A high-speed slender body ice-breaking method and system based on ALE

Abstract

This invention relates to the field of icebreaking for aircraft, specifically to a method and system for high-speed, slender bodies to break ice upon exiting the water, based on ALE (Alternating Lens Array). The method includes: forming a lattice Boltzmann flow field model based on a multi-medium physical parameter set through discretization; embedding and fusing the ice damage propagation mechanism into the lattice Boltzmann flow field model; completing the ALE assignment mapping; dynamically reconstructing the ALE mesh; setting the initial launch parameters of the high-speed, slender body; and performing numerical simulation of icebreaking for the high-speed, slender body based on ALE. This invention uses the combination of a lattice Boltzmann flow field model and structured ALE as its core, embedding the ice damage propagation mechanism and interface capture algorithm to achieve collaborative simulation of flow field evolution, ice layer fracture, and cross-medium interface migration.

Description

Technical Field

[0001] This invention relates to the field of icebreaking for navigable bodies, specifically to a high-speed, slender body icebreaking method and system based on ALE (Air-to-Earth) technology. Background Technology

[0002] Due to their unique location and resource value, polar regions such as the Arctic have become core areas for competition over maritime rights. The sea ice that covers the polar regions year-round provides a natural cover for operational vessels, but it also poses severe challenges to their launches. The traditional method of launching from the surface and breaking ice places extremely high demands on the structural strength and maneuverability of the vessel, and it is easy to expose the target. Relying directly on the kinetic energy of the launcher to break the ice faces technical challenges such as uneven ice thickness and complex impact loads, which can easily lead to structural damage or launch failure.

[0003] In projectile research, the high-speed, slender underwater launch and ice-penetrating technique has attracted much attention due to its advantages such as high concealment and rapid response. However, this process involves complex multidisciplinary issues such as water-ice-gas three-phase coupling, transient impact loads, and dynamic ice fracture. On the one hand, the motion of the high-speed, slender body induces nonlinear fluid phenomena such as cavitation and water cushion effect, which directly affect the impact kinetic energy transfer efficiency. On the other hand, ice has heterogeneous mechanical properties and is resistant to compression but not tension. Its crack initiation and propagation are significantly affected by parameters such as impact velocity and ice thickness. Traditional numerical simulation methods are difficult to accurately capture the two-way coupling relationship between flow field evolution and ice damage.

[0004] In existing technologies, the simple Lagrange method is prone to decreased computational accuracy due to mesh distortion, the Euler method is difficult to accurately track the structure-medium interface, and conventional fluid-structure interaction models do not fully consider the synergistic effect of ice damage propagation mechanism and cross-medium interface evolution. As a result, the numerical simulation of the ice breaking process of slender bodies has problems such as insufficient accuracy and incomplete physical mechanism characterization, which cannot provide reliable theoretical support for actual engineering design. Summary of the Invention

[0005] This invention addresses the technical problems existing in the prior art by providing a high-speed, slender body water effluent ice-breaking method and system based on ALE.

[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A high-speed slender body water effluent ice breaking method based on ALE, including the following steps: Step 1: After deploying the acquisition sensors for ice, water, air and high-speed slender body, the acquired acquisition data is preprocessed to form a standardized multi-media physical parameter set; Step 2: Based on the obtained multi-medium physical parameter set, a lattice Boltzmann flow field model is formed through discretization. Then, the ice physical parameters contained in the multi-medium physical parameter set obtained in Step 1 are used to embed the ice damage propagation mechanism into the lattice Boltzmann flow field model. An interface capture algorithm is introduced to fuse and correct the lattice Boltzmann flow field model. The lattice Boltzmann flow field model is combined with the structured ALE to complete the assignment mapping of ALE. Step 3: Assign the obtained multi-medium physical parameter set to ALE, and after dynamic reconstruction of the ALE mesh, set the initial emission parameters of the high-speed slender body, and use the lattice Boltzmann flow field model to perform multi-physics field evolution of the high-speed slender body's water-breaking process, obtain the multi-physics field evolution dataset of the entire process of the high-speed slender body's water-breaking process, and complete the numerical simulation of high-speed slender body's water-breaking based on ALE.

[0007] In a preferred embodiment, step two, based on the obtained multi-medium physical parameter set, forms a lattice Boltzmann flow field model through discretization processing, specifically as follows: The Boltzmann equation is selected as the governing equation to represent the microscopic motion state of water and air particles. The Cahn-Hilliard equation is introduced to describe the phase distribution of the water-air two phases. The water phase and the gas phase are distinguished by the order parameter. The parameters of the multi-medium physical parameters that include the water phase and the gas phase are assigned values ​​to form a preliminary lattice Boltzmann flow field model. The preliminary lattice Boltzmann flow field model is discretized using the D2Q9 discretization scheme. Taking the flow field nodes in the preliminary lattice Boltzmann flow field model as the center, a stationary direction without velocity components is set, and eight peripheral motion directions are set according to the four corresponding horizontal left and right, vertical up and down, and oblique directions, respectively, to obtain nine discrete directions of the two-dimensional flow field. The weighting coefficients for the nine discrete directions of the two-dimensional flow field are distributed by taking one-ninth of the weighting coefficient for the central stationary direction, one-ninth of the weighting coefficients for each of the four horizontal and vertical directions, and one-thirty-sixth of the weighting coefficients for each of the four oblique directions. The continuous Boltzmann equation and the Cahn-Hilliard equation are simultaneously transformed into particle transport equations in discrete velocity directions, obtaining particle distribution functions including fluid momentum distribution functions and phase sequence parameter distribution functions in each discrete direction, thus forming a lattice Boltzmann flow field model.

