Broken ice channel modeling digital twinning method for ship ice resistance calculation

Abstract

The invention discloses a ship ice resistance calculation-oriented broken ice channel modeling digital twinning method, which comprises the following steps of: obtaining broken ice contour information of a broken ice channel through a broken ice image segmentation and coordinate mapping method, and obtaining a circle center coordinate by using an RDP algorithm and a hexagonal grid algorithm; the method comprises the following steps: acquiring a three-dimensional DEM sphere model of a crushed ice channel by introducing a layer thickness mapping and sphere center coordinate generation technology, and acquiring a ship-ice-water multiphase coupling numerical simulation model simulated in software by adopting a CFD (Computational Fluid Dynamics) method through combining an integral computational domain of a ship three-dimensional model with a computational domain grid with an SST turbulence model and a contact model; by establishing a ship motion model and a transient pressure load database, the resistance and contact force distribution condition of the whole ship body can be obtained in real time when the ship sails in an ice region, so that the construction process of crushed ice channel modeling digital twinning for ship ice resistance calculation is realized; the problems that in an existing method, broken ice geometry simplification, ice loading condition idealization and stress distribution are difficult to obtain are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of polar shipping, and particularly relates to a digital twin method for ice channel modeling for ship ice resistance calculation. BACKGROUND

[0002] In recent years, with the continuous melting of Arctic sea ice and the gradual opening of polar shipping routes, polar shipping and resource exploration activities are becoming increasingly frequent. When a ship sails in an ice channel, the interaction between the ship and the ice will generate complex ice loads and additional resistance, which seriously affects the safety, maneuvering stability and fuel economy of the ship. Therefore, developing high-precision digital modeling and ice resistance prediction methods for ice channels has important engineering significance for the design optimization and navigation decision of polar ships.

[0003] Currently, research on ship ice resistance mainly relies on physical model tests, empirical formulas and numerical simulations. Physical model tests simulate real ice conditions by arranging artificial ice in a towing tank. Although the results are reliable, the cost is high, the period is long, and it is difficult to achieve flexible reproduction and mechanism analysis of complex ice conditions; semi-empirical formulas (such as the Lindqvist model) based on test data regression are simple to calculate, but have insufficient generalization ability when ice conditions are variable and ship types are significantly different, making it difficult to meet the high-precision engineering prediction requirements.

[0004] With the development of computational fluid dynamics (CFD) and discrete element method (DEM), the CFD-DEM coupling strategy has become the mainstream numerical method for simulating ship-ice-water multiphase interaction. This method can simulate the discrete motion, breaking behavior and complex contact process between ice and ship. However, existing numerical methods still have some limitations: first, the ice geometry model is mostly generated based on simple regular shapes (such as cubes and spheres), which does not reflect the non-uniform characteristics of ice in real ice-breaking channels in terms of size, shape and spatial distribution; second, the ice field is generated by a "dynamic injection" method, which relies on water flow to drive ice to impact a stationary ship model. This "water-driven ice" loading mechanism is fundamentally different from the physical process in actual ship active ice-breaking navigation, resulting in deviations in key results such as resistance peak and ice accumulation pattern from tests; third, existing methods have weak capabilities in obtaining real ice shapes from image sources and implementing high-fidelity geometric reconstruction, lacking a complete technical link for converting visual observation data into computable physical models, which limits the application of digital twin models in real ship engineering.

[0005] Therefore, it is urgent to develop a digital twin method that integrates real ice image recognition, high-precision geometric reconstruction, dynamic ship-ice coupling simulation and fine load analysis to improve the accuracy and engineering applicability of ship navigation performance prediction in ice channels. SUMMARY

[0006] The application provides a digital twin method for modeling an ice-breaking channel for calculating ice resistance of a ship to overcome the above technical problems.

