessels, characterized in that, Includes the following steps: S1. Define and construct key feature control lines through feature parameters Pi to establish a parameterized model of the bow; the key feature control lines include: bow mid-longitudinal section line, upper bow mid-longitudinal section line, bow horizontal line, bow maximum concave line, bow widest line, bow waterline, and bow deck line; i = 1, 2, ..., M, where M is the number of feature parameters; S2. Select the parent ship type and measure and obtain the initial value vi of the parent ship corresponding to the characteristic parameter Pi; S3. Construct the range of variation of characteristic parameter Pi relative to the initial value vi of the parent ship during the design process; S4. Using the Latin hypercube design method according to equation (1), generate N sets of characteristic parameter schemes within the range of variation constructed in S3; In the formula, i = 1, 2, ..., M, j = 1, 2, ..., N, LHS represents the Latin hypercube function, UBi is the upper limit of Pi, and LBi is the lower limit of Pi. This represents the value of the characteristic parameter Pi in the j-th group of characteristic parameter schemes; S5. Input the N sets of feature parameter schemes generated in S4 into the parameterized model in S1 to generate N kinds of ship hull geometric model schemes. S6. Calculate the total resistance Fj for each of the N ship hull geometric model schemes obtained in S5, j = 1, 2, ..., N; S7. Calculate the maximum bow impact pressure Yj for the N hull geometry model schemes obtained in S5, j = 1, 2, ..., N; S8. Based on the total resistance and maximum bow impact pressure data collected in S6 and S7, establish a total resistance performance prediction model Mf and a maximum bow impact pressure prediction model My respectively using neural network methods. S9. Construct an improved differential evolution algorithm to optimize the bow design of a river-sea direct vessel with the goal of reducing the total resistance of the ship and the maximum impact pressure on the bow, and obtain the design results.

2. The bow design method for river-sea direct vessels according to claim 1, characterized in that, The characteristic parameters for constructing the longitudinal section of the bow include: the position of the bow bottom rise (P1), the position of the foremost edge of the bow (P2), the bow bottom rise angle (P3), the bow inclination angle (P4), the straight distance of the bow bottom rise (P5), and the straight distance of the bow inclination (P6).

3. The bow design method for river-sea direct vessels according to claim 1, characterized in that, The characteristic parameters for constructing the upper bow longitudinal section include: the inclination angle of the starting position of the upper bow longitudinal section (P7), the straight-line distance of the starting position of the upper bow longitudinal section (P8), the ending position of the upper bow longitudinal section (P9), the inclination angle of the ending position of the upper bow longitudinal section (P10), and the straight-line distance of the ending position of the upper bow longitudinal section (P11).

4. The bow design method for river-sea direct vessels according to claim 1, characterized in that, The characteristic parameters for constructing the bow flat bottom line include: bow flat bottom line contraction position (P12), bow flat bottom line contraction position straight line distance (P13), bow flat bottom line end position (P14), bow flat bottom line end position inclination angle (P15), and bow flat bottom line end position straight line distance (P16).

5. The bow design method for river-sea direct vessels according to claim 1, characterized in that, The characteristic parameters for constructing the maximum concave line of the bow include: the starting position of the maximum concave line of the bow (P17), the inclination angle of the starting position of the maximum concave line of the bow (P18), the straight-line distance of the starting position of the maximum concave line of the bow (P19), the ending position of the maximum concave line of the bow (P20), the inclination angle of the ending position of the maximum concave line of the bow (P21), and the straight-line distance of the ending position of the maximum concave line of the bow (P22).

6. The bow design method for river-sea direct vessels according to claim 1, characterized in that, The characteristic parameters for constructing the bow widest line include: the starting position of the bow widest line (P23), the straight-line distance from the starting position of the bow widest line (P24), the ending position of the bow widest line (P25), the straight-line distance from the ending position of the bow widest line (P26), the inclination angle of the middle position of the bow widest line (P27), and the straight-line distance of the middle position of the bow widest line (P28).

7. The bow design method for river-sea direct vessels according to claim 1, characterized in that, The characteristic parameters for constructing the bow waterline include: the starting position of the bow waterline (P29), the inclination angle of the starting position of the bow waterline (P30), the straight-line distance of the starting position of the bow waterline (P31), the ending position of the bow waterline (P32), the inclination angle of the ending position of the bow waterline (P33), and the straight-line distance of the ending position of the bow waterline (P34).

8. The bow design method for river-sea direct vessels according to claim 1, characterized in that, The characteristic parameters for constructing the bow deck line include: the starting position of the bow deck line (P35), the inclination angle of the starting position of the bow deck line (P36), the straight-line distance of the starting position of the bow deck line (P37), the ending position of the bow deck line (P38), the inclination angle of the ending position of the bow deck line (P39), and the straight-line distance of the ending position of the bow deck line (P40).