[0008] In a preferred embodiment, step two, after forming the lattice Boltzmann flow field model, further includes setting the flow field boundary conditions, specifically: The ice body and the surface of the slender body are set as solid wall boundaries. The rule that particles bounce back in the opposite direction after hitting the solid wall is used as the bounce boundary condition, thereby ensuring that the momentum exchange between the fluid and the solid conforms to the physical laws. The outside of the flow field is set as a non-reflective boundary. The boundary node values ​​are linearly extrapolated based on the physical quantities of the adjacent nodes inside the non-reflective boundary. The water-air initial phase interface is set as a horizontal interface. In some specific implementations, setting it as a horizontal interface means that it is consistent with the liquid level of the actual water tank. In the actual simulation, the interface is adjusted automatically according to the target experiment. The phase interface position is initialized based on the order parameter distribution. After setting the flow field boundary conditions for the lattice Boltzmann flow field model, values ​​are assigned to the parameters in the multi-medium physical parameter set (including water and gas phases) from step one. Initial particle distribution function values ​​are assigned to the nine discrete directions of each node in the flow field. It should be noted that the momentum distribution function is assigned based on hydrostatic pressure and static flow field characteristics, and the order parameter distribution function is assigned 1 and 0 for the water and gas phase regions respectively. The lattice Boltzmann flow field model is then iterated, specifically as follows: The actual values ​​of the fluid momentum distribution function and phase sequence parameter distribution function of the flow field node in each discrete direction in the current iteration are taken as the actual distribution function, which is used to directly reflect the actual motion state and phase properties of the current particle. Then, the actual values ​​of the fluid momentum distribution function and the phase sequence parameter distribution function of the current node under the current physical conditions are fitted to derive the particle motion law of the ideal fluid and obtain the local equilibrium distribution function. Based on the current actual distribution function, the velocity of the flow field node is obtained by weighted summation of the current fluid momentum distribution function in all discrete directions and the velocity component in the corresponding direction, and the pressure of the flow field node is obtained by summing the current fluid momentum distribution functions. Based on the BGK collision model, the difference between the current actual distribution function and the current local equilibrium distribution function is calculated. The difference is added to the actual distribution function of the current iteration, thereby updating the actual distribution function and the local equilibrium distribution function of each discrete direction in the next iteration, completing the exchange of particle momentum and phase properties. The collision-updated actual distribution function is synchronously transmitted to the adjacent flow field nodes according to the corresponding discrete direction, completing the simulation of the spatial motion of particles in the flow field and the evolution of the phase interface. After continuous iterations, when the changes in velocity and pressure at all nodes in the flow field are less than the difference between the actual distribution function and the local equilibrium distribution function, the iteration is considered to have converged. The lattice Boltzmann flow field model finally outputs a real-time dataset containing velocity and pressure, as well as a dataset of the interface position for each node in the flow field.

[0009] In a preferred embodiment, step two is based on the ice body physical parameters included in the multi-medium physical parameter set, discretizing the ice body into near-field dynamic particles, taking each ice body particle as the center, and determining the neighborhood radius based on the density and size of the ice body particles, establishing connections between all ice body particles within the neighborhood radius through bonding, and taking the average value of the compressive strength and yield stress of the ice body, and performing a linear decay correction on the average value according to the distance between particles to obtain the bonding strength between every two ice body particles; In the lattice Boltzmann flow field model, the pressure value of each flow field node is mapped to the corresponding ice particle according to its spatial location, and the product of the pressure of the flow field node corresponding to each ice particle and the area of ​​the ice particle is used as the external load to complete the load transfer between the flow field and the ice particle. When the external load on an ice particle exceeds the bonding strength, the corresponding ice particle is determined to have experienced bond breakage. Based on the location and number of bond breaks, the cumulative bond breakage is taken as the degree of damage to the ice. If the proportion of broken bonds in the neighborhood of an ice particle exceeds 50%, the corresponding ice particle is deemed invalid. The bonding fracture and failure information of ice particles is fed back to the lattice Boltzmann mesoscopic flow field model. The lattice Boltzmann mesoscopic flow field model corrects the flow field boundaries of the corresponding crack region and failure particle region to free boundaries, re-executes iterative calculations, completes the embedding of the ice damage propagation mechanism, updates the flow field velocity and pressure distribution in the region, and realizes the bidirectional coupling between the flow field and ice damage.

[0010] In a preferred embodiment, step two discretizes the cross-medium interfaces of water-ice, ice-gas, and water-emitter in the lattice Boltzmann flow field model into smooth particles, and determines the mass, initial position, and physical properties of the interface particles based on the multi-medium physical parameter set in step one. In some specific embodiments, the density of water interface particles is taken as the density of water, and the density of ice interface particles is taken as the density of ice. With each interface particle as the center, neighboring particles form a neighborhood, and a cubic spline kernel function is constructed based on the spatial position of the interface particles in the neighborhood. The density value of the corresponding interface particle is obtained based on the cubic spline function. The pressure value of the particles in the neighborhood is spatially differentiated by the cubic spline kernel function, and the pressure change rate in each spatial direction is calculated. Then, the change rates in each direction are integrated into the overall pressure gradient. Based on the obtained density and pressure gradient of the interface particles, the ALE mesh interface is corrected. According to the real-time position of each interface particle, the moving least squares method is used to adjust the node position of the ALE mesh, so that the ALE mesh node fits the distribution of interface particles, thereby avoiding mesh distortion and medium leakage problems. At the same time, physical quantities such as the density and pressure gradient of interface particles are mapped to the ALE mesh node to achieve accurate matching between interface features and ALE mesh, and complete the fusion correction of the interface capture algorithm. The pressure, velocity, phase interface data, ice damage data, and interface data of optical interface particles from the lattice Boltzmann mesoscopic flow field model are mapped to the nodes and elements of the ALE mesh using linear interpolation, thus completing the ALE assignment mapping.

