An Adaptive Simulation Method for the Flow Field of a Channel Boat in Navigation

Abstract

This application belongs to the technical field of ship navigation flow field calculation, and provides an adaptive simulation method for the navigation flow field of a channel boat, including the following steps: establishing a fluid model and a channel boat model and setting a calculation domain; setting an initial flow field distribution, the navigation speed of the channel boat and an initial mesh division scheme; dividing the mesh of the fluid model based on the initial mesh division scheme; capturing the gas-liquid interface morphology during the navigation process of the channel boat and tracking droplets and bubbles; dividing the droplet splash zone and the gas-liquid interaction zone based on the tracking results of the droplets and bubbles; performing adaptive mesh division on the droplet splash zone and the gas-liquid interaction zone to obtain a modified mesh division scheme; numerically simulating the time evolution process of the navigation flow field of the channel boat based on the modified mesh division scheme. The method provided by this application ensures the calculation accuracy while optimizing the calculation efficiency, and significantly improves the calculation stability, providing effective support for the accurate modeling of gas-liquid two-phase flow and the structural optimization of channel boats.

Description

Technical Field

[0001] This application belongs to the technical field of ship navigation flow field calculation, and specifically relates to an adaptive simulation method for the navigation flow field of a channeled ship. Background Art

[0002] The channeled ship is a new type of high-speed ship. Compared with traditional ships, the most significant improvement in its structure lies in the unique grooved structure at the bow of the ship. For example, patents CN108945280A, CN202609027U, etc. have all disclosed high-speed ships with bow channels. When the water flow in the grooved area enters the channel, due to pressure changes and velocity differences, part of the air is sucked into the water flow, enters the underwater area, and forms bubbles in the water. These bubbles interact with the water flow and change the flow properties.

[0003] Due to the special channel structure at the bow of the channeled ship, when it sails at a relatively high speed, complex water-air interaction phenomena will occur near the channel. For example, because the mixing speed of water flow and air is relatively fast, the bubbles generated near the liquid-gas interface do not remain stable, and will break, split, and spread, forming a group of fine bubbles. These bubbles not only change the flow characteristics of water, but also cause fluctuations in the gas-liquid interface in the water and increase the phenomenon of liquid splashing. As a result, when the channeled ship sails at high speed, it will be affected by fluids, gases, and splashing droplets at the same time. Obviously, only by comprehensively analyzing various factors affecting the navigation state of the channeled ship can the structure of the channeled ship be optimized targeted, so as to further improve the ship speed and navigation stability of the channeled ship.

[0004] However, there is currently no method for accurately numerically simulating the navigation flow field of a channeled ship for its special structure. The reasons are as follows: The bow structure of the channeled ship is special, and the water-air interaction process generated when the water flow enters the channel is more complex than that of conventional ships. In the computational domain, especially in the computational area around the channeled ship, there will be a liquid flow field and bubbles distributed therein, a gas flow field and splashing droplets distributed therein at the same time, and the flow field, bubbles, and droplets interact with each other and jointly affect the navigation of the channeled ship. Although the numerical simulation accuracy can be improved by the conventional method of increasing the grid density, this will inevitably greatly increase the consumption of computing resources; in addition, from the perspective of the stability of numerical simulation, the liquid splashing phenomenon often makes the simulation process diverge, which also causes the extension of simulation time and the decrease of simulation accuracy, thus interfering with the accurate simulation of the water-air interaction phenomenon. Summary of the Invention

[0005] In order to solve the problems existing in the above-mentioned prior art, this application provides an adaptive simulation method for the navigation flow field of a channeled ship through embodiments to efficiently and accurately analyze the flow field during the navigation of the channeled ship.

[0006] The adaptive simulation method for the flow field of a channel boat provided by this application includes the following steps:

[0007] S1. Establish a fluid model and a channel boat model and set the computational domain. The fluid model includes a liquid part and a gas part;

[0008] S2. Set the initial flow field distribution of the fluid model, the sailing speed of the channel boat, and the initial mesh division scheme;

[0009] S3. Perform mesh division on the fluid model based on the initial mesh division scheme;

[0010] S4. Based on the initial mesh division scheme, capture the gas-liquid interface morphology during the sailing process of the channel boat, and track the droplets and bubbles caused by the change of the gas-liquid interface morphology;

[0011] S5. Based on the tracking results of the droplets and bubbles, divide the droplet splash zone and the gas-liquid interaction zone;

[0012] S6. Use the first mesh division scheme and the second mesh division scheme to perform adaptive mesh division on the droplet splash zone and the gas-liquid interaction zone respectively to obtain a modified mesh division scheme. Among them, the mesh density of the droplet splash zone divided according to the first mesh division scheme is less than the mesh density divided according to the initial mesh division scheme, and the mesh density of the gas-liquid interaction zone divided according to the second mesh division scheme is greater than the mesh density divided according to the initial mesh division scheme;

[0013] S7. Perform numerical simulation on the time evolution process of the flow field of the channel boat based on the modified mesh division scheme.

