A multi-field synchronized control system and method in a laser melting deposition additive manufacturing process

Abstract

The application provides a multi-field synchronous control system and method in a laser melting deposition additive manufacturing process, the method comprising: constructing a non-steady-state cross-scale multi-time-step multi-coupling physical field model; inputting parameters, solving temperature field distribution, velocity field distribution, stress field distribution and strain field distribution; performing secondary grid division on a local paste region with a region temperature being a material melting point temperature, refining a time step, and researching heat and mass transport in a molten pool and an influence of the heat and mass transport on microstructure solidification of a solid / liquid interface; predicting a shape and size of a microstructure of the molten pool region based on a phase field model, solving molten pool composition field distribution and phase field distribution; based on the solved field distribution and microstructure distribution results, constructing a pore evaluation equation, a crack evaluation equation and a microstructure evaluation equation, and using a mass optimization model to iteratively optimize a multi-field parameter set to obtain optimal control parameters. The application can avoid the limitation of multi-field coupling local optimization, and improve production efficiency and manufacturing quality.

Description

Technical Field

[0001] This application relates to the field of multi-field control technology, specifically to a multi-field synchronous control system and method in the laser melting deposition additive manufacturing process. Background Technology

[0002] Laser metal deposition modeling (LMD) is a type of metal additive manufacturing technology that integrates computer-aided design, computer-aided manufacturing, computer numerical control, materials science, and precision servo technology. It can rapidly manufacture complex-shaped metal parts of various materials and densities with high performance through a single processing step, without the need for any hard molds or models, and without being limited by the shape or material of the parts being processed. After more than a decade of development, LMD has seen increasingly wider applications in the manufacturing field.

[0003] Although laser metal deposition modeling (LMD) technology has made significant progress, the concentrated energy input and rapid heating and cooling during the forming process result in large temperature gradients within the formed sample and between the sample and the substrate. This leads to strong directional selectivity in the sample's microstructure, affecting various properties. Simultaneously, the dramatic temperature gradient generates substantial thermal stress during forming, which can cause cracks or even fracture when the thermal stress reaches the material's limit. Therefore, it is crucial to investigate the influence of macroscopic process parameters on the evolution of the microstructure to improve the surface quality and microstructure of the formed parts, enhance their mechanical and physical properties, and prevent cracking during the forming process.

[0004] A literature search of existing technologies revealed that Chinese patent CN120065756A proposes a multi-field synchronous control system and method for laser additive manufacturing, which can explore the intrinsic mapping law between data of each field and finished product quality indicators under multi-physics fields, but does not involve the characterization of the dynamic micro-organism evolution process.

[0005] Most existing technologies for laser metal melting deposition (LMD) study the impact of process parameters on microstructure through experimental methods. However, cross-sections of samples only reflect the final microstructure formation and cannot dynamically explain its evolution. Furthermore, multi-scale models of the LMD process are rarely studied.

[0006] Therefore, there is an urgent need for a synchronous control system and method that can operate under multiple coupled physical fields. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the purpose of this application is to provide a multi-field synchronous control system and method for laser melting deposition additive manufacturing processes, thereby improving production efficiency and manufacturing quality.

[0008] According to one aspect of this application, a multi-field synchronous control method is provided in a laser melting deposition additive manufacturing process, comprising the following steps: Step 1: Construct an unsteady, multi-scale, multi-time-step, multi-coupled physical field model; Step 2: Collect the set of multi-field parameters currently input into the field, which is exactly the same as the set of verification experiments; Step 3: Input a set of multiple field parameters into the unsteady multi-scale multi-time-step multi-coupled physical field model, and use the governing equations and the simple algorithm to solve for the temperature field distribution and velocity field distribution within the molten pool region; load the macroscopic temperature field model, apply mechanical boundary conditions, and solve for the stress field distribution and strain field distribution; Step 4: Identify the regional temperature, perform secondary meshing on the local pasty region where the regional temperature is the melting point temperature of the material, and refine the time step to study the heat and mass transport in the molten pool and their influence on the solidification of the solid / liquid interface microstructure. Step 5: Based on the refined mesh and time step, the phase field model that calculates the temperature gradient G and solidification rate R is used to predict the morphology and size of the microstructure in the molten pool region; and the multi-component mass transfer equation based on the mixing averaging method is introduced into the non-isothermal hydrodynamic framework; the composition field distribution and phase field distribution of the molten pool are solved. Step Six: Based on the temperature field distribution, velocity field distribution, stress field distribution, strain field distribution, molten pool composition field distribution, phase field distribution, and microstructure distribution obtained in Steps Three to Five, construct the porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation. Then, use the quality optimization XG Boost-Bo model to iteratively optimize the set of multi-field parameters to obtain the optimal control parameters, thereby improving production efficiency and manufacturing quality.