[0011] In a preferred embodiment, step two, when constructing the cubic spline kernel function based on the spatial positions of interface particles within the neighborhood, further includes: Extract the spatial coordinates of all interface particles within the neighborhood of the corresponding interface particle. After linearly normalizing the coordinate values, divide the neighborhood space into several continuous segmented intervals in ascending order. The number of intervals matches the number of particles in the neighborhood, ensuring that each interval contains exactly one particle coordinate point. Set an independent cubic polynomial expression for each segmented interval. Note that the polynomial includes quartic coefficients (i.e., cubic, quadratic, linear, and constant terms), which serve as unknowns in the fitting. Establish function value continuity constraints, first derivative continuity constraints, and second derivative continuity constraints for the connection points of adjacent segmented intervals. The function value continuity constraint means that the calculated values ​​of the connection points in the left and right piecewise polynomials are equal. The first derivative continuity constraint means that the first derivative values ​​of the connection point are equal in the two piecewise polynomials on the left and right sides. The second derivative continuity constraint means that the second derivative values ​​of the connection point are equal in the two piecewise polynomials on the left and right sides. At the same time, natural boundary constraints are added to the first and last piecewise intervals of the neighborhood, that is, the second derivative values ​​of the first and last intervals are zero, to ensure that the kernel function is smooth at the boundary. The constraints of function value continuity constraint, first derivative continuity constraint and second derivative continuity constraint are transformed into a system of linear equations, and the unknowns of the linear equations are the coefficients of all piecewise polynomials. The constant terms of the linear equations are determined according to the multi-medium physical parameter set. The linear equations are solved by elimination method to obtain the coefficient values ​​of all piecewise cubic polynomials in the linear equations. The piecewise cubic polynomials with solved coefficients are integrated according to the piecewise intervals to form a complete cubic spline kernel function.

[0012] In some other specific implementations, the verification of the obtained cubic spline kernel function is also included, specifically: Randomly select several coordinate points in the neighborhood and verify whether the function value, first derivative, and second derivative of the kernel function are continuous at that point. If there are discontinuous points, readjust the segmented intervals and repeat the above fitting steps until all continuity requirements are met. In step two, after constructing the cubic spline function, the spatial coordinates of other interface particles in the neighborhood of the corresponding interface particle are substituted into the cubic spline kernel function to obtain the influence weight of each interface particle on the neighborhood of the corresponding interface particle. Then, through the cubic spline kernel function, the mass of all interface particles in the neighborhood is multiplied by the corresponding influence weight and summed to obtain the density value of the corresponding interface particle.

[0013] In a preferred embodiment, step three, when a corresponding ice particle is determined to be faulty, involves dynamic reconstruction of the ALE mesh, including: The momentum and mass conservation governing equations in Lagrange form are spatially discretized using the finite element method. The continuous solution domain is divided into several quadratic grid elements, and the governing equations are transformed into discrete algebraic equations for each quadratic grid element. The stiffness matrix of each quadratic grid element is calculated based on the physical parameters assigned to the ALE according to the multi-medium physical parameter set. At the same time, external loads, including flow field loads and ice damage loads, are transformed into load vectors of the quadratic grid nodes. Apply corresponding displacement, pressure, and velocity boundary conditions to the solid wall boundary, free boundary, and interface boundary of the secondary mesh. Correct the stiffness matrix and load vector, eliminate the unknowns of the boundary nodes, and take the stiffness matrix of all corrected secondary mesh elements as the global stiffness matrix and the load vector as the global load vector to form a complete system of linear algebraic equations. Solve the system by elimination method to obtain the velocity, pressure, and displacement of the secondary mesh nodes. Based on the solution results, update the pressure, velocity, and displacement of all ALE mesh nodes to complete the Lagrange step calculation.

[0014] In a preferred embodiment, step three, in the dynamic reconstruction of the ALE mesh, involves topological reconstruction of the existing distorted mesh, deletion of distorted mesh cells, and adjustment of the mesh cell size and node position in the ALE based on the re-mesh division. Specifically, this includes: Select a new grid node as the center, calculate the Euclidean distance from the new grid node to the surrounding original grid nodes, and remove the original grid nodes whose Euclidean distance exceeds the critical distance value determined by the statistical average value of the grid neighborhood range. For the original grid nodes that have been filtered, calculate the reciprocal of the Euclidean distance between the filtered original grid nodes and the new grid node. The closer the distance, the larger the reciprocal, which indicates that the original grid node has a higher degree of influence on the new grid node. Sum the reciprocals of the distances of all original grid nodes, and then divide the reciprocal of the distance of each original grid node by the sum to obtain the normalized weight corresponding to each original grid node. Multiply the physical quantity of each original grid node by the corresponding normalized weight, and then sum all the products to obtain the physical quantity of the new grid node.

[0015] In a preferred embodiment, after setting the initial launch parameters of the high-speed slender body, including ice contact velocity, launch depth, and aspect ratio, step three involves capturing the generation of flow field cavitation using a lattice Boltzmann flow field model based on a fixed time step, extracting flow field pressure distribution data based on the current state of the ALE grid to draw a spatial distribution map of ice prestress, and extracting damage cloud map data and crack propagation trajectory data based on the distribution of new grid nodes in the ALE, thereby correcting the initial launch parameters of the high-speed slender body and completing the ideal state of water exit and ice breaking of the high-speed slender body.

[0016] This invention also provides a high-speed, slender body ice-breaking system based on ALE, the system comprising: Acquisition module; used to acquire physical parameters of ice, water, air, and high-speed slender bodies via a sensor array; The flow field model construction module is used to construct a lattice Boltzmann flow field model based on a multi-medium physical parameter set, complete the model discretization, boundary condition setting and iterative convergence calculation, and output the real-time dataset of flow field nodes and the dataset of phase interface positions. The ice damage coupling module is used to discretize ice into near-field dynamic particles and establish particle bonding connections to realize the transfer of flow field load to ice particles, determine the bonding breakage and failure state of ice particles, and complete the embedding of the ice damage propagation mechanism into the lattice Boltzmann flow field model and the bidirectional coupling between the flow field and ice damage. The interface fusion correction module is used to discretize the cross-medium interface into smooth particles, construct a cubic spline kernel function to calculate the physical quantities of the interface particles, correct the ALE mesh interface through the interface capture algorithm, achieve accurate matching between interface features and ALE mesh, and complete the assignment mapping of relevant data of lattice Boltzmann flow field model to ALE. The ALE mesh processing module is used to assign multi-medium physical parameter sets to ALE, complete the Lagrange step calculation of ALE mesh, perform topology reconstruction and physical quantity mapping on distorted mesh, and verify the quality of the reconstructed mesh from the dimensions of mesh distortion rate and physical quantity mapping error. The numerical simulation module is used to set the initial launch parameters of the high-speed slender body, perform multiphysics field evolution calculations of the water-breaking process using the lattice Boltzmann flow field model, obtain the multiphysics field evolution dataset of the entire process and visualize it, and correct the initial launch parameters based on the simulation results to complete the numerical simulation of high-speed slender body water-breaking based on ALE.

