Multi-field synchronous control system and method in laser additive manufacturing

Abstract

The present invention discloses a multi-field synchronous control system and method in laser additive manufacturing, which relates to the technical field of multi-field control. By constructing an electromagnetic-thermal-mechanical coupling model, a mass diffusion model, and a stress-strain model, inputting a multi-field parameter set into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool region; inputting the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model to solve the molten pool composition field distribution; inputting the multi-field parameter set and temperature field distribution into the stress-strain model to solve the stress field distribution and strain field distribution, constructing a pore evaluation equation, a crack evaluation equation, and a microstructure evaluation equation, then constructing a quality optimization function, using a swarm intelligence algorithm to solve the quality optimization function, solving the quality optimization model, and obtaining multi-field control parameters; improving production efficiency and manufacturing quality.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-field control, and particularly to a multi-field synchronous control system and method in laser additive manufacturing. Background Art

[0002] Multi-field laser additive manufacturing technology has received extensive attention and application in recent years. However, due to its complex physical process and the mutual influence of different fields, the control of the finished product quality has become a major challenge.

[0003] The multi-field laser additive manufacturing process involves the coupling of multiple physical fields such as heat conduction, fluid dynamics, phase change, solute diffusion, stress and strain. The temperature field directly affects the melting degree and microstructure morphology of the material; the disturbance of the flow field will lead to the instability of the molten pool and the generation of defects; the uneven distribution of the composition field will cause tissue segregation. At the same time, thermal stress and residual stress will also have a significant impact on the mechanical properties of the finished product.

[0004] Existing technologies have deficiencies in multi-field control. The common practice is to simply control each field at the optimal level, but the close coupling between fields is ignored. For example, only optimizing the thermal cycle parameters based on the temperature field without considering the response of the stress-strain field and the tissue field may ultimately result in defects and performance degradation. Other methods use empirical coefficients and are difficult to achieve precise control.

[0005] In order to break through the restriction of multi-field coupling on quality control, a new technical solution that can comprehensively model and optimize the multi-field influence is urgently needed.

[0006] The Chinese patent with the authorization announcement number CN107368642B discloses a multi-scale multi-physical field coupling simulation method for metal additive manufacturing. The method includes the following steps: S1, establishing a process data model for metal additive manufacturing; S2, at the microscale, carrying out first-principles calculations through first-principles calculation software to obtain the microphysical properties of the additive metal material; S3, establishing an N×N×N supercell model of the additive metal material and carrying out molecular dynamics simulation calculations through molecular dynamics simulation software; S4, at the mesoscale, studying the plasma generated during the melting process of metal powder heated by an electron beam or a laser; S5, carrying out simulation calculations using a fluid-thermal-solid-magnetic multi-physical field coupling simulation platform; S6, establishing a process parameter feedback control model for different types and distribution conditions of defects to optimize the process parameters of metal additive manufacturing. However, the quantitative correlation model established by this method is between multiple process parameters and defect characteristics, and does not involve multi-field distribution, nor can it control process parameters by analyzing multi-field distribution.

[0007] Therefore, the present invention proposes a multi-field synchronous control system and method in laser additive manufacturing. Summary of the Invention

[0008] The present invention aims to solve at least one of the technical problems existing in the prior art. For this purpose, the present invention proposes a multi-field synchronous control system and method in laser additive manufacturing, which improves production efficiency and manufacturing quality.

[0009] To achieve the above object, a multi-field synchronous control method in laser additive manufacturing is proposed, including the following steps:

[0010] Step 1: Construct an electromagnetic-thermal-mechanical coupling model, a mass diffusion model, and a stress-strain model;

[0011] Step 2: Collect the multi-field parameter set input into the current field;

[0012] Step 3: Input the multi-field parameter set into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool area; input the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model to solve the composition field distribution of the molten pool; input the multi-field parameter set and temperature field distribution into the stress-strain model to solve the stress field distribution and strain field distribution.