[0009] Optionally, the construction of the unsteady, multi-scale, multi-time-step, multi-coupled physical field model in step one includes: Based on the heat source model, the heat conduction control equations are established using the laser-molten pool-powder / wire interaction model and boundary conditions, and a heat conduction model is constructed based on these control equations. A set of macroscopic-scale governing equations is established based on the continuity equation, momentum equation, and energy equation. A macroscopic thermodynamic coupling field model and a flow field model are then constructed based on the set of macroscopic-scale governing equations. A set of microscale governing equations was established using the phase field governing equations, the concentration field model, and the microscale temperature field.

[0010] Optionally, the heat conduction model is constructed as follows: the heat conduction model, heat source model, and laser beam-powder / wire-molten pool interaction model are described by the heat conduction control equation; the macroscopic thermo-mechanical coupling field distribution is calculated by giving boundary conditions and setting material thermophysical parameters; The flow field model is constructed as follows: the flow field model is established by constructing the continuity equation, momentum equation and energy equation; the flow field distribution is calculated by giving boundary conditions and setting material flow parameters.

[0011] Optionally, step two involves collecting a set of multi-field parameters currently input into the field, including: Collect laser parameters, workpiece material parameters, environmental parameters, and process parameters; The composition and phase diagram data, diffusion coefficient, distribution coefficient, and flow field and temperature field distribution required for the microscopic phase field and concentration field model are collected in advance to form a set of diffusion parameters; The mechanical property parameters, thermophysical parameters, boundary and loading conditions required for the macroscopic thermo-mechanical coupled field model are collected in advance to form a set of stress and strain parameters.

[0012] Optionally, step three involves inputting a set of multiple field parameters into the unsteady, multi-scale, multi-time-step, multi-coupled physics model to solve for the temperature and velocity field distributions within the molten pool region. Specifically, this includes: The laser parameters, workpiece material parameters, and boundary conditions from the multi-field parameter set are input into the unsteady multi-scale multi-time step multi-coupled physical field model. The finite element method or boundary element method is used to solve the model to obtain the macroscopic thermo-mechanical coupling field distribution and temperature field distribution. Input the process parameters and thermo-coupled field distribution results from the multi-field parameter set into the flow field model of the molten pool region, and use the finite element or boundary element method to solve the continuity equation, momentum equation and energy equation to obtain the velocity field distribution in the molten pool region.

[0013] Optionally, in step four, Based on the microscopic control equations, a second mesh generation and time step reselection are performed in the local mushy region where the temperature is the melting point of the cladding material.

[0014] Optionally, in step five, The finite element method or boundary element method is used to solve the set of governing equations at the microscale to obtain the molten pool concentration field distribution and phase field distribution within the molten pool region.

[0015] Optionally, in step six, The construction of the porosity evaluation equation includes: collecting the temperature field distribution and velocity field distribution of the sample workpiece as the first input features, correspondingly labeling whether each workpiece has porosity defects as the first label data, constructing a first training dataset, and using an artificial neural network to fit the mapping relationship between the first input features and the first label in the first training dataset as the porosity evaluation equation. The construction of the crack evaluation equation includes: collecting temperature field distribution, stress field distribution, and strain field distribution data of workpieces with and without cracks as second input features, correspondingly labeling the number of cracks in each workpiece as second label data, constructing a second training dataset, and using an artificial neural network to fit the mapping relationship between the second input features and the second label in the second training dataset as the crack evaluation equation. The construction of the microstructure evaluation equation includes: collecting temperature field distribution and melt pool composition field distribution data of the sample workpiece as the third input feature, collecting corresponding microstructure images or data, labeling grain size as the third label data, constructing a third training dataset, and using an artificial neural network to fit the mapping relationship between the third input feature and the third label in the third training dataset as the microstructure evaluation equation. Based on the constructed porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation, the quality optimization XGBoost-Bo model is solved and iterated continuously to obtain the suggested distribution values ​​of macroscopic thermo-mechanical coupling field distribution, velocity field distribution, concentration field distribution, and phase field distribution. Then, based on the suggested distribution values ​​of each field distribution and the corresponding model, the suggested boundary conditions that need to be controlled are deduced as multi-field control parameters.

[0016] Optionally, step six further includes: Based on the constitutive equations and stress field distribution results of the formed material, the metallic bonds are regarded as spring mode. Based on Hooke's law, porosity evaluation equations, crack evaluation equations, and microstructure evaluation equations are constructed to study the stress release process and crack propagation.

[0017] According to a second aspect of this application, a multi-field synchronous control system is provided for a laser melting deposition additive manufacturing process. This system implements a multi-field synchronous control method for a laser melting deposition additive manufacturing process, comprising a non-steady-state multi-scale multi-timestep multi-coupled physical field model construction module, a parameter collection module, a distributed solution module, and a conditional control module. These modules are electrically connected. The non-steady-state multi-scale multi-timestep multi-coupled physical field model construction module constructs a non-steady-state multi-scale multi-timestep multi-coupled physical field model and sends it to the distributed solution module. The parameter collection module collects the current set of multi-field parameters input into the field and sends the set to the distributed solution module to obtain the macroscopic thermo-coupling field distribution, flow field distribution, microscopic phase field distribution, microscopic concentration field distribution, and microscopic temperature field distribution. The conditional control module constructs a quality optimization function based on the porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation, solves the quality optimization XG Boost-Bo model, and then obtains the multi-field control parameters.