[0017] The beneficial effects of this invention are as follows: By combining the lattice Boltzmann flow field model with the structured ALE as the core, and embedding the ice damage propagation mechanism and interface capture algorithm, the collaborative simulation of flow field evolution, ice layer fracture, and cross-medium interface migration is realized. Compared with traditional methods, it can characterize the influence of water cushion effect and cavitation on ice layer prestress, as well as the dynamic feedback of ice crack propagation and flow field pressure distribution, making the physical mechanism characterization of multi-medium coupling process more complete, and significantly improving the fit between numerical simulation results and actual working conditions. By discretizing ice into particles and establishing bonded connections, the bonding strength and damage state of the particles are dynamically calculated based on multi-medium physical parameters, accurately reconstructing the complete damage process of the ice layer from bond breakage to particle failure. At the same time, a cross-medium interface particle model is constructed using cubic spline kernel functions, and the ALE mesh is corrected by moving least squares method, which effectively avoids mesh distortion and medium leakage problems, and achieves accurate matching between the water-ice, ice-gas, and water-emitter interface features and the ALE mesh. Attached Figure Description

[0018] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] As attached Figure 1 As shown, this embodiment provides a high-speed, slender body ice-breaking method based on ALE, comprising the following steps: Step 1: After deploying sensors for collecting data on ice, water, air, and high-speed slender bodies, the collected data is preprocessed to form a standardized set of multi-media physical parameters. Furthermore, for ice, a low-temperature servo loading testing machine, a laser porosimeter, and a high-speed topography scanner were used to collect mechanical and topographic parameters such as compressive strength, flexural strength, elastic modulus, yield stress, porosity, and crystal structure. For water, pressure sensors, temperature sensors, and a cavitation characteristic tester were used to collect fluid characteristic parameters such as water pressure, water temperature, water compressibility, and critical cavitation pressure at different depths in polar water. For air, a pressure sensor and a density tester were used to collect gas characteristic parameters such as atmospheric pressure and air density above the ice surface. For high-speed slender bodies, a materials mechanics tester and a laser dimension measuring instrument were used to collect material and geometric parameters such as elastic modulus, density, yield strength, diameter, length, and head curvature.

[0021] After acquiring the collected data, the moving average method is used to denoise all the collected data, remove outliers and isolated values, and the linear normalization method is used to map parameters of different dimensions to a unified numerical range. In some other specific implementations, in view of the heterogeneous characteristics of ice, the homogeneity level of ice is divided according to porosity and crystal structure data, and a corresponding heterogeneity correction coefficient is assigned to different levels. All parameters of pre-processed ice, water, air, and high-speed slender bodies are classified and organized according to medium type, parameter category, and operating conditions to form a standardized multi-medium physical parameter set.

[0022] Step 2: Based on the obtained multi-medium physical parameter set, a lattice Boltzmann flow field model is formed through discretization. Then, the ice physical parameters contained in the multi-medium physical parameter set in Step 1 are used to embed the ice damage propagation mechanism into the lattice Boltzmann flow field model. Finally, an interface capture algorithm is introduced to fuse and correct the lattice Boltzmann flow field model. The lattice Boltzmann flow field model is combined with the structured ALE to complete the assignment mapping of ALE. Step two involves discretizing the obtained multi-medium physical parameter set to form a lattice Boltzmann flow field model, specifically as follows: The Boltzmann equation is selected as the governing equation to represent the microscopic motion state of water and air particles. The Cahn-Hilliard equation is introduced to describe the phase distribution of the water-air two phases. The water phase and the gas phase are distinguished by the order parameter. The parameters of the multi-medium physical parameters that include the water phase and the gas phase are assigned values ​​to form a preliminary lattice Boltzmann flow field model. The preliminary lattice Boltzmann flow field model is discretized using the D2Q9 discretization scheme. Taking the flow field nodes in the preliminary lattice Boltzmann flow field model as the center, a stationary direction without velocity components is set, and eight peripheral motion directions are set according to the four corresponding horizontal left and right, vertical up and down, and oblique directions, respectively, to obtain nine discrete directions of the two-dimensional flow field. The weighting coefficients for the nine discrete directions of the two-dimensional flow field are distributed by taking one-ninth of the weighting coefficient for the central stationary direction, one-ninth of the weighting coefficients for each of the four horizontal and vertical directions, and one-thirty-sixth of the weighting coefficients for each of the four oblique directions. The continuous Boltzmann equation and the Cahn-Hilliard equation are simultaneously transformed into particle transport equations in discrete velocity directions, obtaining particle distribution functions including fluid momentum distribution functions and phase sequence parameter distribution functions in each discrete direction, thus forming a lattice Boltzmann flow field model.