[0014] Preferably, in the initial mesh division scheme, the mesh density of the gas-liquid interface region and the region where the fluid intersects with the channel boat is greater than that of other regions.

[0015] Further, step S4 is implemented by the Level Set algorithm, which specifically includes the following steps:

[0016] S41. Define a marker function, and the zero isosurface of the marker function is the gas-liquid interface;

[0017] S42. Solve the evolution equation of the marker function and capture the gas-liquid interface morphology and the real-time flow field distribution of the fluid model at several sampling moments;

[0018] S43. Through interface reconstruction and flow field evolution of the gas-liquid interface, generate bubble and droplet structures in the liquid part and the gas part respectively based on the interface instability mechanism, and track their distributions at several sampling moments.

[0019] Further, step S5 includes the following steps:

[0020] S51, divide an initial grid correction region in the computational domain around the channel boat according to the tracking results of the bubble and droplet structures, where the initial grid correction region is the union of regions where at least one bubble structure and at least one droplet structure appear among the several sampling moments;

[0021] S52, divide several statistical grids in the initial grid correction region according to a preset statistical grid density;

[0022] S53, based on the tracking results of the droplet structures at several sampling moments, statistically analyze the droplet splash trend in each statistical grid;

[0023] S54, divide a droplet splash region from the initial grid correction region based on the statistical results of the droplet splash trend;

[0024] S55, divide a gas-liquid interaction region from the initial grid correction region based on the division result of the droplet splash region.

[0025] Preferably, for each statistical grid, statistically analyze its droplet splash trend based on the following formula:

[0026] ,

[0027] where, is the splash separation index, is the total number of sampling moments, is the total number of droplets that appear in this statistical grid at the th sampling moment, is the velocity vector of the th droplet that appears in this statistical grid at the th sampling moment, is the unit vector of the th droplet that appears in this statistical grid at the th sampling moment relative to the channel boat, and its direction is from the centroid position of the bow of the channel boat to the droplet position, is the maximum value function, is a null value.

[0028] Further, the step of dividing a droplet splash region from the initial grid correction region based on the statistical results of the droplet splash trend is specifically: divide the statistical grids in the initial grid correction region where the value is greater than a preset threshold into the droplet splash region.

[0029] Preferably, the preset threshold increases with the increase of the relative speed of the channel boat, and the maximum limit it can reach is 1.

[0030] Further, based on the division result of the droplet splash zone, a gas-liquid interaction zone is divided from the initial mesh correction region, specifically: the gas-liquid interaction zone is divided from the remaining region after removing the droplet splash zone from the initial mesh correction region.

[0031] Preferably, the initial mesh correction region is larger than the union of the droplet splash zone and the gas-liquid interaction zone, and the remaining region after removing the droplet splash zone and the gas-liquid interaction zone from the initial mesh correction region is located behind the channel boat.

[0032] Preferably, the gas-liquid interaction zone is further divided into a first gas-liquid interaction zone and a second gas-liquid interaction zone;

[0033] The first gas-liquid interaction zone is a region where the value is not and less than or equal to the preset threshold, and the second gas-liquid interaction zone is a region where the value is ;

[0034] The grid density of the first gas-liquid interaction zone divided according to the second grid division scheme is greater than the grid density of the second gas-liquid interaction zone divided according to the second grid division scheme.

[0035] Preferably, it further includes the step of performing a grid density smoothing operation on the droplet splash zone and other regions or on the gas-liquid interaction zone and other regions in the modified grid division scheme.

[0036] The adaptive simulation method for the flow field of a channel boat provided by the embodiment of the present application first comprehensively considers bubble breakage, droplet splash during the high-speed navigation of the channel boat and their impacts on the navigation process of the channel boat, tracks the gas-liquid interface to accurately obtain the dynamic process of the two-phase flow, and then uses the adaptive grid division technology based on the splash separation index (SSI) to dynamically adjust the grid size according to the flow field changes, coarsen the grid in the droplet splash region far from the boat body to achieve the purpose of dissipating unnecessary droplets, and refine the grid in the regions where water and gas interact violently such as the grooved region, thereby optimizing the calculation efficiency and ensuring the calculation accuracy, and significantly improving the stability of the calculation, providing effective support for the accurate modeling of the gas-liquid two-phase flow and the structural optimization of the channel boat. Description of the Drawings