[0018] This application provides a multi-field synchronous control method for laser melting deposition additive manufacturing. First, an unsteady, multi-scale, multi-time-step, multi-coupled physical field model is established. The distributions of temperature, velocity, stress, strain, molten pool composition, phase, and microstructure are obtained. Based on these field distributions, porosity, crack, and microstructure evaluation equations are constructed. Then, the evaluation results of these equations are used to optimize the sets of parameters for each multi-field, achieving intelligent closed-loop control throughout the entire process. On the one hand, this method fully considers the strong coupling between multiple fields, avoiding the limitations of local optimization; on the other hand, it eliminates the need for manually compiling complex evaluation models through big data fitting, thus possessing broad applicability and improving production efficiency and manufacturing quality.

[0019] Other technical effects resulting from the additional features will be further illustrated in the corresponding embodiments. Attached Figure Description

[0020] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart of a multi-field synchronization control method in one embodiment of this application; Figure 2 This is a diagram illustrating the cladding defect identification process based on the XG Boost-Bo model in one embodiment of this application. Figure 3 This is a flowchart of a method for calculating the distribution results of various fields based on an unsteady, multi-scale, multi-time-step, multi-coupled physical field model in one embodiment of this application. Figure 4This is a schematic diagram of a multi-field synchronization control system according to an embodiment of this application. Detailed Implementation

[0021] The present application will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application, and these all fall within the protection scope of the present application. Parts not described in detail in the following embodiments can be implemented using existing technology.

[0022] Laser metal deposition modeling (LMD) technology has undergone more than a decade of development, and its applications in the manufacturing field are becoming increasingly widespread. However, most existing technologies for LMD study the influence of process parameters on microstructure through experimental methods. These methods only demonstrate the final microstructure formation through sample cross-sections and cannot dynamically explain the evolution of the microstructure. Furthermore, multi-scale models of the LMD process are rarely studied. To address these issues, this paper proposes a multi-field synchronous control system and method for LMD additive manufacturing to solve these problems.

[0023] Reference Figure 1 As shown, this application provides a multi-field synchronous control method in a laser melting deposition additive manufacturing process, including: Step 1: Construct an unsteady, multi-scale, multi-time-step, multi-coupled physical field model; Step 2: Collect the set of multi-field parameters currently input into the field. The set of multi-field parameters is exactly the same as that used in the verification experiment. Step 3: Input a set of multi-field parameters into the unsteady multi-scale multi-time-step multi-coupled physics field model, and use the governing equations and the simple algorithm to solve for the temperature field distribution and velocity field distribution within the molten pool region; load the macroscopic temperature field model, apply mechanical boundary conditions, and solve for the stress field distribution and strain field distribution; Step 4: Identify the regional temperature, perform secondary meshing on the local pasty region where the regional temperature is the melting point temperature of the material, and refine the time step to study the heat and mass transport in the molten pool and their influence on the solidification of the solid / liquid interface microstructure. Step 5: Based on the refined mesh and time step, the phase field model that calculates the temperature gradient G and solidification rate R is used to predict the morphology and size of the microstructure in the molten pool region; and the multi-component mass transfer equation based on the mixture-averaged method is introduced into the non-isothermal hydrodynamic framework; the composition field distribution and phase field distribution of the molten pool are solved. Step Six: Based on the temperature field distribution, velocity field distribution, stress field distribution, strain field distribution, molten pool composition field distribution, phase field distribution, and microstructure distribution obtained in Steps Three to Five, construct the porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation. Then, use the quality optimization XG Boost-Bo model to iteratively optimize the set of multi-field parameters to obtain the optimal control parameters, thereby improving production efficiency and manufacturing quality.

[0024] In some specific embodiments of this application, the construction of the unsteady, multi-scale, multi-time-step, multi-coupled physical field model in step one may further include: Based on the heat source model, the heat conduction control equations are established using the laser-molten pool-powder / wire interaction model and boundary conditions, and a heat conduction model is constructed based on these control equations. A set of macroscopic governing equations is established based on the continuity equation, the momentum equation (Navier-Stokes equations), and the energy equation. Based on the set of macroscopic governing equations, a macroscopic thermodynamic coupling field model and a flow field model are constructed. A set of microscale governing equations was established using the phase field governing equations (Karma phase field model), the concentration field model, and the microscale temperature field.

[0025] In some specific embodiments of this application, the heat conduction model is constructed as follows: the heat conduction model, the heat source model, and the laser beam-powder / wire-molten pool interaction model are described by the heat conduction control equation; the macroscopic thermo-mechanical coupling field distribution is calculated by giving boundary conditions and setting material thermophysical parameters; The flow field model is constructed as follows: the flow field model is established by constructing the continuity equation, momentum equation and energy equation; the flow field distribution is calculated by giving boundary conditions and setting material flow parameters.