[0023] Step two, after forming the lattice Boltzmann flow field model, also includes setting the flow field boundary conditions, specifically: The ice body and the surface of the slender body are set as solid wall boundaries. The rule that particles bounce back in the opposite direction after hitting the solid wall is used as the bounce boundary condition, thereby ensuring that the momentum exchange between the fluid and the solid conforms to the physical laws. The outside of the flow field is set as a non-reflective boundary. The boundary node values ​​are linearly extrapolated based on the physical quantities of the adjacent nodes inside the non-reflective boundary. The water-air initial phase interface is set as a horizontal interface. In some specific implementations, setting it as a horizontal interface means that it is consistent with the liquid level of the actual water tank. In the actual simulation, the interface is adjusted automatically according to the target experiment. The phase interface position is initialized based on the order parameter distribution. After setting the flow field boundary conditions for the lattice Boltzmann flow field model, values ​​are assigned to the parameters in the multi-medium physical parameter set (including water and gas phases) from step one. Initial particle distribution function values ​​are assigned to the nine discrete directions of each node in the flow field. It should be noted that the momentum distribution function is assigned based on hydrostatic pressure and static flow field characteristics, and the order parameter distribution function is assigned 1 and 0 for the water and gas phase regions respectively. The lattice Boltzmann flow field model is then iterated, specifically as follows: The actual values ​​of the fluid momentum distribution function and phase sequence parameter distribution function of the flow field node in each discrete direction in the current iteration are taken as the actual distribution function, which is used to directly reflect the actual motion state and phase properties of the current particle. Then, the actual values ​​of the fluid momentum distribution function and the phase sequence parameter distribution function of the current node under the current physical conditions are fitted to derive the particle motion law of the ideal fluid and obtain the local equilibrium distribution function. Based on the current actual distribution function, the velocity of the flow field node is obtained by weighted summation of the current fluid momentum distribution function in all discrete directions and the velocity component in the corresponding direction, and the pressure of the flow field node is obtained by summing the current fluid momentum distribution functions. Based on the BGK collision model, the difference between the current actual distribution function and the current local equilibrium distribution function is calculated. The difference is added to the actual distribution function of the current iteration, thereby updating the actual distribution function and the local equilibrium distribution function of each discrete direction in the next iteration, completing the exchange of particle momentum and phase properties. The collision-updated actual distribution function is synchronously transmitted to the adjacent flow field nodes according to the corresponding discrete direction, completing the simulation of the spatial motion of particles in the flow field and the evolution of the phase interface. After continuous iterations, when the changes in velocity and pressure at all nodes in the flow field are less than the difference between the actual distribution function and the local equilibrium distribution function, the iteration is considered to have converged. The lattice Boltzmann flow field model finally outputs a real-time dataset containing velocity and pressure, as well as a dataset of the interface position for each node in the flow field.

[0024] Step 2: Based on the ice body physical parameters contained in the multi-medium physical parameter set, the ice body is discretized into near-field dynamic particles. With each ice body particle as the center, the neighborhood radius is determined based on the density and size of the ice body particles. All ice body particles within the neighborhood radius are connected through bonding. The compressive strength and yield stress of the ice body are averaged, and the average value is linearly attenuated according to the distance between particles to obtain the bonding strength between every two ice body particles. In the lattice Boltzmann flow field model, the pressure value of each flow field node is mapped to the corresponding ice particle according to its spatial location, and the product of the pressure of the flow field node corresponding to each ice particle and the area of ​​the ice particle is used as the external load to complete the load transfer between the flow field and the ice particle. When the external load on an ice particle exceeds the bonding strength, the corresponding ice particle is determined to have experienced bond breakage. Based on the location and number of bond breaks, the cumulative bond breakage is taken as the degree of damage to the ice. If the proportion of broken bonds in the neighborhood of an ice particle exceeds 50%, the corresponding ice particle is deemed invalid. The bonding fracture and failure information of ice particles is fed back to the lattice Boltzmann mesoscopic flow field model. The lattice Boltzmann mesoscopic flow field model corrects the flow field boundaries of the corresponding crack region and failure particle region to free boundaries, re-executes iterative calculations, completes the embedding of the ice damage propagation mechanism, updates the flow field velocity and pressure distribution in the region, and realizes the bidirectional coupling between the flow field and ice damage.

[0025] Step 2 discretizes the cross-medium interfaces of water-ice, ice-gas, and water-emitter in the lattice Boltzmann flow field model into smooth particles, and determines the mass, initial position, and physical properties of the interface particles based on the multi-medium physical parameter set in Step 1. In some specific implementations, the density of water interface particles is taken as the density of water, and the density of ice interface particles is taken as the density of ice. With each interface particle as the center, neighboring particles form a neighborhood, and a cubic spline kernel function is constructed based on the spatial position of the interface particles in the neighborhood. The density value of the corresponding interface particle is obtained based on the cubic spline function. The pressure value of the particles in the neighborhood is spatially differentiated by the cubic spline kernel function, and the pressure change rate in each spatial direction is calculated. Then, the change rates in each direction are integrated into the overall pressure gradient. Based on the obtained density and pressure gradient of the interface particles, the ALE mesh interface is corrected. According to the real-time position of each interface particle, the moving least squares method is used to adjust the node position of the ALE mesh, so that the ALE mesh node fits the distribution of interface particles, thereby avoiding mesh distortion and medium leakage problems. At the same time, physical quantities such as the density and pressure gradient of interface particles are mapped to the ALE mesh node to achieve accurate matching between interface features and ALE mesh, and complete the fusion correction of the interface capture algorithm. The pressure, velocity, phase interface data, ice damage data, and interface data of optical interface particles from the lattice Boltzmann mesoscopic flow field model are mapped to the nodes and elements of the ALE mesh using linear interpolation, thus completing the ALE assignment mapping.

[0026] Step 3: Assign the obtained multi-medium physical parameter set to ALE, and after dynamic reconstruction of the ALE mesh, set the initial emission parameters of the high-speed slender body, and use the lattice Boltzmann flow field model to perform multi-physics field evolution of the high-speed slender body's water-breaking process, obtain the multi-physics field evolution dataset of the entire process of the high-speed slender body's water-breaking process, and complete the numerical simulation of high-speed slender body's water-breaking based on ALE.

[0027] Step 3 involves dynamically reconstructing the ALE mesh when a corresponding ice particle fails to be identified, including: The momentum and mass conservation governing equations in Lagrange form are spatially discretized using the finite element method. The continuous solution domain is divided into several quadratic grid elements, and the governing equations are transformed into discrete algebraic equations for each quadratic grid element. The stiffness matrix of each quadratic grid element is calculated based on the physical parameters assigned to the ALE according to the multi-medium physical parameter set. At the same time, external loads, including flow field loads and ice damage loads, are transformed into load vectors of the quadratic grid nodes. Apply corresponding displacement, pressure, and velocity boundary conditions to the solid wall boundary, free boundary, and interface boundary of the secondary mesh. Correct the stiffness matrix and load vector, eliminate the unknowns of the boundary nodes, and take the stiffness matrix of all corrected secondary mesh elements as the global stiffness matrix and the load vector as the global load vector to form a complete system of linear algebraic equations. Solve the system by elimination method to obtain the velocity, pressure, and displacement of the secondary mesh nodes. Based on the solution results, update the pressure, velocity, and displacement of all ALE mesh nodes to complete the Lagrange step calculation.