[0013] Step 4: Construct a pore evaluation equation, a crack evaluation equation, and a microstructure evaluation equation through the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution;

[0014] Step 5: Construct a quality optimization function based on the pore evaluation equation, crack evaluation equation, and microstructure evaluation equation, use a swarm intelligence algorithm to solve the quality optimization function, solve the quality optimization model, and obtain multi-field control parameters;

[0015] The construction process of the electromagnetic-thermal-mechanical coupling model includes:

[0016] Construct an electromagnetic field model based on Maxwell's equations;

[0017] Construct a heat conduction model based on the heat conduction control equation;

[0018] Construct a flow field model based on the energy, momentum, and mass conservation control equations;

[0019] The electromagnetic field model, heat conduction model, and flow field model together constitute the electromagnetic-thermal-mechanical coupling model;

[0020] The method for constructing the electromagnetic field model is:

[0021] Describe the electromagnetic field distribution with Maxwell's equations;

[0022] Calculate the electromagnetic field distribution J by giving boundary conditions and setting the electromagnetic parameters of the material;

[0023] The method for constructing the heat conduction model is:

[0024] Describe the heat conduction model with the heat conduction control equation;

[0025] By giving the boundary conditions and setting the material heat conduction parameters, calculate the temperature field distribution;

[0026] The method for constructing the flow field model is as follows:

[0027] Construct the flow field model with the energy, momentum, and mass conservation control equations;

[0028] By giving the boundary conditions and setting the material flow parameters, calculate the flow field distribution;

[0029] The method for constructing the mass diffusion model is as follows:

[0030] Use the diffusion equation to describe the mass diffusion process;

[0031] Given the initial conditions, convective boundary conditions, and convective / diffusive boundary conditions;

[0032] Introduce the phase change conditions;

[0033] Based on the diffusion equation and the phase change conditions, establish the mass control equations, and use the finite element or finite difference method to solve the mass control equations, and iteratively update the temperature field and velocity field until convergence to obtain the converged molten pool composition field distribution;

[0034] The method for constructing the stress-strain model is as follows:

[0035] Construct the generalized deflection constitutive equation, and introduce the strain-displacement relationship condition and the stress equilibrium equation as the constraint conditions to form the stress-strain model;

[0036] The method for collecting the multi-field parameter set input into the current field is as follows:

[0037] Pre-collect the laser parameters, workpiece material parameters, environmental parameters, and process parameters required by the electromagnetic-thermal-mechanical coupling model to form the electromagnetic parameter set;

[0038] Pre-collect the composition and phase diagram data, diffusion coefficient, distribution coefficient, and flow field and temperature field distribution required by the mass diffusion model to form the diffusion parameter set;

[0039] Pre-collect the mechanical property parameters, thermophysical parameters, boundary and loading conditions required by the stress-strain model to form the stress-strain parameter set;

[0040] The electromagnetic parameter set, the diffusion parameter set, and the stress-strain parameter set form the multi-field parameter set;

[0041] The method of inputting the multi-field parameter set into the electromagnetic-thermal coupling model and solving the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool area is as follows:

[0042] Input the laser parameters and workpiece material parameters in the electromagnetic parameter set of the multi-field parameter set into the electromagnetic field model, and use the finite element or boundary element method to solve the Maxwell equations to obtain the electromagnetic field distribution J of the current density in the molten pool area;

[0043] Input the environmental parameters, workpiece material parameters, and electromagnetic field distribution J in the electromagnetic parameter set of the multi-field parameter set into the heat conduction model, and use the finite element or boundary element method to solve the heat conduction control equations to obtain the temperature field distribution T in the molten pool area;

[0044] Input the process parameters in the electromagnetic parameter set of the multi-field parameter set and the temperature field distribution T into the flow field model, and use the finite element or boundary element method to solve the energy, momentum, and mass conservation control equations to obtain the velocity field distribution u in the molten pool area;

[0045] The method of inputting the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model and solving the molten pool composition field distribution is as follows:

[0046] Input the diffusion parameter set in the multi-field parameter set and the temperature field distribution T into the mass diffusion model, and use the finite element or boundary element method to solve the mass control equations to obtain the molten pool composition field distribution C in the molten pool area;

[0047] The method of inputting the multi-field parameter set and temperature field distribution into the stress-strain model and solving the stress field distribution and strain field distribution is as follows:

[0048] By inputting the temperature field distribution and the stress-strain parameter set in the multi-field parameter set into the stress-strain model, and using the finite element or finite difference method to solve the stress-strain equations, the stress field distribution and strain field distribution are obtained;

[0049] The method of constructing the porosity evaluation equation, crack evaluation equation, and microstructure evaluation equation is as follows:

[0050] The construction of the porosity evaluation equation includes:

[0051] Collect the temperature field distribution, velocity field distribution, and electromagnetic field distribution of the sample workpieces as the first input features, and correspondingly mark whether each workpiece has porosity defects as the first label data, construct the first training data set, and use an artificial neural network to fit the mapping relationship between the first input features and the first labels in the first training data set as the porosity evaluation equation;

[0052] The construction of the crack evaluation equation includes:

[0053] Collect the temperature field distribution, stress field distribution, and strain field distribution data of cracked and non-cracked workpieces as the second input features, and correspondingly mark the number of cracks in each workpiece as the second label data. Construct a second training dataset, and use 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 evaluation equation;