[0037] Figure 1 is a flowchart of the adaptive simulation method for the flow field of a channel boat provided by the embodiment of the present application;

[0038] Figure 2 is a structural schematic diagram of the channel boat model in the embodiment of the present application;

[0039] Figure 3It is the specific implementation flowchart of step S4 in the embodiments of this application;

[0040] Figure 4 It is the specific implementation flowchart of step S5 in the embodiments of this application;

[0041] Figure 5 It is the schematic diagram of the preset threshold varying with the relative ship speed in the embodiments of this application;

[0042] Figure 6 It is the schematic diagram of meshing the fluid model using the initial mesh division scheme in the specific embodiments of this application;

[0043] Figure 7 It is the schematic diagram of meshing the fluid model using the modified mesh division scheme in the specific embodiments of this application;

[0044] Figure 8 It is the side view of the flow field of the channel boat during navigation obtained by meshing the fluid model using the modified mesh division scheme in the specific embodiments of this application;

[0045] Figure 9 It is the top view of the flow field of the channel boat during navigation obtained by meshing the fluid model using the modified mesh division scheme in the specific embodiments of this application. Specific implementation manner

[0046] Hereinafter, this application will be further described based on the preferred implementation manners and with reference to the accompanying drawings.

[0047] As analyzed in the background art, there is currently no method for accurately numerically simulating the flow field of a channel boat. Therefore, it is difficult to optimize the structure of the channel boat targeted based on the simulation results. For this reason, this application provides an adaptive simulation method for the flow field of a channel boat through embodiments, referring to Figure 1 , and this simulation method includes the following steps:

[0048] S1. Establish a fluid model and a channel boat model and set the computational domain, where the fluid model includes a liquid part and a gas part;

[0049] S2. Set the initial flow field distribution of the fluid model, the ship speed of the channel boat, and the initial mesh division scheme;

[0050] S3. Mesh the fluid model based on the initial mesh division scheme;

[0051] S4. Capture the gas-liquid interface morphology during the navigation of the channel boat based on the initial mesh division scheme, and track the liquid droplets and bubbles caused by the change of the gas-liquid interface morphology;

[0052] S5. Segment the droplet splash region and the gas-liquid interaction region based on the tracking results of the droplets and bubbles.

[0053] S6. Use the first mesh division scheme and the second mesh division scheme to perform adaptive mesh division on the droplet splash region and the gas-liquid interaction region respectively to obtain a modified mesh division scheme. Among them, the mesh density of the droplet splash region divided according to the first mesh division scheme is less than the mesh density divided according to the initial mesh division scheme, and the mesh density of the gas-liquid interaction region divided according to the second mesh division scheme is greater than the mesh density divided according to the initial mesh division scheme.

[0054] S7. Numerically simulate the time evolution process of the flow field of the channel boat based on the modified mesh division scheme.

[0055] The following combines the drawings and specific embodiments to elaborate on the steps of this method in detail.

[0056] Step S1:

[0057] Step S1 is used to establish relevant models for numerically simulating the flow field during navigation and set the numerical simulation environment.

[0058] In some embodiments, a fluid model can be established according to indicators such as the simulation accuracy requirements. Among them, the fluid model includes a liquid part located below and a gas part located above, and the interface between the two parts is called the gas-liquid interface, which continuously changes according to the navigation conditions of the channel boat during subsequent numerical simulations.

[0059] In some embodiments, a channel boat model can be established according to the structural parameters of the channel boat used for analysis. Obviously, the channel boat model for numerical simulation only needs to finely characterize the hull shell shape of the part located in the liquid part and near the gas-liquid interface, and set the hull shell of the above parts as a rigid body to provide a rigid constraint at the junction of the hull and the fluid. In this way, the complexity of the model can be greatly simplified without losing the numerical simulation accuracy. Figure 2 Shows a schematic diagram of the channel boat model for numerical simulation in some specific embodiments, as Figure 2 shown. The structure of the lower shell of this channel boat model is relatively fine and can accurately characterize the M-shaped double-channel structure located at the bow of the boat, and its upper part is simply set as a flat-top structure.

[0060] After obtaining the fluid model and the channel boat model, the computational domain can be further set. The size of the computational domain can be determined according to information such as the simulation accuracy, the fluid flow rate used for simulation, and the ship speed. For example, when the flow rate of the liquid part or the ship speed is large, the size of the computational domain can be appropriately enlarged, and vice versa, the size of the computational domain can be reduced. The above setting methods are well known to those skilled in the art and will not be elaborated here.