[0026] In some specific embodiments of this application, step two, collecting the set of multi-field parameters currently input into the field, may further include: Collect laser parameters, workpiece material parameters, environmental parameters, and process parameters; The composition and phase diagram data, diffusion coefficient, distribution coefficient, and flow field and temperature field distribution required for the microscopic phase field and concentration field model are collected in advance to form a set of diffusion parameters; The mechanical property parameters, thermophysical parameters, boundary and loading conditions required for the macroscopic thermo-mechanical coupled field model are collected in advance to form a set of stress and strain parameters.

[0027] In some specific embodiments of this application, step three involves inputting a set of multiple field parameters into the unsteady, multi-scale, multi-time-step, multi-coupled physics model to solve for the temperature and velocity field distributions within the molten pool region. Specifically, this includes: The laser parameters, workpiece material parameters and boundary conditions from the multi-field parameter set are input into the unsteady multi-scale multi-time step multi-coupled physical field model. The finite element method or boundary element method is used to solve the model and obtain the macroscopic thermo-mechanical coupling field distribution and temperature field distribution. Input the process parameters and thermo-coupled field distribution results from the multi-field parameter set into the flow field model of the molten pool region, and use the finite element or boundary element method to solve the continuity equation, momentum equation and energy equation to obtain the velocity field distribution in the molten pool region.

[0028] In some specific embodiments of this application, step four may further include: Based on the microscopic control equations, a second mesh generation and time step reselection are performed in the local mushy region where the temperature is the melting point of the cladding material.

[0029] In some specific embodiments of this application, step five may further include: The finite element method or boundary element method is used to solve the set of governing equations at the microscale to obtain the molten pool concentration field distribution and phase field distribution within the molten pool region.

[0030] In some specific embodiments of this application, in step six, The construction of the porosity evaluation equation includes: collecting the temperature field distribution and velocity field distribution of the sample workpiece as the first input features, correspondingly labeling each workpiece to indicate whether there are porosity defects as the first label data, constructing the first training dataset, and using an artificial neural network to fit the mapping relationship between the first input features and the first label in the first training dataset as the porosity evaluation equation; The construction of the crack assessment equation includes: collecting temperature field distribution, stress field distribution, and strain field distribution data of workpieces with and without cracks as second input features, correspondingly labeling the number of cracks in each workpiece as second label data, constructing a second training dataset, and using an artificial neural network to fit the mapping relationship between the second input features and the second labels in the second training dataset as the crack assessment equation. The construction of the microstructure evaluation equation includes: collecting temperature field distribution and melt pool composition field distribution data of sample workpieces as third input features, collecting corresponding microstructure images or data, labeling grain size as third label data, constructing a third training dataset, and using an artificial neural network to fit the mapping relationship between the third input features and the third label in the third training dataset as the microstructure evaluation equation. Based on the constructed porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation, the quality optimization XGBoost-Bo model is solved and iterated continuously to obtain the suggested distribution values ​​of macroscopic thermo-mechanical coupling field distribution, velocity field distribution, concentration field distribution, and phase field distribution. Then, based on the suggested distribution values ​​of each field distribution and the corresponding model, the suggested boundary conditions that need to be controlled are deduced as multi-field control parameters.

[0031] In some specific embodiments of this application, step six further includes: Based on the constitutive equations and stress field distribution results of the formed material, the metallic bonds are regarded as spring mode. Based on Hooke's law, porosity evaluation equations, crack evaluation equations, and microstructure evaluation equations are constructed to study the stress release process and crack propagation.

[0032] In summary, the multi-field synchronous control method in the laser melting deposition additive manufacturing process of this application specifically includes the following steps: constructing an unsteady, multi-scale, multi-time-step, multi-coupled physical field model; inputting a set of multi-field parameters (including laser parameters, processing parameters, environmental parameters, and material thermophysical parameters, etc.) that are identical to those in the verification experiment into the constructed unsteady, multi-scale, multi-time-step, multi-coupled physical field model; solving for the macroscopic temperature field and temperature gradient distribution, and solving for the temperature field distribution and velocity field distribution within the molten pool region through the control equations and the Simple algorithm; identifying the regional temperature, performing secondary meshing on the local mushy region where the regional temperature is the melting point temperature of the material, and refining the time step to study the heat and mass transport within the molten pool and their effects on the molten pool. The influence of solid / liquid interface microstructure on solidification: Based on the refined mesh and time step, the morphology and size of the molten pool microstructure are predicted by calculating the temperature gradient G and solidification rate R using a phase-field model. A multi-component mass transfer equation based on the mixture-averaged method is introduced into the non-isothermal hydrodynamic framework to solve for the molten pool composition field distribution and phase field distribution. Simultaneously, based on the temperature field distribution, velocity field distribution, stress field distribution, strain field distribution, molten pool composition field distribution, phase field distribution, and microstructure distribution obtained from the above steps, mechanical boundary conditions are applied to solve for the stress field distribution and strain field distribution, constructing porosity evaluation equations, crack evaluation equations, and microstructure evaluation equations. Then, based on the XG Boost-Bo model, a quality optimization XGBoost-Bo method is constructed to obtain multi-field control parameters. The XG Boost-Bo model (Bo-XGBoost model) can be used to identify cladding defects to optimize multi-field control parameters.