[0007] To achieve the above-mentioned purpose, the technical scheme of the application is: A digital twin method for modeling an ice-breaking channel for calculating ice resistance of a ship, comprising the following steps: S1: obtaining ice fragment contour information in an ice-breaking channel image; S2: preprocessing the ice fragment contour information to obtain ice fragment physical contour information of the ice fragment in a preset physical coordinate system; and the preprocessing comprises coordinate mapping and scaling; Simplifying the ice fragment boundary according to the ice fragment physical contour information by using an RDP algorithm to obtain a simplified ice fragment contour, and obtaining a DXF vector diagram according to the simplified ice fragment contour; Discretizing the ice fragment polygon, i.e., the simplified ice fragment contour, in the DXF vector diagram based on a hexagonal grid algorithm to obtain a center coordinate of the ice fragment polygon; Based on the actually measured ice layer thickness, the center coordinate of the ice fragment polygon is used to obtain a vertical coordinate of the sphere center of the ice-breaking channel; S3: importing a pre-established ship three-dimensional model of a target ship into software to obtain a whole calculation domain and a calculation domain grid of the ship three-dimensional model; and using a CFD method to numerically simulate a flow field environment of the ship sailing in the ice-breaking channel based on an SST K- turbulence model combined with the whole calculation domain and the calculation domain grid to obtain a simulated flow field; S4: creating spherical particles based on a Lagrangian multiphase model by using a DEM method, and obtaining a DEM spherical model of the ice fragments according to material properties of the ice blocks; The material properties of the ice blocks at least include density, Poisson's ratio, and elastic modulus; S5: setting interaction parameters in the simulated flow field based on a contact model; The interaction parameters at least include a friction coefficient between the ice fragments, a friction coefficient between the ice fragments and the ship body, a normal restitution coefficient and a tangential restitution coefficient for describing mechanical properties of contact, sliding, and rebounding of the ship body and the ice fragments during the sailing of the ship; Based on the interaction parameters, the contact mechanical response between the ice fragment particles and between the ice fragments and the ship body is obtained according to the DEM spherical model, and then a ship-ice contact model is obtained; S6: calling an Injector component in the software to perform a table injection operation based on the preset ice thickness according to the vertical coordinate of the sphere center to obtain a numerical ice-breaking channel model for reproducing a spatial distribution state of the ice fragments. S7: obtaining the ship motion model through calling the DFBI component in the software, obtaining the transient pressure load data corresponding to the ship-ice contact when the simulated ship motion model sails in the numerical ice channel model according to the ship-ice contact model by calling the monitoring function, and obtaining the total resistance time history curve of the ship sailing in the ice area according to the transient pressure load data, thereby realizing the construction process of the ice channel modeling digital twinning for ship ice resistance calculation.

[0008] Further, the expression for preprocessing the ice profile information in S2 is , In the formula: the numerical coordinates of the ice profile information; , the pixel coordinates of the ice image in the ice channel image; the heading angle of the actual ship; , the scaling ratio; , the pixel coordinate translation amount of the ice image.

[0009] Further, the method for obtaining the vertical coordinates of the spherical center of the ice channel according to the center coordinates of the ice polygon in S2 is as follows: S01: obtaining the actual measured ice thickness, and setting a plurality of spherical bodies according to the ice thickness; and the distance between the centers of the adjacent two spherical bodies is the radius of the spherical body; the center of each spherical body is located on a straight line along the vertical direction according to the center coordinates of the ice polygon; S02: recording and obtaining the center coordinates of each spherical body, i.e., obtaining the vertical coordinates of the spherical center of the ice channel.

[0010] Further, S7 specifically includes the following steps: S71: obtaining the ship motion model through calling the DFBI component in the software, S72: obtaining the transient pressure load data corresponding to the ship-ice contact when the simulated ship motion model sails in the numerical ice channel model through the monitoring function Contact Force[i] in the software and i representing the speed direction; S73: outputting the transient pressure load data through the Report component in the software; S74: mapping the transient pressure load data to the preset ship surface grid to obtain the real-time contact force distribution of each grid in the ship surface grid. The contact force time history curves of each grid during ship navigation in ice areas are obtained based on the real-time contact force distribution. S75: Perform time arithmetic averaging on the time history curves of each contact force to obtain the total resistance time history curve for ship navigation in ice zones, thereby realizing the construction process of a digital twin for ice channel modeling for ship ice resistance calculation.

[0011] Furthermore, the method in S2 for discretizing the ice-fragment polygons in the DXF vector image based on the hexagonal mesh algorithm to obtain the center coordinates of the ice-fragment polygons specifically includes the following steps: S21: Obtain the area of ​​each ice fragment polygon, i.e., each simplified ice fragment outline, in the DXF vector image. And select the ice-crush polygon with the smallest area in the corresponding region as the reference polygon; S22: Obtain the length values ​​of each side of the reference polygon, and define the radius of the structural circle corresponding to the hexagonal mesh built based on the hexagonal mesh algorithm according to the length values ​​of each side. ,and > > ;in, This represents the side length corresponding to the shortest side of the reference polygon. This represents the side length corresponding to the longest side of the reference polygon; S23: Divide the ice fragments in the DXF vector image into a grid using a hexagonal grid to obtain the gridded ice fragment region; The coordinates of the hexagonal vertices located within the gridded ice fragmentation area are used as the center coordinates of the ice fragmentation polygon.

[0012] Furthermore, S2 employs the RDP algorithm to simplify the ice fragment boundaries based on the physical contour information of the ice fragments. This method specifically includes the following steps: S100: Obtain a set of directed position coordinates of the physical contour information of the broken ice according to the preset contour direction. And a set of directed position coordinates ; S101: Obtain the set of directed position coordinates initial coordinates in and define the line segment And the initial coordinate point These are the two coordinate points with the largest positional distance. Based on the initial coordinates Set of directed position coordinates Partition the first set and With the second set and ; S102: Obtain the first set The vertical distance of each coordinate point to the straight line segment ; The expression of the vertical distance of each coordinate point to the straight line segment , In the formula: Indicates the vertical distance; S103: Obtain the current maximum distance , and define the corresponding coordinate point as ; Determine whether the current maximum distance satisfies the preset distance threshold ; If , discard the intermediate coordinate points between the initial coordinate points ; If , keep the coordinate points , and obtain the first straight line segment and the second straight line segment according to the initial coordinate points ; S104: According to the coordinate points , divide the first set into a third set and a fourth set , and , ; Obtain the vertical distance of each coordinate point in the third set or the fourth set to the first straight line segment or the second straight line segment , and continue to repeat the method based on step S103; S105: Based on the method of steps S103-S104, optimize the coordinate points of the second set , and further realize the simplified acquisition of the ice boundary and the simplified ice profile.