[0061] Steps S2 - S3:

[0062] Step S2 is used for initializing the numerical simulation. Specifically, in this step, the initial flow field distributions can be set for the liquid part and the gas part of the fluid model respectively. Among them, the flow field distribution can include the fluid flow velocity at each position in the fluid model. Obviously, the flow field distribution of the liquid part corresponds to the water flow velocity, and the flow field distribution of the gas part corresponds to the wind speed. Here, the flow field distribution of the fluid model can also include information such as the pressure or pressure at each position in the fluid model. Through the above information, the water resistance and wind resistance suffered by the channel boat during navigation can be calculated.

[0063] In step S2, the boat speed of the channel boat can also be set. In some embodiments, a coordinate system can be established with the channel boat as the origin, and the computational domain can be bound to the channel boat to keep the channel boat stationary relative to the computational domain. The fluid model is driven to move in the computational domain with the combined velocity of the flow velocities of the liquid part and the gas part and the boat speed, so as to facilitate subsequent numerical calculations.

[0064] Step S2 is also used to set the grid division scheme for initial grid division of the fluid model in the computational domain. In the embodiments of the present application, the grid division scheme set in step S2 is called the initial grid division scheme.

[0065] The initial grid division scheme can be formulated according to the grid setting schemes familiar to those skilled in the art. For example, in some preferred embodiments, the gas - liquid interface region (i.e., the gas - liquid interface and the regions within a certain distance above and below it), and the fluid - channel boat interface region are respectively subjected to grid encryption processing to ensure the accuracy of numerical simulation in the above regions.

[0066] After completing the setting of the initial grid division scheme, in step S3, the fluid model in the computational domain can be grid - divided according to the initial grid division scheme.

[0067] Step S4:

[0068] Step S4 is used for preliminarily simulating the flow field of the channel boat's navigation of the fluid model divided by the initial grid division scheme to obtain the changes in the gas - liquid interface caused by the navigation of the channel boat and track the bubbles and droplets caused by the above - mentioned changes in the gas - liquid interface. The result of this preliminary simulation will be used as the basis for correcting the grid division scheme in the subsequent step S5.

[0069] As Figure 3 shown, in the embodiments of the present application, step S4 can be implemented by the Level Set algorithm, which specifically includes the following steps:

[0070] S41. Define a marker function, where the zero isosurface of the marker function is the gas-liquid interface.

[0071] In the Level Set algorithm, by defining a marker function to describe the interface between different phase fluids, it can accurately handle the shape change of the interface and is suitable for simulating complex fluid dynamics. Especially in phenomena such as splashing droplets and bubble breakup, the part where < 0 can be regarded as the liquid phase region, and the region where > 0 can be regarded as the gas phase region. Obviously, the zero isosurface of the marker function ( = 0) represents the gas-liquid interface.

[0072] S42. Solve the evolution equation of the marker function and capture the gas-liquid interface morphology and the real-time flow field distribution of the fluid model at several sampling times.

[0073] The Level Set algorithm captures the dynamic changes of the interface by solving the evolution equation of the above-mentioned marker function. The evolution equation of the marker function is shown as follows:

[0074] (1),

[0075] In the above formula, is the velocity field of the fluid, is the time variable, is the gradient operator. This equation indicates that the marker function evolves as the fluid moves.

[0076] By solving the evolution equation shown in (1) during a preset sampling period, the corresponding at several sampling times can be obtained, so as to capture the morphology of its zero isosurface (i.e., the gas-liquid interface). At the same time, the real-time distribution of the flow field at each sampling time can also be obtained.

[0077] S43. Through interface reconstruction and flow field evolution of the gas-liquid interface, generate bubble and droplet structures in the liquid part and gas part respectively based on the interface instability mechanism and track their distributions at several sampling times.

[0078] When the interface is sufficiently perturbed and the distance changes, it will cause instability at the gas-liquid interface. The above instability mechanism causes the gas-phase component to enter the liquid-phase component, thus generating multiple bubbles in the liquid region. The bubbles may also break up and generate multiple droplets. In the Level Set algorithm, by performing interface reconstruction and flow field evolution on the gas-liquid interface, the dynamic evolution of the interface can be accurately captured, and the bubble breakup process can be accurately simulated.

[0079] In some specific embodiments, the process of the Level Set algorithm simulating the dynamic changes of the interfaces of liquid droplets and bubbles in a gas-liquid two-phase flow includes the following operations:

[0080] Interface deformation tracking: Calculate the normal velocity of the gas-liquid interface through the evolution of the marker function, and update the position of the zero isosurface;

[0081] Topological change processing: When the interface curvature exceeds the critical value, the interface breaks, generating bubbles in the liquid part of the fluid model and droplets in the gas part of the fluid model.