[0028] Step 3 involves dynamic reconstruction of the ALE mesh, performing topology reconstruction on existing distorted meshes, deleting distorted mesh cells, and then adjusting the mesh cell size and node positions in the ALE mesh based on the re-mesh. Specifically: Select a new grid node as the center, calculate the Euclidean distance from the new grid node to the surrounding original grid nodes, and remove the original grid nodes whose Euclidean distance exceeds the critical distance value determined by the statistical average value of the grid neighborhood range. For the original grid nodes that have been filtered, calculate the reciprocal of the Euclidean distance between the filtered original grid nodes and the new grid node. The closer the distance, the larger the reciprocal, which indicates that the original grid node has a higher degree of influence on the new grid node. Sum the reciprocals of the distances of all original grid nodes, and then divide the reciprocal of the distance of each original grid node by the sum to obtain the normalized weight corresponding to each original grid node. Multiply the physical quantity of each original grid node by the corresponding normalized weight, and then sum all the products to obtain the physical quantity of the new grid node.

[0029] This application further proposes to quantify the degree of distortion by the side length ratio and interior angle deviation of the grid cells. When the side length ratio and interior angle deviation of all grid cells are less than the critical distortion value determined by the variance value based on the statistics of the geometric characteristics of the grid cells, the distortion rate is deemed to meet the requirements. Select several key verification nodes and calculate the relative error between the physical quantities of the new grid nodes and the physical quantities of the original grid nodes. When the relative error of all key nodes is less than the critical error value determined based on the root mean square value of the errors of all verification nodes, the mapping error is deemed qualified. Finally, the quality of the reconstructed mesh was verified from two dimensions: mesh distortion rate and physical quantity mapping error.

[0030] Step 3: After setting the initial launch parameters of the high-speed slender body, including ice contact velocity, launch depth, and aspect ratio, the generation of flow field cavitation is captured by the lattice Boltzmann flow field model based on a fixed time step. The flow field pressure distribution data is extracted according to the current state of the ALE grid to draw a spatial distribution map of ice prestress. After extracting damage cloud map data and crack propagation trajectory data according to the distribution of new grid nodes in ALE, the initial launch parameters of the high-speed slender body are corrected to complete the ideal state of water breaking of the high-speed slender body.

[0031] This invention also provides a high-speed, slender body ice-breaking system based on ALE, the system comprising: Acquisition module; used to acquire physical parameters of ice, water, air, and high-speed slender bodies via a sensor array; The flow field model construction module is used to construct a lattice Boltzmann flow field model based on a multi-medium physical parameter set, complete the model discretization, boundary condition setting and iterative convergence calculation, and output the real-time dataset of flow field nodes and the dataset of phase interface positions. The ice damage coupling module is used to discretize ice into near-field dynamic particles and establish particle bonding connections to realize the transfer of flow field load to ice particles, determine the bonding breakage and failure state of ice particles, and complete the embedding of the ice damage propagation mechanism into the lattice Boltzmann flow field model and the bidirectional coupling between the flow field and ice damage. The interface fusion correction module is used to discretize the cross-medium interface into smooth particles, construct a cubic spline kernel function to calculate the physical quantities of the interface particles, correct the ALE mesh interface through the interface capture algorithm, achieve accurate matching between interface features and ALE mesh, and complete the assignment mapping of relevant data of lattice Boltzmann flow field model to ALE. The ALE mesh processing module is used to assign multi-medium physical parameter sets to ALE, complete the Lagrange step calculation of ALE mesh, perform topology reconstruction and physical quantity mapping on distorted mesh, and verify the quality of the reconstructed mesh from the dimensions of mesh distortion rate and physical quantity mapping error. The numerical simulation module is used to set the initial launch parameters of the high-speed slender body, perform multiphysics field evolution calculations of the water-breaking process using the lattice Boltzmann flow field model, obtain the multiphysics field evolution dataset of the entire process and visualize it, and correct the initial launch parameters based on the simulation results to complete the numerical simulation of high-speed slender body water-breaking based on ALE.

Claims

1. An ALE-based high-speed slender body out-of-water icebreaking method, characterized in that, The method includes: Step 1: After deploying sensors for collecting data on ice, water, air, and high-speed slender bodies, the collected data is preprocessed to form a standardized set of multi-media physical parameters. Step 2: Based on the obtained multi-medium physical parameter set, a lattice Boltzmann flow field model is formed through discretization. Then, the ice physical parameters contained in the multi-medium physical parameter set obtained in Step 1 are used to embed the ice damage propagation mechanism into the lattice Boltzmann flow field model. An interface capture algorithm is introduced to fuse and correct the lattice Boltzmann flow field model. The lattice Boltzmann flow field model is combined with the structured ALE to complete the assignment mapping of ALE. Step 3: Assign the obtained multi-medium physical parameter set to ALE, and after dynamic reconstruction of the ALE mesh, set the initial emission parameters of the high-speed slender body, and use the lattice Boltzmann flow field model to perform multi-physics field evolution of the high-speed slender body's water-breaking process, obtain the multi-physics field evolution dataset of the entire process of the high-speed slender body's water-breaking process, and complete the numerical simulation of high-speed slender body's water-breaking based on ALE.