[0013] Beneficial effects: The application provides a digital twin method for ice channel modeling of ship ice resistance calculation, obtains ice profile information of the ice channel through ice image segmentation and coordinate mapping method, and obtains the center coordinates by using RDP algorithm and hexagonal grid algorithm; By introducing layer thickness mapping and sphere center coordinate generation technology, obtain the three-dimensional DEM sphere model of the ice channel, and adopt CFD method through the combination of the overall calculation domain of the ship three-dimensional model and the calculation domain grid SST turbulence model, contact model, obtain in The software simulates a ship-ice-water multi-phase coupling numerical simulation model; by establishing a ship motion model and a transient pressure load database, the overall resistance and contact force distribution of the ship body during the ship sailing in the ice area can be obtained in real time, so as to realize the construction process of the ice-breaking channel modeling digital twin for ship ice resistance calculation, and greatly improve the accuracy and engineering applicability of the ship sailing performance prediction in the ice-breaking channel. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings described below are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0015] Figure 1 The flow chart of the ice-breaking channel modeling digital twin method for ship ice resistance calculation of the present application; Figure 2 The schematic diagram of the ice-breaking channel in the present embodiment; Figure 3 The ice-breaking image segmentation schematic diagram in the present embodiment; Figure 4 The simplified ice-breaking profile schematic diagram in the present embodiment; Figure 5 The hexagonal grid schematic diagram in the present embodiment; Figure 6 The calculation domain schematic diagram in the present embodiment; Figure 7 The calculation domain grid schematic diagram in the present embodiment; Figure 8 The spherical particle filling ice block schematic diagram in the present embodiment; Figure 9 The particle contact model schematic diagram in the present embodiment; Figure 10 The numerical ice-breaking channel schematic diagram in the present embodiment; Figure 11 The sailing resistance curve at different densities at 0.53m / s in the present embodiment; Figure 12 The maximum contact force distribution of the ship model at 45% ice-breaking density (a, b, c represent the maximum contact force distribution at 0.22, 0.33, 0.53m / s, respectively) in the present embodiment; Figure 13 The ship sailing resistance average value under different ice-breaking densities in the present embodiment; Figure 14Fig. 1 is a schematic diagram of the ship transverse segmentation (left: forward-looking bow view; right: overhead stern view) in the embodiment; Figure 15 Fig. 2 is a schematic diagram of the average resistance distribution of each region in the embodiment; Figure 16 Fig. 3 is a schematic diagram of the resistance change over time of B1-B10 in the embodiment. DETAILED DESCRIPTION