9. The bow design method for river-sea direct vessels according to claim 1, characterized in that, In step S8, the total resistance performance prediction model Mf and the maximum bow impact pressure prediction model My are established, which specifically includes the following steps: S8.

1. Standardize the collected resistance and maximum bow impact pressure data into input matrices, namely MatrixF and MatrixY; S8.2 Normalize MatrixF and MatrixY using the following formula; In the formula: MatrixF k Let MatrixY be the element in the k-th column of MatrixF. k Let MatrixY be the element in the k-th column, where k = 1, 2, ..., M+1, and MatrixF be the element in the k-th column. k ′ and MatrixY k ′ represents the normalized result; S8.3, MatrixF k ′ and MatrixY k The dataset is divided into training and testing sets. S8.4 Design a neural network structure, including an input layer, a hidden layer, and an output layer; set the number of neurons in the input layer to nm, the number of neurons in the hidden layer to num, and the number of neurons in the output layer to 1; use ReLU as the activation function and set the loss function to the mean squared error between the neural network output and the actual calculation result; S8.5, MatrixF k The training set obtained by dividing the data into ' and ' is used as training data to train the neural network model. The trained neural network model is then tested using the test set to obtain the total drag performance prediction model Mf. S8.6, MatrixY k The training set obtained by dividing the data into ' and ' is used as training data to train the neural network model. The trained neural network model is then tested using the test set to obtain the maximum impact pressure prediction model My for the bow.

10. The bow design method for river-sea direct vessels according to claim 1, characterized in that, In step S9, an improved differential evolution algorithm is constructed to optimize the bow design of a river-sea direct-route vessel, aiming to reduce the total resistance of the ship and the maximum impact pressure on the bow. The specific steps are as follows: S9.1 Set the maximum number of iterations of the algorithm gmax, and generate I initial optimized design schemes for the bow of the river-sea direct-access vessel by randomizing the variation range of the characteristic parameters defined in S3 according to equation (2); In the formula: i = 1, 2, ..., M, m = 1, 2, ..., I, For the m-th initial river-sea direct-access vessel bow optimization design scheme, r m,g The value is a random value between 0 and 1, and g is the iteration number identifier, which is recorded as 1 in this step; S9.

2. According to Equation (3), perform chaotic mapping on the I initial optimized design schemes for the bow of the randomized river-sea direct-access vessel; In the formula: For the m-th optimized design scheme of the bow section of a river-sea direct-access vessel after chaotic mapping, r m,g A random value between 0 and 1; S9.

3. Randomly divide the I optimized design schemes for the bow of the river-sea direct-access vessel after chaotic mapping into two groups, and label them as I1 and I2 respectively. S9.

4. The I / 2 optimized design schemes for the bow of the river-sea direct vessel after chaotic mapping marked as I1 are used to generate a further optimized design scheme for the bow of the river-sea direct vessel using Equation (4). In the formula: n1 = 1, 2, ..., I / 2, For the n1th generated further optimized design scheme of the bow of the river-sea direct vessel, r1, r2 and r3 are random integers from 1 to 1 / 2; F is the scaling factor with a value of 0.5; S9.

5. The I / 2 river-sea direct-ship bow optimization design schemes marked as I2 after chaotic mapping are used to generate a further river-sea direct-ship bow optimization design scheme using Equation (5). In the formula: n2=I / 2+1,…,I, For the n2th generated further optimized design scheme of the bow of the river-sea direct vessel, r4, r5, r6, r7 and r8 are random integers from I / 2+1 to I, and F is a scaling factor of 0.5; S9.6, Merging and for S9.

7. The further optimized design scheme of the bow section of the river-sea direct vessel is refined by formula (6) to generate a refined optimized design scheme of the bow section of the river-sea direct vessel. In the formula, i rand is a random integer from 1 to M, and r is a random value from 0 to 1; S9.8, will The input is obtained in Mf of S8 The total resistance performance of the generated initial and refined river-sea direct vessel bow optimization design schemes are calculated using the ship total resistance performance prediction model Mf constructed in S8. S9.9, will The input is obtained in My in S8 Even if we use the maximum bow impact pressure prediction model My built in S8 to calculate the maximum bow impact pressure of the generated initial and refined river-sea direct-ship bow optimization design schemes; S9.10, according to formula (7) The corresponding optimized design scheme for the bow of the ship is selected, and the selected scheme is assigned a value. The iteration number marker g is updated to iter, and iter = g + 1, meaning that the schemes with smaller total resistance and maximum bow impact pressure generated in this iteration are retained for the next iteration calculation; In the formula, This represents the initial optimized design scheme for the bow of the river-sea direct-access vessel in the next iteration calculation process; S9.11, Determine if iter is greater than gmax. If less, jump to S9.3 and set g in S9.3 to the value of iter for the next iteration. If greater, then... The scheme with the lowest total resistance and the lowest maximum impact pressure on the ship's bow is determined as the optimal design.

Images