[0033] For example, refer to Figure 2As shown, the XGBoost model is initialized with parameters. The initialization of the model is performed based on BO (Bayesian optimization), initial points are randomly generated, a Gaussian process is introduced, the search space boundary is checked and modified, and the optimization is iterated until the terminal condition is met, finally obtaining the optimal Bo-XGBoost prediction model. The performance is evaluated using this prediction model.

[0034] This application constructs a non-steady-state, multi-scale, multi-time-step, multi-coupled physical field model encompassing temperature, flow, phase, concentration, and stress-strain fields. It introduces machine learning algorithms to uncover the inherent mapping patterns between data from each field and finished product quality indicators, and constructs predictive equations such as porosity assessment equations, crack assessment equations, and microstructure assessment equations. Based on the assessment results, it optimizes various process parameters to achieve intelligent closed-loop control throughout the entire process. This application, on the one hand, fully considers the strong coupling effects between multiple fields, avoiding the limitations of local optimization and clarifying the influence of macroscopic process parameters on microstructure, elemental distribution, and stress-strain. On the other hand, through big data fitting, it eliminates the need for manually developing complex assessment models, thus possessing broad applicability and improving production efficiency and manufacturing quality.

[0035] In the above embodiments of this application, the construction principle of the unsteady multi-scale multi-time-step multi-coupled physical field model is as follows: 1) Heat source model: During laser processing, the laser heat source transfers heat energy to the sample through a specific area, called the heating spot. The heat distribution on the heating spot is non-uniform, with more heat at the center and less at the edges. Friedmann approximates the distribution of heat flux density on the heating spot using the following Gaussian mathematical model: (1) In equation (1), The Gaussian distribution density of the heat source, Indicates the laser radius. The total intensity of laser energy. The effective radius of the laser beam is denoted as .

[0036] 2) Macroscopic scale governing equations: At a certain moment during continuous laser scanning, in a Cartesian coordinate system, the classical laser melting and solidification control equation can be expressed as: Continuity equation (mass conservation equation): (2) Momentum equation (Navier-Stokes equation): (3a) (3b) (3c) Energy equation (enthalpy form): (4) In equations (2)-(4), Let t represent density, and t represent time. These are the projections of the velocity vector U onto the x, y, and z axes, respectively; For manifest enthalpy; For pressure: in the equation It is the source item. The distribution consists of liquid phase viscosity and thermal diffusivity. , Thermal conductivity, Specific heat capacity.

[0037] The aforementioned set of fundamental macroscopic governing equations is a system of differential equations written with the control volume on the matrix as the object, describing the conservation relationships of the control volume's mass, momentum, and energy. The source term of the momentum equation... To account for the buoyancy effect caused by the non-uniformity of liquid alloy density due to temperature differences, the Boussinesq assumption is adopted, expressed as: (5) In equation (5), The coefficient of thermal expansion is Here is the visible enthalpy reference value; g is the acceleration due to gravity, and this source term only exists in the direction of gravity. The specific heat is at constant pressure. This system of equations holds true throughout the entire computational domain, meaning the entire computational domain is considered a continuous unified body. The Level Set method is then used to solve the problem of tracing the free surface of the molten pool.

[0038] 3) Interaction between laser and powder: When the laser passes through the coaxial flow of metal powder emitted from the nozzle, the energy decreases due to the reflection, absorption, and scattering effects of the metal powder. Simultaneously, the powder's temperature rises and may even undergo a phase change (melting and sublimation) before reaching the substrate. Experiments have verified that the spatial distribution of the coaxial powder flow, viewed from the side, is approximately a Gaussian distribution, as expressed by equation (6): (6) In equation (6), N The quantity of powder per unit volume is denoted by , and the radial distance is denoted by , i.e., the laser radius. r and axial distance l The function, At the center of the powder flow ( peak mass fraction In axial distance The effective radius at that point. According to the beer-Lambert law, the attenuation rate of the laser beam decreases as it travels a certain distance. This can then be represented as: (7) In equation (7), It is the energy density after decay. It is the undiminished energy density. The melting rate of the powder is given. The intensity of the laser gradually decreases with the axial distance (from the nozzle exit to the substrate surface). The decrease in the power density of the upper laser layer is taken as the laser power density in the adjacent layer.

[0039] When metal powder particles absorb laser energy, their temperature rises. This temperature rise can be calculated using the following heat balance equation: (8) In equation (8), It is the absorption coefficient of the powder. It is the radius of the powder particles. It's the powder speed. It is the difference in axial distance. It is the density of the powder. It is the specific heat capacity at constant pressure of the powder. This is the temperature rise of the powder. The temperature rise of the powder can be calculated using the energy density decay layer by layer along the axial direction. Absorption coefficient With temperature T The formula for the relationship between them is as follows: (9) In equation (9): Laser power; The stimulated absorption coefficient; The density of the total number of atoms; The energy level degeneracy; Planck's constant (6.626 × 10⁻⁶) -34 J·s ); The speed of light in a vacuum; For electronic quality; It is the elementary charge; The central angular frequency of atomic transitions; ω is the angular frequency of the incident light; The damping coefficient; Boltzmann constant (1.38 × 10⁻⁶) -23 J / K ).