[0016] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0017] The embodiment provides a broken ice channel modeling digital twin method for ship ice resistance calculation, as shown in Figure 1 , comprising the following steps: S1: obtaining broken ice contour information in a broken ice channel image; Specifically, the method for obtaining broken ice contour information in a broken ice channel image in the embodiment is a known technology, including but not limited to obtaining a broken ice channel and a broken ice channel image through a broken ice observation camera, as shown in Figure 2 , and performing image recognition of broken ice contour information as input. First, a deep learning model SAM is used to realize preliminary segmentation, obtain a coarse mask of the broken ice channel domain, and capture small ice blocks missed through residual compensation. Then, the noise and false targets are removed by combining the geometric feature constraint Solidity with the area threshold value, and the effective candidate region is reserved. On this basis, a active contour model MGAC is introduced to take the coarse segmentation result as an initial boundary, and iterative optimization is performed under the action of gradient field and energy function, so that the contour can still approach the real ice edge in the weak boundary and adhesion case, thereby realizing fine segmentation of the broken ice image, as shown in Figure 3 , and finally the extracted broken ice contour information is saved as a CSV file. The specific implementation content will not be described in detail here; S2: In order to convert the broken ice contour information recognized by the image into a numerical coordinate system, the broken ice contour information is preprocessed to obtain the broken ice physical contour information of the broken ice in the preset physical coordinate system; and the preprocessing includes coordinate mapping and scaling. Specifically, the expression for preprocessing the broken ice contour information is , In the formula, represents the numerical coordinates of the broken ice contour information; , pixel coordinate of the ice image in the ice channel image; indicates the bow angle of the real ship, and rotating the angle makes the bow direction in the ice pool numerical experiment or ; indicates the scaling ratio, which is used to convert the pixel scale to the ice pool scale, that is, first, the pixels are one-to-one corresponding to the real ship scale, and second, the real ship scale also needs to be converted to the ice pool ship model scale according to the scaling ratio. In this embodiment, the preferred scale ratio of the real ship and the ship model is 1:21.43; indicates the pixel coordinate translation amount of the ice image, that is, the ice channel is placed at a position 1.5 times the ship length away from the bow of the numerical experiment ice pool; The RDP algorithm is used to simplify the ice boundary according to the ice physical contour information, and the DXF vector diagram is obtained according to the simplified ice contour; After completing the numerical coordinate mapping, the ice boundary is simplified based on the RDP algorithm. The RDP is a classic curve simplification algorithm, which is used to reduce the number of polyline or boundary points while maintaining the overall shape of the curve. This algorithm can significantly reduce the number of vertices while maintaining the accuracy of geometric features by iteratively deleting redundant points; In this embodiment, the method for simplifying the ice boundary according to the ice physical contour information by using the RDP algorithm includes the following steps: S100: According to the preset contour direction, obtain a set of directed position coordinate points of the ice physical contour information , and the set of directed position coordinate points ; S101: Obtain the initial coordinate points in the set of directed position coordinate points , and define a straight line segment , and the initial coordinate points are the two coordinate points with the largest position distance; According to the initial coordinate points , divide the set of directed position coordinate points into a first set and a second set and ; Then, the first set and the second set are recursively processed, specifically: S102: Obtain the perpendicular distance of each coordinate point in the first set from the straight line segment ; ​​And the expression for the perpendicular distance from each coordinate point to the line segment is: , In the formula: Indicates vertical distance; S103: Get the current maximum distance And define the corresponding coordinate point as ; Determine the current maximum distance Does it meet the preset distance threshold? ; like This indicates that all coordinate points are near the line segment and the entire curve corresponding to that line segment can be used. If this is the case, then the initial coordinates are discarded. The midpoint between them; like This indicates that the entire curve segment corresponding to the straight line deviates too much from the straight line, and this point must be retained. That is, retain the coordinate points And based on the initial coordinates Obtain the first straight line segment With the second straight segment ; S104: Based on coordinate points The first set Divided into the third set With the fourth set ,and , ; And obtain the third set Or the fourth set The distance of each coordinate point from the first straight line segment Or the second straight segment The vertical distance is calculated, and the process is repeated based on the method in step S103. S105: Based on the methods of steps S103 to S104, process the second set. The coordinate points are optimized to simplify the ice fragment boundaries and obtain a simplified ice fragment outline.

[0018] In this embodiment, the ice fragment outline information includes a boundary of 10787 points. Under the threshold, it can be simplified to 1909 points, such as Figure 4 The simplified ice crush outline shown is used to effectively reduce the calculation time. The simplified outline can form a script for DXF vector graphics and then obtain a DXF file. The method of obtaining DXF vector graphics based on the simplified ice crush outline in this embodiment is a well-known existing technology and will not be described in detail here. Discretize the ice fragment polygons in the DXF vector diagram, i.e. simplify the ice fragment contour, based on a hexagonal grid algorithm, obtain the center coordinates of the ice fragment polygons, specifically including the following steps: S21: Obtain the area of each ice fragment polygon in the DXF vector diagram, i.e. the area of each simplified ice fragment contour; And select the ice fragment polygon with the smallest area as the reference polygon; S22: Obtain the length of each side of the reference polygon, and define the radius of the structure circle corresponding to the hexagonal grid algorithm based on the length of each side , and ; wherein, represents the length of the shortest side of the reference polygon; represents the length of the longest side of the reference polygon; The principle of the hexagonal grid algorithm is: As shown in Figure 5 , generate a circle with a radius of inside any polygon area, the distance between adjacent circles is , and form a hexagonal grid by staggering the odd and even rows (assuming is the overall horizontal offset of each row and ), forming a closely arranged hexagon for discretization modeling; in this embodiment, the circle radius , so the distance between adjacent circles is , about 11320 circle center points can be generated, thereby forming a closely arranged hexagon. All circle center coordinates falling within the polygon boundary are retained and output as a CSV file, providing ice discretization data for CFD-DEM coupling simulation; S23: Divide the grid of each ice fragment polygon in the DXF vector diagram by the hexagonal grid, obtain the grid ice area, and take the coordinates of the hexagonal vertex located in the grid ice area as the center coordinates of the ice fragment polygon; Based on the actual measured ice thickness, obtain the spherical center vertical coordinates of the ice channel according to the center coordinates of the ice fragment polygon, specifically: S01: Obtain the actual measured ice thickness, and set a plurality of spherical bodies according to the ice thickness; The distance between the centers of the adjacent two spherical bodies is the radius of the spherical body; the centers of the spherical bodies are located on a straight line in the vertical direction according to the center coordinates of the ice fragment polygon; S02: Record and obtain the center coordinates of each spherical body, i.e. obtain the spherical center vertical coordinates of the ice channel.