[0082] Interface reconstruction: Re-initialize the marker function regularly to maintain numerical stability.

[0083] Through the above steps, the reconstructed gas-liquid interface, as well as the shapes and trajectories of the generated bubbles and droplets, can be accurately tracked at each sampling moment.

[0084] Steps S5 - S6:

[0085] In the embodiments of the present application, in step S5, by statistically analyzing the droplet and bubble structures tracked at multiple sampling moments, a droplet splash zone and a gas-liquid interaction zone are further divided in the computational domain. In step S6, grid sparsification processing and grid densification processing are respectively performed on the droplet splash zone and the gas-liquid interaction zone, thereby realizing the correction of the initial grid division scheme.

[0086] The reasons for further dividing the droplet splash zone and the gas-liquid interaction zone in the computational domain and respectively performing grid sparsification and grid densification on them are as follows:

[0087] First, when the channel boat sails at a high speed, the bubbles formed by the gas compressed into the channel do not remain stable. They will continuously break, split, and spread, forming a group of fine bubbles. These bubbles not only affect the hydrodynamic characteristics of water but also increase the phenomenon of liquid splash, resulting in the coexistence of liquid-phase fluid, gas-phase fluid, bubbles in the liquid-phase fluid, and droplets in the gas-phase fluid in some regions. Since the bubbles generated by this gas-liquid interaction can further lift the hull to achieve a drag reduction effect, it is necessary to analyze their evolution more accurately. Second, the applicant found during the simulation that when calculating the interaction between the splashing droplets, bubbles, and the hull, the large-scale calculation of droplet evolution will bring certain instability to the numerical simulation, often causing the simulation process to diverge. Although the grid can be densified for all regions where there are bubbles or droplets, this method will cause a huge increase in the amount of calculation, resulting in a significant increase in calculation time and calculation cost.

[0088] To this end, the present application adopts a different grid construction method to solve the above problems: First, an initial region is delimited based on the distributions of droplets and bubbles at several sampling moments. Then, the evolution behaviors of the droplets in this region are statistically analyzed to determine the distribution region of the droplets far from the hull. For this part of the region, since the movement and evolution of its droplets do not affect the navigation of the channel boat, there is no need to consider its interaction with the channel boat. Therefore, the grid of this part is coarsened, which not only reduces the computational complexity of this part of the region but also further ensures that it will not affect the computational stability. For the region where there are droplets with a movement state not much different from that of the channel boat, and the region where there are droplets mixed with bubbles, since the droplets therein involve complex interaction calculations, it is necessary to perform grid encryption processing on them to improve the computational accuracy and avoid numerical divergence.

[0089] Reference Figure 4 , in the embodiment of the present application, step S5 includes the following steps:

[0090] S51, divide an initial grid correction region in the computational domain around the channel boat according to the tracking results of the bubble and droplet structures, and the initial grid correction region is the union of the regions where at least one bubble structure and at least one droplet structure appear in at least one of the several sampling moments.

[0091] S52, divide several statistical grids in the initial grid correction region according to a preset statistical grid density.

[0092] S53, based on the tracking results of the droplet structures at several sampling moments, statistically analyze the droplet splashing trend in each statistical grid.

[0093] In some embodiments, the splashing separation index shown by the following formula can be used to characterize the statistical results of the droplet splashing trend in each statistical grid:

[0094] (2),

[0095] where, is the total number of sampling moments, is the total number of droplets that appear in this statistical grid at the th sampling moment, is the velocity vector of the th droplet that appears in this statistical grid at the th sampling moment, is the unit vector of the th droplet that appears in this statistical grid at the th sampling moment relative to the channel boat, and its direction is from the centroid position of the bow of the channel boat to the droplet position. is the maximum value function, is a null value.

[0096] According to the meanings of the above variables, for any statistical grid, represents the component of the velocity of each droplet in the direction of the bow of the boat, and this component is used to measure the magnitude of the velocity of the droplet in the direction away from the boat body, The exponent represents the ratio of the component of the droplet moving in the direction away from the boat body in the statistical grid to the total velocity of the droplet:

[0097] When approaches 1, it means that most of the movement directions of the droplets are away from the boat body, that is, most of the splashing droplets are away from the boat body, which indicates that the splashing phenomenon is occurring and the droplets have an obvious splashing trend;

[0098] When approaches 0, it means that the droplets do not have an obvious movement trend away from the boat body. It may be due to the water-vapor interaction phenomenon or other disturbances, and the droplets still remain near the boat body, or the movement direction is not obvious;

[0099] When is between 0 and 1, it means that the droplets have both a trend to move away from the boat body and may also have a movement component approaching the boat body. This situation may involve complex water-vapor interaction or partial droplet backflow phenomenon;

[0100] When , it means that at each sampling moment, no droplets have appeared in the statistical grid, and only bubbles exist. In this case, the value of .