[0040] 4) Interaction between powder and molten pool: The boundary conditions are shown in equation (10): (10) In equation (10), This represents the energy density value at the surface of the molten pool. The Gaussian distribution density of the heat source, Indicates the laser radius. This refers to the scanning speed. The total intensity of laser energy. The effective radius of the laser beam is given; the actual effective absorption of laser energy is determined by the absorptivity of the workpiece material. and laser incident angle The empirical formula for the absorption coefficient is determined as follows: ), For the logarithm of thermal forced convection, For ambient temperature, It is the Stefan-Poltzmann constant. Emissivity.

[0041] (11) In equation (11), It is the melting point temperature. For the latent enthalpy of phase transition, This is due to excess energy derived from heated powder particles. The rate at which mass is added to the molten pool. These represent the mass percentages of the liquid phase and the solid phase, respectively. The density of the material in its liquid state. These are the specific heat capacities of the solid and liquid phases, respectively. This represents the mass distribution of the powder before it reaches the substrate.

[0042] 5) Phase-field method simulation model of dendrite growth: In the phase-field method simulation of dendrite growth, a diffusion interface model is adopted, and phase-field variables are introduced. This represents the physical state (solid or liquid) of a system in time and space. The phase field has a constant value for each phase in the system; for example, =1 indicates the solid phase region. =0 indicates the liquid phase region, at the solid-liquid interface. The value changes continuously between 1 and 0.

[0043] The following conditions and assumptions are set: 1. Since the solidification process in this application exists only in a very small selected unit, the effect of fluid flow is not considered.

[0044] 2. Thermophysical parameters are calculated using the continuum method.

[0045] 3. Considering that the diffusion of solid solutes is much smaller than that of liquids, we only consider the diffusion of liquid solutes.

[0046] 4. For simplicity, the solid-liquid partition coefficient is set to the equilibrium state.

[0047] The governing equations are expressed as follows: Since the main experimental material involved in this application is TiAl6V4, which is a binary titanium-aluminum alloy, its phase-field model is as follows: 1. The phase-field governing equations (Karma phase-field model) are as follows: (12) In equation (12), For free energy; A function constructed to account for the latent heat released at the interface during dendrite growth. Definition and Introducing interface anisotropy, where, Interface thickness; This refers to the relaxation time; The variable characterizing the time of atomic motion at the solid-liquid interface; For anisotropy factor, , The angle between the dendrite principal axis and the interface normal. , The angle between the dendrite principal axis and the tangential direction of the interface is taken as 45°; The anisotropy index; Let the modulus be anisotropic, and take... Among them, the double potential well function Phase transition thermodynamic driving force term of phase A Phase transition thermodynamic driving force term of phase B The expression is as follows: (13a) (13b) (13c) In equation (13), , These are the latent heat of phase change for phase A and phase B, respectively. , These are the latent heat of phase change of phase A and the phase change temperature of phase B, respectively. 2. Concentration field model: (14) In equation (14), Let c be the gas constant, and c represent the concentration. This refers to the solute diffusion and migration rate. Volume is the molar volume; The regularization length parameter, The solid-phase solute diffusion coefficient is... is the diffusion coefficient of the solute in the liquid phase.

[0048] 3. Microscopic temperature field model: (15) In equation (15), is the thermal diffusivity.

[0049] Numerical stability: By using macroscopic grid computation, the linear numerical stability of the numerical difference equation can be expressed as: (16) In equation (16), This represents the maximum steady-state time step of the heat conduction equation; The thermal diffusivity of the material; , These are the spatial grid step sizes in the x and y directions, respectively. is the maximum stable time step of the phase-field equation.

[0050] In the phase-field model, the time step is taken as: (17) Reference Figure 3 As shown, the method for obtaining different physical field distributions based on the unsteady, multi-scale, multi-time-step, multi-coupled physical field model is as follows: First, initialize the temperature T, pressure, velocity, and VOF (volume fraction method) equations. Then, solve the momentum equation, VOF equation, and energy equation sequentially, and update the temperature T, pressure p, the projection u of the velocity vector U on the x-axis, and the projection v of the velocity vector U on the y-axis. When the preset supercooling is reached, initialize the temperature and component distribution phase field variables (temperature T, concentration c, relaxation time). The phase-field equation and energy equation are then derived to update the temperature T, concentration c, and relaxation time. .