2. The ALE-based high-speed slender body ice-breaking method of claim 1, wherein, Step two, based on the obtained multi-medium physical parameter set, forms a lattice Boltzmann flow field model through discretization. Specifically, it involves: The Boltzmann equation is selected as the governing equation to represent the microscopic motion state of water and air particles. The Cahn-Hilliard equation is introduced to describe the phase distribution of the water-air two phases. The water phase and the gas phase are distinguished by the order parameter. The parameters of the multi-medium physical parameters that include the water phase and the gas phase are assigned values ​​to form a preliminary lattice Boltzmann flow field model. The preliminary lattice Boltzmann flow field model is discretized using the D2Q9 discretization scheme. Taking the flow field nodes in the preliminary lattice Boltzmann flow field model as the center, a stationary direction without velocity components is set, and eight peripheral motion directions are set according to the four corresponding horizontal left and right, vertical up and down, and oblique directions, respectively, to obtain nine discrete directions of the two-dimensional flow field. The weighting coefficients for the nine discrete directions of the two-dimensional flow field are distributed by taking one-ninth of the weighting coefficient for the central stationary direction, one-ninth of the weighting coefficients for each of the four horizontal and vertical directions, and one-thirty-sixth of the weighting coefficients for each of the four oblique directions. The continuous Boltzmann equation and the Cahn-Hilliard equation are simultaneously transformed into particle transport equations in discrete velocity directions, obtaining particle distribution functions including fluid momentum distribution functions and phase sequence parameter distribution functions in each discrete direction, thus forming a lattice Boltzmann flow field model.

3. The ALE-based high-speed slender body ice-breaking method according to claim 2, wherein, Step two, after forming the lattice Boltzmann flow field model, also includes setting the flow field boundary conditions, specifically: The ice body and the surface of the slender body are set as solid wall boundaries. The rule that the particles bounce back in the opposite direction after hitting the solid wall is used as the bounce boundary condition. The outside of the flow field is set as a non-reflective boundary. The boundary node values ​​are linearly extrapolated based on the physical quantities of the adjacent nodes inside the non-reflective boundary. The water-air initial phase interface is set as a horizontal interface. The phase interface position is initialized based on the order parameter distribution. After setting the flow field boundary conditions for the lattice Boltzmann flow field model, values ​​are assigned based on the parameters of the water and gas phases included in the multi-medium physical parameter set from step one. Initial particle distribution function values ​​are assigned to the nine discrete directions of each node in the flow field, and the lattice Boltzmann flow field model is iterated, specifically as follows: The actual values ​​of the fluid momentum distribution function and phase sequence parameter distribution function of the flow field nodes in each discrete direction in the current iteration are taken as the actual distribution function from the previous iteration. Then, the actual values ​​of the fluid momentum distribution function and the phase sequence parameter distribution function of the current node under the current physical conditions are fitted to obtain the local equilibrium distribution function; Based on the current actual distribution function, the velocity of the flow field node is obtained by weighted summation of the current fluid momentum distribution function in all discrete directions and the velocity component in the corresponding direction, and the pressure of the flow field node is obtained by summing the current fluid momentum distribution functions. Based on the BGK collision model, the difference between the current actual distribution function and the current local equilibrium distribution function is calculated. The difference is added to the actual distribution function of the current iteration, thereby updating the actual distribution function and the local equilibrium distribution function of each discrete direction in the next iteration. The collision-updated actual distribution function is synchronously transmitted to the adjacent flow field nodes according to the corresponding discrete direction, thus completing the simulation of the spatial motion of particles in the flow field and the evolution of the phase interface. After continuous iterations, when the changes in velocity and pressure at all nodes in the flow field are less than the difference between the actual distribution function and the local equilibrium distribution function, the iteration is considered to have converged. The lattice Boltzmann flow field model finally outputs a real-time dataset containing velocity and pressure, as well as a dataset of the interface position for each node in the flow field.

4. The ALE-based high-speed slender body ice-breaking method of claim 1, wherein, Step two is based on the ice body physical parameters contained in the multi-medium physical parameter set. The ice body is discretized into near-field dynamic particles. With each ice body particle as the center, the neighborhood radius is determined based on the density and size of the ice body particles. All ice body particles within the neighborhood radius are connected through bonding. The compressive strength and yield stress of the ice body are averaged, and the average value is linearly attenuated according to the distance between the particles to obtain the bonding strength between every two ice body particles. In the lattice Boltzmann flow field model, the pressure value of each flow field node is mapped to the corresponding ice particle according to its spatial location, and the product of the pressure of the flow field node corresponding to each ice particle and the area of ​​the ice particle is used as the external load to complete the load transfer between the flow field and the ice particle. When the external load on an ice particle exceeds the bonding strength, the corresponding ice particle is determined to have experienced bond breakage. Based on the location and number of bond breaks, the cumulative bond breakage is taken as the degree of damage to the ice. If the proportion of broken bonds in the neighborhood of an ice particle exceeds 50%, the corresponding ice particle is deemed invalid. The bonding fracture and failure information of ice particles is fed back to the lattice Boltzmann mesoscopic flow field model. The lattice Boltzmann mesoscopic flow field model corrects the flow field boundaries of the corresponding crack region and failure particle region to free boundaries, and re-executes iterative calculations to complete the embedding of the ice damage propagation mechanism.

5. The ALE-based high-speed slender body ice-breaking method of claim 1, wherein, Step two discretizes the cross-medium interfaces of water-ice, ice-gas, and water-emitter in the lattice Boltzmann flow field model into smooth particles, and determines the mass, initial position, and physical properties of the interface particles based on the multi-medium physical parameter set in step one. With each interface particle as the center, neighboring particles form a neighborhood, and a cubic spline kernel function is constructed based on the spatial position of the interface particles in the neighborhood. The density value of the corresponding interface particle is obtained based on the cubic spline function. The pressure value of the particles in the neighborhood is spatially differentiated by the cubic spline kernel function, and the pressure change rate in each spatial direction is calculated. Then, the change rates in each direction are integrated into the overall pressure gradient. Based on the obtained density and pressure gradient of the interface particles, the mesh interface of ALE is corrected. According to the real-time position of each interface particle, the node position of ALE mesh is adjusted by moving least squares method, so that ALE mesh nodes are aligned with the distribution of interface particles, thus completing the fusion correction of the interface capture algorithm. The pressure, velocity, phase interface data, ice damage data, and interface data of optical interface particles from the lattice Boltzmann mesoscopic flow field model are mapped to the nodes and elements of the ALE mesh using linear interpolation, thus completing the ALE assignment mapping.