[0019] ​​This embodiment establishes the number of coordinate layers based on the actual measured ice thickness, using the coordinates of the center of the ice fragments in the two-dimensional plane. In this embodiment, the measured ice thickness is 0.86m, and according to the model scaling ratio of 1:21.43, the ice fragment thickness is 40mm. Therefore, the ice fragment channel area can be divided into 2 layers. In each layer, a copy of the center plane coordinates is made and assigned a corresponding height z value. Finally, 22,640 center coordinates are obtained and output to a CSV file. Each data point contains (x,y,z) three-dimensional coordinates, which can completely describe the distribution of the ice fragment centers in physical space. At the same time, the CSV file can be directly input into the CFD-DEM coupled simulation software for modeling ice fragments in the ice fragment channel.

[0020] S3: Import the pre-built 3D model of the target ship. The software acquires the overall computational domain and computational domain mesh of the ship's 3D model; and uses the CFD method based on SST K- The turbulence model combines the overall computational domain with the computational domain grid to numerically simulate the flow field environment of ships navigating in ice-breaking channels and obtain the simulated flow field. Specifically, the following steps are included: S31: Import the pre-built 3D ship model In the software, based on the ship's dimensions, speed, and the characteristics of the ice-breaking channel, a comprehensive computational domain is first established around the ship. This domain is ensured to be large enough to cover both the ship's navigation area and the ice-breaking movement range. Simultaneously, inlet, outlet, and boundary conditions are set according to simulation requirements. In this implementation, the forward bow is set as the velocity inlet, the stern and top as pressure outlets, and the sides and bottom as walls. The computational domain is as follows: Figure 6 As shown; the overall computational domain is 33m*3.9m*10m in length, width and height; then the overall fluid computational domain is established: the hull volume is subtracted using Boolean operations to obtain the fluid computational domain, and finally an overlapping region is established according to the hull size to ensure that the ship is completely wrapped in the overlapping region and the overlapping region is contained in the computational domain, thus obtaining the overall computational domain. S32: Based on the obtained overall computational domain, mesh the computational domain and overlapping domain: Use a surface reconstruction generator to generate high-quality surface meshes for the hull and ice contact surfaces; use a cut-volume mesh generator to generate volume meshes for the fluid domain to ensure the analytical accuracy of the flow field near the ship; apply a prism layer mesh to the fluid domain near the hull and ice contact surfaces to improve the accuracy of wall shear force calculations; and call the generated mesh. The Mesh component in the Mesh module performs mesh quality checks to ensure that mesh cells are free of negative volumes, non-manifold cells, or extreme distortions, thus guaranteeing the stability of numerical computation and obtaining the computational domain mesh. In this example, the meshes of the computational domain and the overlapping domain are as follows: Figure 7 As shown; S33: Adopting CFD method, based on SST K- turbulence model, through blending function and implementation turbulence model to simulate the flow field environment of ship sailing in broken ice channel, where blending function and According to the wall distance, local turbulent Reynolds number and flow field gradient information, the model weight is adaptively adjusted to realize the smooth transition of the two, so as to balance the wall flow precision and the stability of free flow calculation.

[0021] Specifically, in the boundary layer region near the wall, the turbulence model can accurately predict the wall turbulent shear stress and fluid near-wall behavior, while in the free flow region away from the wall, the turbulence model can effectively describe the turbulent diffusion and energy dissipation characteristics.

[0022] and SST K- The core equation of turbulence model is: turbulent kinetic energy equation : , specific dissipation rate equation : , turbulent kinetic energy dissipation rate equation : , turbulent viscosity coefficient : , turbulent production term : , blending function : , , In turbulence model, the definitions of various physical quantities and symbols are as follows: denotes time; denotes spatial coordinate component and ; denotes fluid density; denotes fluid velocity component; denotes flow energy; denotes specific dissipation rate; denotes empirical constant; express Prandt number; Indicates the molecular viscosity coefficient; Indicates the turbulent viscosity coefficient; Represents the empirical coefficients of the turbulence model; Represents the constants of the turbulence model; Indicates the distance from the wall; Indicates dynamic viscosity; This represents the correction term used to ensure numerical stability in the near-wall region; This represents the turbulent kinetic energy generation term; In this embodiment, the fluid component employs the incompressible fluid control equations and uses a multiphase flow model based on the Volume of Fluid (VOF) method to handle the gas-liquid interface. In this experimental example, the fluid boundary conditions are set as follows: inlet velocity is 0 to simulate a still water environment; pressure outlet boundary conditions are used at the outlet and top; wall boundary conditions are used at the bottom and sides; and a no-slip wall is set on the hull surface to construct a flow field model for ship navigation in ice zones, providing a basic fluid environment, i.e., a simulated flow field, for subsequent ship-ice interactions. And the fluid control equations are Continuity equation: , in: Indicates density; Indicates fluid velocity; Momentum equation: , in: Indicates pressure; Indicates dynamic viscosity; Represents gravitational acceleration; S4: Use the DEM method to create spherical particles based on the Lagrange multiphase model, and obtain the DEM spherical model that constitutes the ice block according to the material properties of the ice block; Furthermore, the material properties of the ice block include at least density, Poisson's ratio, and elastic modulus; Specifically, in the numerical modeling of the broken ice in this embodiment, the ice material properties consistent with those of the experimental ice pool are first selected, including density, Poisson's ratio, and elastic modulus; the density of the broken ice is approximately determined through a pre-set experiment. Poisson's ratio and elastic modulus ; and input the above material parameters to The DEM material property library was used; subsequently, based on the Lagrange multiphase model and the DEM method, a broken ice model was generated according to the spherical particle method, resulting in a discrete spherical particle DEM spherical model that can reflect the geometric morphology and mechanical properties of broken ice.Figure 8 As shown.