[0101] S54. Based on the statistical results of the droplet splashing trend, a droplet splashing area is divided from the initial grid correction area.

[0102] Specifically, after obtaining the value of each statistical grid through step S53, in step S54, a preset threshold between 0 and 1, such as 0.7, can be set to determine the statistical grids in the initial grid correction area whose value is greater than the preset threshold as the droplet splashing area.

[0103] In some preferred embodiments, the preset threshold will increase with the increase in the relative speed of the channel boat (i.e., the result of the vector synthesis of the flow velocity and the boat speed). This is because as the boat speed continues to increase, more droplets may participate in the interaction with the boat body. Therefore, the threshold should be increased to reduce the size of the region with coarse meshing, which also meets the requirement of performing fine calculations on more regions during high-speed movement. However, obviously, no matter how it is increased, the maximum limit that the preset threshold can reach is 1.

[0104] Therefore, a preset threshold function as shown in the following formula can be constructed to determine the preset threshold corresponding to different relative speeds:

[0105] (3),

[0106] where, is the preset threshold, is the relative speed (in knots), is the proportionality coefficient. The above formula indicates that the minimum value of the preset threshold is 0.7. As the relative speed increases, the preset threshold gradually rises, and its upper limit is 1. The situation of the preset threshold changing with the increase in the relative speed can be referred to Figure 5 , where the proportionality coefficient takes a value of 0.1. It should be noted that formula (3) is only an optional form of the preset threshold function. Under the condition that the preset threshold increases with the increase in the relative speed of the channel boat and is subject to the maximum value constraint, other preset threshold functions that meet the conditions can also be constructed.

[0107] S55. Based on the division result of the droplet splash zone, a gas-liquid interaction zone is divided from the initial mesh correction region.

[0108] In step S55, the gas-liquid interaction zone can be divided from the remaining region after removing the droplet splash zone from the initial mesh correction region.

[0109] In some embodiments, the entire remaining region after removing the droplet splash zone from the initial mesh correction region can be directly divided into the gas-liquid interaction zone, that is: the initial mesh correction region is equal to the union of the droplet splash zone and the gas-liquid interaction zone.

[0110] Considering that during the navigation of a channel boat, the area with bubbles may extend from the channels of the channel boat to a rather long area behind the channel boat. In fact, these areas with bubbles behind the channel boat do not affect the navigation of the channel boat. Therefore, in some preferred embodiments, the initial mesh correction area is larger than the union of the droplet splash area and the gas-liquid interaction area. That is, only the area of the channel boat hull and the area in front of it in the remaining area of the initial mesh correction area after removing the droplet splash area needs to be divided into the gas-liquid interaction area. In this case, the remaining area of the initial mesh correction area after removing the droplet splash area and the gas-liquid interaction area is located behind the channel boat. For this part of the area, there is no need to correct its mesh division scheme, and the initial mesh division scheme can be used for mesh division.

[0111] After completing the division of the droplet splash area and the gas-liquid interaction area, in step S6, the droplet splash area and the gas-liquid interaction area can be adaptively remeshed according to the first mesh division scheme and the second mesh division scheme, so as to obtain the corrected mesh division scheme. As analyzed above, the first mesh division scheme sparsifies the mesh of the droplet splash area. By simplifying the simulation of the evolution process of droplets that do not interact with the channel boat, it can reduce the computational amount while effectively alleviating the problem of divergence of the calculation results during the numerical simulation process. For the gas-liquid interaction area, the second mesh division scheme encrypts the mesh to improve the simulation accuracy of the interaction process of bubbles, droplets and the channel boat hull in this area as much as possible, so as to ensure the accuracy of the overall numerical simulation results.

[0112] In some preferred embodiments, after re-dividing the meshes of the droplet splash area and the gas-liquid interaction area respectively to obtain the corrected mesh division scheme, the corrected mesh division scheme can be further optimized. For example, steps of performing mesh density smoothing operations on the droplet splash area and other areas, or on the gas-liquid interaction area and other areas in the corrected mesh division scheme can be added to ensure that the meshes in different areas can transition naturally.