[0051] Reference Figure 4As shown, based on the same inventive concept, another embodiment of this application provides a multi-field synchronous control system for the laser melting deposition additive manufacturing process, used to realize a multi-field synchronous control method in the laser melting deposition additive manufacturing process. The multi-field synchronous control system 100 includes an unsteady multi-scale multi-time step multi-coupled physical field model construction module 110, a parameter collection module 120, a distributed solution module 130, and a condition control module 140; wherein, each module is electrically connected; the unsteady multi-scale multi-time step multi-coupled physical field model construction module is used to construct an unsteady multi-scale multi-time step multi-coupled physical field model and send it to the distributed solution module; the parameter collection module is used to collect the set of multi-field parameters currently input into the field and send the set of multi-field parameters to the distributed solution module to obtain the macroscopic thermo-coupling field distribution, flow field distribution, microscopic phase field distribution, microscopic concentration field distribution, and microscopic temperature field distribution, etc.; wherein, the parameter collection module includes an acquisition module, a control equation module, a partitioning module, and a microstructure prediction module, the acquisition module is used to collect The system currently inputs a set of multiple field parameters into the field. The governing equation module is used to input this set of parameters into the unsteady, multi-scale, multi-time-step, multi-coupled physical field model. It uses the governing equations and the simple algorithm to solve for the temperature and velocity field distributions within the molten pool region. It then loads the macroscopic temperature field model, applies mechanical boundary conditions, and solves for the stress and strain field distributions. The meshing module identifies the regional temperature, performs secondary meshing on localized mushy regions where the temperature is the material's melting point, and refines the time step to study heat and mass transport within the molten pool and their impact on the solidification of the solid / liquid interface microstructure. The microstructure prediction module, based on the refined mesh and time step, uses a phase-field model that calculates the temperature gradient G and solidification rate R to predict the morphology and size of the microstructure in the molten pool region. It also introduces a multi-component mass transfer equation based on the mixed averaging method into the non-isothermal hydrodynamic framework, solving for the molten pool composition field distribution and phase field distribution. Finally, the condition control module constructs a mass optimization function based on the porosity assessment equation, crack assessment equation, and microstructure assessment equation to solve for the mass optimization XG Boost-Bo model, thereby obtaining the multiple field control parameters.

[0052] The parts not explicitly described in the above embodiments can be implemented with reference to existing technologies.

[0053] Furthermore, the parts of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration. The preset parameters or preset thresholds mentioned above have all been set by those skilled in the art based on actual conditions or obtained through simulation with a large amount of data.

[0054] The preferred features in the above embodiments can be used individually in any embodiment, or in any combination thereof, provided they do not conflict with each other. Furthermore, parts not described in detail in the embodiments can be implemented using existing technologies.

[0055] The foregoing has described some specific embodiments of this application. It should be understood that this application is not limited to the specific embodiments described above, and those skilled in the art can make various modifications or variations within the scope of the claims, which do not affect the substantive content of this application. The above-described preferred features can be used in any combination without conflict.

Claims

1. A multi-field synchronous control method in a laser melting deposition additive manufacturing process, characterized in that, include: Step 1: Construct an unsteady, multi-scale, multi-time-step, multi-coupled physical field model; Step 2: Collect the set of multi-field parameters currently input into the field, which is exactly the same as the set of parameters in the verification experiment; Step 3: Input a set of multiple field parameters into the unsteady multi-scale multi-time-step multi-coupled physical field model, and use the governing equations and the simple algorithm to solve for the temperature field distribution and velocity field distribution within the molten pool region; load the macroscopic temperature field model, apply mechanical boundary conditions, and solve for the stress field distribution and strain field distribution; Step 4: Identify the regional temperature, perform secondary meshing on the local pasty region where the regional temperature is the melting point temperature of the material, and refine the time step to study the heat and mass transport in the molten pool and their influence on the solidification of the solid / liquid interface microstructure. Step 5: Based on the refined mesh and time step, the phase field model that calculates the temperature gradient G and solidification rate R is used to predict the morphology and size of the microstructure in the molten pool region; and the multi-component mass transfer equation based on the mixing averaging method is introduced into the non-isothermal hydrodynamic framework; the composition field distribution and phase field distribution of the molten pool are solved. Step Six: Based on the temperature field distribution, velocity field distribution, stress field distribution, strain field distribution, molten pool composition field distribution, phase field distribution, and microstructure distribution obtained in Steps Three to Five, construct the porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation. Then, use the quality optimization XG Boost-Bo model to iteratively optimize the set of multi-field parameters to obtain the optimal control parameters, thereby improving production efficiency and manufacturing quality.

2. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 1, characterized in that, Step one involves constructing an unsteady, multi-scale, multi-time-step, multi-coupled physical field model, including: Based on the heat source model, the heat conduction control equations are established using the laser-molten pool-powder / wire interaction model and boundary conditions, and a heat conduction model is constructed based on these control equations. A set of macroscopic-scale governing equations is established based on the continuity equation, momentum equation, and energy equation. Based on the set of macroscopic-scale governing equations, a macroscopic thermo-mechanical coupled field model and a flow field model are constructed. A set of microscale governing equations was established using the phase field governing equations, the concentration field model, and the microscale temperature field.

3. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 2, characterized in that, The heat conduction model is constructed as follows: the heat conduction model, heat source model, and laser beam-powder / wire-molten pool interaction model are described by the heat conduction control equation; the macroscopic thermo-mechanical coupling field distribution is calculated by giving boundary conditions and setting material thermophysical parameters. The flow field model is constructed as follows: the flow field model is established by constructing the continuity equation, momentum equation and energy equation; the flow field distribution is calculated by giving boundary conditions and setting material flow parameters.

4. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 2, characterized in that, Step two involves collecting the set of multi-field parameters currently being input into the field, including: Collect laser parameters, workpiece material parameters, environmental parameters, and process parameters; The composition and phase diagram data, diffusion coefficient, distribution coefficient, and flow field and temperature field distribution required for the microscopic phase field and concentration field model are collected in advance to form a set of diffusion parameters; The mechanical property parameters, thermophysical parameters, boundary and loading conditions required for the macroscopic thermo-mechanical coupled field model are collected in advance to form a set of stress and strain parameters.

5. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 4, characterized in that, Step three involves inputting a set of multiple field parameters into the unsteady, multi-scale, multi-time-step, multi-coupled physics model to solve for the temperature and velocity field distributions within the molten pool region. Specifically, this includes: The laser parameters, workpiece material parameters, and boundary conditions from the multi-field parameter set are input into the unsteady multi-scale multi-time step multi-coupled physical field model. The finite element method or boundary element method is used to solve the model to obtain the macroscopic thermo-mechanical coupling field distribution and temperature field distribution. Input the process parameters and thermo-coupled field distribution results from the multi-field parameter set into the flow field model of the molten pool region, and use the finite element or boundary element method to solve the continuity equation, momentum equation and energy equation to obtain the velocity field distribution in the molten pool region.

6. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 2, characterized in that, In step four, based on the microscopic control equation, a second mesh division and time step reselection are performed in the local pasty region where the temperature is the melting point of the cladding material.

7. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 6, characterized in that, In step five, the finite element method or boundary element method is used to solve the set of microscale governing equations to obtain the molten pool concentration field distribution and phase field distribution within the molten pool region.

8. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 1, characterized in that, In step six, The construction of the porosity evaluation equation includes: collecting the temperature field distribution and velocity field distribution of the sample workpiece as the first input features, correspondingly labeling whether each workpiece has porosity defects as the first label data, constructing a first training dataset, and using an artificial neural network to fit the mapping relationship between the first input features and the first label in the first training dataset as the porosity evaluation equation. The construction of the crack evaluation equation includes: collecting temperature field distribution, stress field distribution, and strain field distribution data of workpieces with and without cracks as second input features, correspondingly labeling the number of cracks in each workpiece as second label data, constructing a second training dataset, and using an artificial neural network to fit the mapping relationship between the second input features and the second label in the second training dataset as the crack evaluation equation. The construction of the microstructure evaluation equation includes: collecting temperature field distribution and melt pool composition field distribution data of the sample workpiece as the third input feature, collecting corresponding microstructure images or data, labeling grain size as the third label data, constructing a third training dataset, and using an artificial neural network to fit the mapping relationship between the third input feature and the third label in the third training dataset as the microstructure evaluation equation. Based on the constructed porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation, the quality optimization XGBoost-Bo model is solved and iterated continuously to obtain the suggested distribution values ​​of macroscopic thermo-mechanical coupling field distribution, velocity field distribution, concentration field distribution, and phase field distribution. Then, based on the suggested distribution values ​​of each field distribution and the corresponding model, the suggested boundary conditions that need to be controlled are deduced as multi-field control parameters.

9. The multi-field synchronous control method in the laser melting deposition additive manufacturing process according to claim 1, characterized in that, Step six also includes: Based on the constitutive equations and stress field distribution results of the formed material, the metallic bonds are regarded as spring mode. Based on Hooke's law, porosity evaluation equations, crack evaluation equations, and microstructure evaluation equations are constructed to study the stress release process and crack propagation.

10. A multi-field synchronous control system for laser melting deposition additive manufacturing process, used to implement the multi-field synchronous control method for laser melting deposition additive manufacturing process according to any one of claims 1-9, characterized in that, The system comprises a nonsteady-state, multi-scale, multi-time-step, multi-coupled physical field model construction module, a parameter collection module, a distributed solution module, and a condition control module. These modules are electrically connected. The nonsteady-state, multi-scale, multi-time-step, multi-coupled physical field model construction module constructs the nonsteady-state, multi-scale, multi-time-step, multi-coupled physical field model and sends it to the distributed solution module. The parameter collection module collects the set of multi-field parameters currently input into the field and sends this set to the distributed solution module to obtain the macroscopic thermo-coupling field distribution, flow field distribution, microscopic phase field distribution, microscopic concentration field distribution, and microscopic temperature field distribution. The condition control module constructs a quality optimization function based on the porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation to solve the quality optimization XG Boost-Bo model, thereby obtaining the multi-field control parameters.

Images