[0023] S5: Based on Contact model, setting the interaction parameters in the simulated flow field; Furthermore, the interaction parameters include at least the friction coefficient between ice fragments, the friction coefficient between ice fragments and the hull, and the normal and tangential recovery coefficients used to describe the mechanical properties of the hull contacting, slipping, and rebounding with ice fragments during ship navigation. Based on the interaction parameters, the contact mechanical response between ice fragments and between ice fragments and the ship hull is obtained according to the DEM spherical model, thereby obtaining the ship-ice fragment contact model; Specifically, in this embodiment based on The contact model sets interaction parameters between ice fragments and between ice fragments and the ship hull to obtain the contact mechanical response between ice fragments and between ice fragments and the ship hull. In this embodiment, the friction coefficient between ice fragments is set to 0.3, and the normal and tangential recovery coefficients are both set to 0.2 to reflect the energy dissipation characteristics during the collision process between ice fragments. The friction coefficient between ice fragments and the ship hull surface is set to 0.4, and the normal and tangential recovery coefficients are both set to 0.5 to describe the mechanical characteristics of the ship hull contacting, slipping, and rebounding with ice fragments during ship navigation. At the same time, based on a pre-set simple failure model, the spheres that make up the ice fragments are connected by adhesive bonds to obtain the failure behavior between ice fragments. This experimental example does not consider the secondary breakage of ice fragments, so the bonding strength is set to infinity, and finally a ship-ice fragment contact model is established. And the The principle of the contact model is as follows: like Figure 9 The interparticle contact forces shown include normal and tangential forces, expressed as follows: , In the formula: normal force With tangential force Based on the amount of contact overlap between particles and material parameters, this model can more realistically reflect the nonlinear stiffness characteristics and energy dissipation behavior of ice fragment collisions than traditional linear models.

[0024] S6: Call The Injector component in the software performs a table injection operation based on the preset ice thickness and the vertical coordinates of the sphere's center to obtain a numerical ice channel model for reproducing the spatial distribution of ice. Specifically, in this embodiment, the vertical coordinate data of the center of the ice sphere obtained in step S3 is first organized into a CSV table format that can be recognized by the software; then, the coordinate file is imported. The software, simultaneously invoking the table spraying function, uses the table spraying operation to accurately generate corresponding DEM spherical elements within the computational domain defined in S41, based on the preset ice fragment thickness and the provided vertical coordinates of the sphere center. This allows for the reproduction of the spatial distribution of ice fragments in the numerical environment, thereby obtaining a numerical ice fragment channel model, such as... Figure 10 As shown; S7: By calling The DFBI component in the software acquires the ship motion model. By calling monitoring functions, it obtains transient pressure load data corresponding to ship-ice contact when the simulated ship motion model navigates in a numerical ice-breaking channel model. Based on the transient pressure load data, it obtains the total resistance time history curve for ship navigation in ice zones, thus realizing the construction process of a digital twin for ice-breaking channel modeling oriented towards ship ice resistance calculation. Specifically, it includes the following steps: S71: By calling The DFBI component in the software acquires the ship's motion model; Specifically, by calling The components yield a six-degree-of-freedom kinematic model of the ship, whose expression is: , , in: Indicates the mass of the rigid body (kg); Represents the rigid body velocity vector (m / s); The resultant force (N) exerted by the fluid on the surface of the rigid body includes pressure and shear force; This represents other external forces (N), such as gravity, buoyancy, and impact forces. Rigid body inertia tensor (kg·m) 2 ), describing the inertial properties of a rigid body's rotational motion; This represents the rigid body angular velocity vector (rad / s). The torque (N·m) represents the force exerted by a fluid on a rigid body. It represents other external torques (N·m), such as buoyancy torque, collision torque, etc. Represents the surface of a rigid body; Indicates fluid pressure (Pa); Represents the surface normal vector; Represents the fluid shear stress tensor (Pa); This represents the vector from the center of mass to the surface element; In this experimental example, to simplify the calculation process and highlight the force characteristics under the interaction between ice floes and the ship hull, the ship hull is defined as a rigid body and is only allowed to maintain a constant speed in the bow direction. The other five degrees of freedom (sway, heave, roll, pitch, and bow roll) are all constrained and fixed. Through this setting, the ship motion model is obtained. S72: Through The monitoring function Contact Force[i] in the software, where i represents the velocity direction, acquires the transient pressure load data corresponding to the ship-ice contact when the simulated ship motion model navigates in the numerical ice-breaking channel model; and the transient pressure load is the force in S5 of the mesh information under ship-ice interaction extracted in real time during the transient calculation process. ; S73: Through The Report component in the software outputs transient pressure load data; S74: Map transient pressure load data to a preset hull surface grid to obtain the real-time contact force distribution of each grid in the hull surface grid; The contact force time history curves of each grid during ship navigation in ice areas are obtained based on the real-time contact force distribution. S75: Perform time arithmetic averaging on the time history curves of each contact force to obtain the total resistance time history curve for ship navigation in ice conditions. The total resistance time history curve for ship navigation in ice conditions is shown below. Figure 11 As shown, this demonstrates the process of constructing a digital twin for modeling ice-breaking channels for calculating ship ice resistance. In this experimental example, the maximum contact force at the bow is as follows: Figure 12 As shown, the average resistance values ​​for ships navigating in ice-covered areas at different ice fragment concentrations and speeds are as follows: Figure 13 As shown; This embodiment also includes calling... The Threshold component in the software's Derived Parts module obtains the pressure distribution of each part of the hull after dividing it into ten equal parts along the width of the ship. Call In the software, the Threshold component of the Derived Parts module filters and divides the existing hull surface mesh based on a specific location, creating a new mesh component containing only that specific location interval. This achieves secondary processing of the hull surface mesh. In this embodiment, using the hull's central longitudinal section as a symmetry reference, the hull is divided into ten equal-width segments along the width direction. Each segment is treated as an independent component, thus achieving equal partitioning of the hull surface in the transverse direction, ensuring a refined analysis of the hull's transverse force distribution. In this experimental example, the ship partitioning is as follows: Figure 14 As shown; Simultaneously, after obtaining ten components, the same method as S82 and S83 was used to extract the resistance change of each section under the transient interaction between ice fragments and the hull, thus obtaining the resistance time history of the ten sections. By performing arithmetic averaging on the transient load curves of each section, the average resistance of each section of the hull was obtained, which can intuitively reflect the force differences in different lateral regions. In this experimental example, the average resistance of each section of the hull and the time history curves of the resistance of each section are as follows: Figures 15 to 16 As shown.