[0113] Step S7:

[0114] After completing step S6, in step S7, the numerical simulation of the time evolution process of the flow field of the channel boat during navigation can be carried out. Obviously, in this step, the corrected mesh division scheme will be used to divide the fluid model in the computational domain, and then basically the same operations as in step S4 will be adopted. At each numerical simulation moment, the gas-liquid interface will be captured by the Level Set algorithm, bubbles and droplets will be tracked, and the flow field distributions of the liquid part and the gas part at this numerical simulation moment will be calculated. According to the numerical simulation results of the navigation flow field, the flow field evolution and gas-liquid interaction situation around the hull of the channel boat during its navigation can be effectively obtained, providing a reliable basis for optimizing the hull structure and further improving the drag reduction effect.

[0115] In some preferred embodiments, the force condition of the channel boat hull during navigation can be further calculated based on the flow field distribution at each simulation moment, and the above-mentioned hull force condition can further provide effective information for optimizing the hull operation.

[0116] In some preferred embodiments, the duration of the numerical simulation in step S7 is greater than the duration covered by each sampling moment in step S4, that is, the mesh division scheme can be corrected based on the sampling results within a short time period, and then the numerical simulation of the long-term navigation flow field can be carried out based on the corrected mesh division scheme.

[0117] In addition, since there are still regions in the gas-liquid interaction region generated in step S55 that simultaneously contain bubbles and droplets (that is, the value is less than or equal to the preset threshold but not ), and regions that only contain bubbles (that is, the value is ), therefore, in some preferred embodiments, the mesh division scheme can be further optimized. Specifically, the gas-liquid interaction region can be further divided into a first gas-liquid interaction region and a second gas-liquid interaction region, where the first gas-liquid interaction region is the region where the value is not and is less than or equal to the preset threshold, and the second gas-liquid interaction region is the region where the value is .

[0118] Different mesh division densities can be further set for the first gas-liquid interaction region and the second gas-liquid interaction region: the mesh density of the first gas-liquid interaction region divided according to the second mesh division scheme is greater than the mesh density of the second gas-liquid interaction region divided according to the second mesh division scheme. Through the above settings, three different regions are actually formed in the second mesh division scheme: for the droplet splash region that does not interact with the hull, a coarser mesh is used for division; for the region where bubbles and droplets coexist and interact with the hull, the densest mesh is used for division; for the region where only bubbles exist, since the problem of numerical divergence does not need to be considered in this region, its mesh density is between the previous two.

[0119] Figures 6 to 9 Shows the results of each step of calculating the navigation flow field of a channel boat using the adaptive simulation method for the navigation flow field of a channel boat provided by the present application in a specific embodiment. This embodiment is a seakeeping test of a channel boat. The speed of the channel boat is 25 knots, and the fluid model contains waves that appear at a period of 5.5 s. The wave height, wavelength, and wave speed are 1.2 m, 47.23 m, and 8.59 m / s, respectively.

[0120] Among them, Figure 6Schematic diagram of meshing the fluid model using the initial meshing scheme Figure 7 Schematic diagram of meshing the fluid model using the modified meshing scheme, comparison Figure 6 and Figure 7 It can be seen that the meshing is thickened in the region of the splashing liquid that does not come into contact with the channel boat in front of the channel boat to reduce the numerical simulation calculation amount and improve the numerical simulation speed. For the region in front of the channel boat and inside the channel where there are complex gas-liquid interactions and splashing liquid droplets in contact with the channel boat, the meshing is refined to ensure the numerical simulation accuracy of the above complex interaction regions. Figure 8 and Figure 9 Side view and top view of the flow field distribution at the 11th second obtained by simulating the evolution process of the flow field of the channel boat using the modified meshing scheme are respectively shown. From the above figures, it can be seen that by using the method provided in the present application, through more reasonable adaptive meshing of the key regions affecting the flow field analysis results during the simulation of the flow field of the channel boat, the balance between the analysis accuracy and speed of the flow field of the channel boat is achieved.