6. The ALE-based high-speed slender body ice-breaking method according to claim 5, wherein, Step two, in constructing the cubic spline kernel function based on the spatial positions of interface particles within the neighborhood, also includes: Extract the spatial coordinates of all interface particles in the neighborhood of the corresponding interface particle. After linearly normalizing the coordinate values, divide the neighborhood space into several continuous segmented intervals in ascending order. Set an independent cubic polynomial expression for each segmented interval. Establish function value continuity constraints, first derivative continuity constraints, and second derivative continuity constraints at the connection points of adjacent segmented intervals. Transform the constraints of function value continuity constraints, first derivative continuity constraints, and second derivative continuity constraints into a system of linear equations. Determine the constant terms of the linear equation system based on the multi-medium physical parameter set. Solve the linear equation system by elimination method to obtain the coefficient values ​​of all piecewise cubic polynomials in the linear equation system. The piecewise cubic polynomials with the solved coefficients are integrated according to the piecewise intervals to form a complete cubic spline kernel function; In step two, after constructing the cubic spline function, the spatial coordinates of other interface particles in the neighborhood of the corresponding interface particle are substituted into the cubic spline kernel function to obtain the influence weight of each interface particle on the neighborhood of the corresponding interface particle. Then, through the cubic spline kernel function, the mass of all interface particles in the neighborhood is multiplied by the corresponding influence weight and summed to obtain the density value of the corresponding interface particle.

7. The method for high-speed, slender body ice breaking based on ALE according to claim 1, characterized in that, Step three involves dynamic reconstruction of the ALE mesh when a corresponding ice particle is deemed to be ineffective, including: The momentum and mass conservation governing equations in Lagrange form are spatially discretized using the finite element method. The continuous solution domain is divided into several quadratic grid elements, and the governing equations are transformed into discrete algebraic equations for each quadratic grid element. The stiffness matrix of each quadratic grid element is calculated based on the physical parameters assigned to the ALE according to the multi-medium physical parameter set. At the same time, external loads, including flow field loads and ice damage loads, are transformed into load vectors of the quadratic grid nodes. Apply corresponding displacement, pressure, and velocity boundary conditions to the solid wall boundary, free boundary, and interface boundary of the secondary mesh, correct the stiffness matrix and load vector, take the stiffness matrix of all corrected secondary mesh elements as the global stiffness matrix and the load vector as the global load vector to form a complete system of linear algebraic equations, and solve them by elimination method to obtain the velocity, pressure, and displacement of the secondary mesh nodes. Based on the solution results, update the pressure, velocity, and displacement of all ALE mesh nodes to complete the Lagrange step calculation.

8. The method for high-speed, slender body ice breaking based on ALE according to claim 7, characterized in that, Step three, in the dynamic reconstruction of the ALE mesh, involves topological reconstruction of the existing distorted mesh, deletion of distorted mesh cells, and then, based on the re-mesh division, adjustment of the mesh cell size and node position in the ALE. Specifically: Select a new grid node as the center, calculate the Euclidean distance from the new grid node to the surrounding original grid nodes, and remove the original grid nodes whose Euclidean distance exceeds the critical distance value determined by the statistical average value based on the grid neighborhood range. For the original grid nodes that have been filtered, calculate the reciprocal of the Euclidean distance between the retained original grid nodes and the new grid node. Sum the reciprocals of the distances of all original grid nodes, and then divide the reciprocal of the distance of each original grid node by the sum to obtain the normalized weight corresponding to each original grid node. Multiply the physical quantity of each original grid node by the corresponding normalized weight, and then sum all the products to obtain the physical quantity of the new grid node.

9. The method for high-speed, slender body ice breaking based on ALE according to claim 1, characterized in that, In step three, after setting the initial launch parameters of the high-speed slender body, including ice contact velocity, launch depth, and aspect ratio, the generation of flow field cavitation is captured by the lattice Boltzmann flow field model based on a fixed time step. The flow field pressure distribution data is extracted according to the current state of the ALE grid to draw a spatial distribution map of ice prestress. After extracting damage cloud map data and crack propagation trajectory data according to the distribution of new grid nodes in ALE, the initial launch parameters of the high-speed slender body are corrected, and the ideal state of water exit and ice breaking of the high-speed slender body is completed.

10. A high-speed, slender body ice-breaking system based on ALE, applied to the high-speed, slender body ice-breaking method based on ALE as described in any one of claims 1-9, characterized in that, The system includes: Acquisition module; used to acquire physical parameters of ice, water, air, and high-speed slender bodies via a sensor array; The flow field model construction module is used to construct a lattice Boltzmann flow field model based on a multi-medium physical parameter set, complete the model discretization, boundary condition setting and iterative convergence calculation, and output the real-time dataset of flow field nodes and the dataset of phase interface positions. The ice damage coupling module is used to discretize ice into near-field dynamic particles and establish particle bonding connections to realize the transfer of flow field load to ice particles, determine the bonding breakage and failure state of ice particles, and complete the embedding of the ice damage propagation mechanism into the lattice Boltzmann flow field model and the bidirectional coupling between the flow field and ice damage. The interface fusion correction module is used to discretize the cross-medium interface into smooth particles, construct a cubic spline kernel function to calculate the physical quantities of the interface particles, correct the ALE mesh interface through the interface capture algorithm, achieve accurate matching between interface features and ALE mesh, and complete the assignment mapping of relevant data of lattice Boltzmann flow field model to ALE. The ALE mesh processing module is used to assign multi-medium physical parameter sets to ALE, complete the Lagrange step calculation of ALE mesh, perform topology reconstruction and physical quantity mapping on distorted mesh, and verify the quality of the reconstructed mesh from the dimensions of mesh distortion rate and physical quantity mapping error. The numerical simulation module is used to set the initial launch parameters of the high-speed slender body, perform multiphysics field evolution calculations of the water-breaking process using the lattice Boltzmann flow field model, obtain the multiphysics field evolution dataset of the entire process and visualize it, and correct the initial launch parameters based on the simulation results to complete the numerical simulation of high-speed slender body water-breaking based on ALE.

Images