[0025] The method described in this embodiment uses ice-observation cameras to acquire channel images and combines deep learning and active contour models to extract ice fragment contours. The contours are mapped to a numerical coordinate system and simplified using the RDP algorithm. Hexagonal mesh filling is used to obtain the center coordinates, and a 3D DEM ice fragment model is generated by overlaying ice layer thickness. The ship's 3D model is then imported. In the software, a computational domain is constructed and CFD-DEM coupled simulation is performed, based on... The contact model is used to obtain the contact mechanical response of broken ice; then, the ship motion model is obtained by constructing a six-degree-of-freedom motion model of the ship, and the ship's sailing resistance and pressure distribution along the ship's width are extracted to realize high-precision digital twin calculation of ship ice resistance in broken ice channels. This solves the problems of broken ice geometry simplification, idealization of ice loading conditions and difficulty in obtaining force distribution in existing methods, and greatly improves the accuracy and engineering applicability of ship navigation performance prediction in broken ice channels.

[0026] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A digital twin method for modeling broken ice channels for calculating ship ice resistance, characterized in that, Specifically, the following steps are included: S1: Obtain ice debris outline information from the ice debris channel image; S2: Preprocess the ice fragment outline information to obtain the ice fragment physical outline information in a preset physical coordinate system; and the preprocessing includes coordinate mapping and scaling. The RDP algorithm is used to simplify the ice fragment boundary based on the physical contour information of the ice fragment, obtain the simplified ice fragment contour, and then obtain the DXF vector image based on the simplified ice fragment contour. Based on the hexagonal mesh algorithm, the ice-breaking polygons in the DXF vector image, i.e. the simplified ice-breaking outline, are discretized to obtain the center coordinates of the ice-breaking polygons. Based on the actual measured ice thickness, the vertical coordinates of the ice break channel's center are obtained from the coordinates of the center of the ice break polygon. S3: Import the pre-built 3D model of the target ship. The software acquires the overall computational domain and computational domain mesh of the ship's 3D model; and uses the CFD method based on SST K- The turbulence model combines the overall computational domain with the computational domain grid to numerically simulate the flow field environment of ships navigating in ice-breaking channels and obtain the simulated flow field. S4: Use the DEM method to create spherical particles based on the Lagrange multiphase model, and obtain the DEM spherical model that constitutes the ice crush based on the material properties of the ice. Furthermore, the material properties of the ice block include at least density, Poisson's ratio, and elastic modulus; S5: Based on Contact model, setting the interaction parameters in the simulated flow field; Furthermore, the interaction parameters include at least the friction coefficient between ice fragments, the friction coefficient between ice fragments and the hull, and the normal and tangential recovery coefficients used to describe the mechanical properties of the hull contacting, slipping, and rebounding with ice fragments during ship navigation. Based on the interaction parameters, the contact mechanical response between ice fragments and between ice fragments and the ship hull is obtained according to the DEM spherical model, thereby obtaining the ship-ice fragment contact model; S6: Call The Injector component in the software performs a table injection operation based on the preset ice thickness and the vertical coordinates of the sphere's center to obtain a numerical ice channel model for reproducing the spatial distribution of ice. S7: By calling The DFBI component in the software acquires the ship motion model. By calling the monitoring function, it obtains the transient pressure load data corresponding to the ship-ice contact when the simulated ship motion model is navigating in the numerical ice channel model. Based on the transient pressure load data, it obtains the total resistance time history curve of the ship navigating in the ice zone, thereby realizing the construction process of the digital twin for ice channel modeling oriented towards ship ice resistance calculation.