[0121] The specific embodiments of the present application have been described in detail above. For those skilled in the art of the present technology, without departing from the principle of the present application, several improvements and modifications can still be made to the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. An adaptive simulation method for the flow field of a channel boat, characterized in that: The following steps are involved: S1, establishing a fluid model and a channel boat model and setting a calculation domain, wherein the fluid model includes a liquid part and a gas part; S2, setting the initial flow field distribution of the fluid model, the speed of the channel boat and the initial grid division scheme; S3, meshing the fluid model based on the initial meshing scheme; S4, capturing the gas-liquid interface morphology during the navigation of the channel boat based on the initial grid division scheme, and tracking the droplets and bubbles caused by the change of the gas-liquid interface morphology; S5, segmenting the droplet splashing area and the gas-liquid interaction area based on the tracking results of the droplets and bubbles; S6, using the first grid division scheme and the second grid division scheme to perform adaptive grid division on the droplet splash area and the gas-liquid interaction area, respectively, to obtain a revised grid division scheme, wherein the grid density of the droplet splash area divided according to the first grid division scheme is less than the grid density divided according to the initial grid division scheme, and the grid density of the gas-liquid interaction area divided according to the second grid division scheme is greater than the grid density divided according to the initial grid division scheme; S7, numerically simulating the time evolution process of the channel boat navigation flow field based on the modified grid division scheme; Step S5 includes the following steps: S51, dividing an initial grid correction area in the calculation domain around the channel boat according to the tracking results of the bubble and droplet structures, wherein the initial grid correction area is a union of areas where at least one bubble structure and at least one droplet structure appear at a plurality of sampling moments; S52, dividing a plurality of statistical grids in the initial grid correction area according to a preset statistical grid density; S53, based on the tracking results of the droplet structure at a plurality of sampling moments, counting the droplet splashing trend in each statistical grid; S54, dividing a droplet splashing area from the initial grid correction area based on the statistical result of the droplet splashing trend; S55, dividing a gas-liquid interaction area from the initial grid correction area based on the division result of the droplet splash area; For each statistical grid, its droplet splashing trend is calculated based on the following formula: ， in, is the splash separation index, is the total number of sampling moments, For the The total number of droplets that appear in the statistical grid at the sampling moment, For the The sampling time appears in the statistical grid The velocity vector of the droplet, For the The sampling time appears in the statistical grid The unit vector of a droplet relative to the channel boat is directed from the center of mass of the bow of the channel boat to the droplet position. is the maximum value function, Is a null value.

2. The adaptive simulation method of the channel boat navigation flow field according to claim 1 is characterized in that: In the initial meshing scheme, the mesh density in the gas-liquid interface region and the interface region between the fluid and the channel boat is greater than the mesh density in other regions.

3. The adaptive simulation method of the channel boat navigation flow field according to claim 1 is characterized in that: Step S4 is implemented by the Level Set algorithm, which specifically includes the following steps: S41, defining a labeling function, wherein the zero isosurface of the labeling function is a gas-liquid interface; S42, solving the evolution equation of the labeling function and capturing the gas-liquid interface morphology and the real-time flow field distribution of the fluid model at a plurality of sampling moments; S43, by reconstructing the gas-liquid interface and evolving the flow field, bubbles and droplet structures are generated in the liquid part and the gas part respectively based on the interface instability mechanism, and their distribution at several sampling moments is tracked.

4. The adaptive simulation method of the channel boat navigation flow field according to claim 1 is characterized in that: The droplet splashing area is divided from the initial grid correction area based on the statistical results of the droplet splashing trend, specifically: The initial mesh correction area The statistical grids with values ​​greater than the preset threshold are divided into droplet splash areas.

5. The adaptive simulation method of the channel boat navigation flow field according to claim 4 is characterized in that: The preset threshold value increases with the increase of the relative speed of the channel boat, and the maximum value limit that can be reached is 1.

6. The adaptive simulation method of the channel boat navigation flow field according to claim 4 is characterized in that: The gas-liquid interaction area is divided from the initial grid correction area based on the division result of the droplet splash area, specifically: The gas-liquid interaction area is divided into a remaining area after the droplet splashing area is removed from the initial grid correction area.

7. The adaptive simulation method of the channel boat navigation flow field according to claim 6 is characterized in that: The initial grid correction area is larger than the combination of the droplet splash area and the gas-liquid interaction area, and the remaining area of ​​the initial grid correction area after excluding the droplet splash area and the gas-liquid interaction area is located behind the channel boat.

8. The adaptive simulation method for the flow field of a channel boat according to claim 6 is characterized in that: The gas-liquid interaction zone is further divided into a first gas-liquid interaction zone and a second gas-liquid interaction zone; The first gas-liquid interaction zone is Value is not and is less than or equal to the preset threshold, the second gas-liquid interaction area is Value is area; The grid density of the first gas-liquid interaction zone divided according to the second grid division scheme is greater than the grid density of the second gas-liquid interaction zone divided according to the second grid division scheme.

9. The adaptive simulation method of the channel boat navigation flow field according to claim 1 is characterized by: The modified grid division scheme also includes a step of performing a grid density smoothing operation on the droplet splashing area and other areas or on the gas-liquid interaction area and other areas.

Images