2. The digital twin method for modeling broken ice channels for ship ice resistance calculation as described in claim 1, characterized in that, The expression for preprocessing the ice fragment contour information described in S2 is as follows: In the formula: Numerical coordinates representing the outline information of ice fragments; , Represents the pixel coordinates of the ice debris image in the ice channel image; Indicates the bow angle of a real ship; , Indicates the scaling ratio; , This represents the pixel coordinate translation amount of the ice fragment image.

3. The digital twin method for modeling ice-breaking channels for ship ice resistance calculation as described in claim 2, characterized in that, The method in S2 for obtaining the vertical coordinates of the center of the ice break channel from the center coordinates of the ice break polygon is as follows: S01: Obtain the actual measured ice thickness and set several spheres according to the ice thickness; Furthermore, the distance between the centers of two adjacent spheres is equal to the radius of the sphere; the center of each sphere lies on a straight line along the vertical direction, based on the coordinates of the center of the ice polygon. S02: Record and obtain the coordinates of the center of each sphere, that is, obtain the vertical coordinates of the center of the sphere in the ice break channel.

4. The digital twin method for modeling broken ice channels for ship ice resistance calculation according to claim 3, characterized in that, S7 specifically includes the following steps: S71: By calling The DFBI component in the software acquires the ship's motion model. S72: Through The monitoring function Contact Force[i] in the software, where i represents the velocity direction, obtains the transient pressure load data corresponding to the ship-ice contact when the simulated ship motion model is navigating in the numerical ice-breaking channel model. S73: Through The Report component in the software outputs the transient pressure load data; S74: Map transient pressure load data to a preset hull surface grid to obtain the real-time contact force distribution of each grid in the hull surface grid; The contact force time history curves of each grid during ship navigation in ice areas are obtained based on the real-time contact force distribution. S75: Perform time arithmetic averaging on the time history curves of each contact force to obtain the total resistance time history curve for ship navigation in ice zones, thereby realizing the construction process of a digital twin for ice channel modeling for ship ice resistance calculation.

5. A digital twin method for modeling ice-breaking channels for ship ice resistance calculation according to claim 3, characterized in that, The method in S2 for discretizing ice-fragment polygons in a DXF vector image based on a hexagonal mesh algorithm and obtaining the center coordinates of the ice-fragment polygons includes the following steps: S21: Obtain the area of ​​each ice fragment polygon, i.e., each simplified ice fragment outline, in the DXF vector image. And select the ice-crush polygon with the smallest area in the corresponding region as the reference polygon; S22: Obtain the length values ​​of each side of the reference polygon, and define the radius of the structural circle corresponding to the hexagonal mesh built based on the hexagonal mesh algorithm according to the length values ​​of each side. ,and > > ;in, This represents the side length corresponding to the shortest side of the reference polygon. This represents the side length corresponding to the longest side of the reference polygon; S23: Divide the ice fragments in the DXF vector image into a grid using a hexagonal grid to obtain the gridded ice fragment region; The coordinates of the hexagonal vertices located within the gridded ice fragmentation area are used as the center coordinates of the ice fragmentation polygon.

6. The digital twin method for modeling ice-breaking channels for ship ice resistance calculation as described in claim 1, characterized in that, The method used in S2 to simplify the ice fragment boundary based on the physical contour information of the ice fragments includes the following steps: S100: Obtain a set of directed position coordinates of the physical contour information of the broken ice according to the preset contour direction. And a set of directed position coordinates ; S101: Obtain the set of directed position coordinates initial coordinates in and define the line segment And the initial coordinate point These are the two coordinate points with the largest positional distance. Based on the initial coordinates Set of directed position coordinates Partition the first set and With the second set and ; S102: Obtain the first set The distance between each coordinate point and the straight line segment The vertical distance; And the expression for the perpendicular distance from each coordinate point to the line segment is: In the formula: Indicates vertical distance; S103: Get the current maximum distance And define the corresponding coordinate point as ; Determine the current maximum distance Does it meet the preset distance threshold? ; like Then discard the initial coordinate point. The midpoint between them; like Then retain the coordinates. And based on the initial coordinates Obtain the first straight line segment With the second straight segment ; S104: Based on coordinate points The first set Divided into the third set With the fourth set ,and , ; And obtain the third set Or the fourth set The distance of each coordinate point from the first straight line segment Or the second straight segment The vertical distance is calculated, and the process is repeated based on the method in step S103. S105: Based on the methods of steps S103 to S104, process the second set. The coordinate points are optimized to simplify the ice fragment boundaries and obtain a simplified ice fragment outline.

Images