[0054] The construction of the microstructure evaluation equation includes:

[0055] Collect the temperature field distribution and molten pool composition field distribution data of sample workpieces as the third input features, collect the corresponding microstructure images or data, and label the grain size as the third label data. Construct a third training dataset, and use an artificial neural network to fit the mapping relationship between the third input features and the third labels in the third training dataset as the microstructure evaluation equation;

[0056] The method for solving the quality optimization model to obtain the multi-field control parameters is as follows:

[0057] Through the use of the gradient descent algorithm for backpropagation of the error of the quality optimization function and continuous iteration, obtain the proposed distribution values of the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution. Then, based on the proposed distribution values of each field distribution and the corresponding models, inversely deduce the proposed boundary conditions of each input that need to be controlled as the multi-field control parameters.

[0058] Propose a multi-field synchronous control system in laser additive manufacturing, including a multi-field model construction module, a parameter collection module, a distribution solution module, and a condition control module; among them, each module is connected electrically;

[0059] The multi-field model construction module constructs an electromagnetic-thermal-mechanical coupling model, a mass diffusion model, and a stress-strain model, and sends the electromagnetic-thermal-mechanical coupling model, the mass diffusion model, and the stress-strain model to the distribution solution module;

[0060] The parameter collection module collects the multi-field parameter set input into the current field and sends the multi-field parameter set to the distribution solution module;

[0061] The distribution solution module inputs the multi-field parameter set into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool area; inputs the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model to solve the molten pool composition field distribution; inputs the multi-field parameter set and temperature field distribution into the stress-strain model to solve the stress field distribution and strain field distribution, and sends the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution to the condition control module;

[0062] The condition control module constructs a quality optimization function based on the pore evaluation equation, crack evaluation equation, and tissue evaluation equation, and uses the swarm intelligence algorithm to solve the quality optimization function, solve the quality optimization model, and obtain multi-field control parameters.

[0063] A kind of electronic device is proposed, including: a processor and a memory, wherein, a computer program that can be called by the processor is stored in the memory;

[0064] The processor executes the multi-field synchronous control method in the above-mentioned laser additive manufacturing by calling the computer program stored in the memory.

[0065] A computer-readable storage medium is proposed, on which a rewritable computer program is stored;

[0066] When the computer program runs on a computer device, the computer device is made to execute the multi-field synchronous control method in the above-mentioned laser additive manufacturing.

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] The present application first establishes numerical models of multiple physical fields such as temperature field, flow field, phase field, diffusion field, stress and strain field, etc.; then by introducing machine learning algorithms, it mines the internal mapping rules between the data of each field and the finished product quality index, and constructs prediction equations such as defect, tissue, and stress evaluation; furthermore, according to the evaluation results, it feedback-optimizes each process parameter to achieve full-process intelligent closed-loop control. On the one hand, it fully considers the strong coupling effect between multiple fields and avoids the limitation of local optimization; on the other hand, through big data fitting, there is no need to manually compile complex evaluation models, so it has broad applicability; it improves production efficiency and manufacturing quality. Brief Description of the Drawings

[0069] Figure 1 It is the flow chart of the multi-field synchronous control method in laser additive manufacturing in Embodiment 1 of the present invention;

[0070] Figure 2 It is the module connection relationship diagram of the multi-field synchronous control system in laser additive manufacturing in Embodiment 2 of the present invention;

[0071] Figure 3 It is the schematic diagram of the structure of the electronic device in Embodiment 3 of the present invention;

[0072] Figure 4 It is the schematic diagram of the structure of the computer-readable storage medium in Embodiment 4 of the present invention. Detailed Embodiments

[0073] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0074] Embodiment 1

[0075] As Figure 1 shown, a multi-field synchronous control method in laser additive manufacturing includes the following steps:

[0076] Step 1: Construct an electromagnetic-thermal-mechanical coupling model, a mass diffusion model, and a stress-strain model;

[0077] Step 2: Collect the multi-field parameter set input into the current field;

[0078] Step 3: Input the multi-field parameter set into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool region; input the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model to solve the composition field distribution of the molten pool; input the multi-field parameter set and temperature field distribution into the stress-strain model to solve the stress field distribution and strain field distribution;

[0079] Step 4: Construct a pore evaluation equation, a crack evaluation equation, and a microstructure evaluation equation through the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution;

[0080] Step 5: Construct a quality optimization function based on the pore evaluation equation, crack evaluation equation, and microstructure evaluation equation, use a swarm intelligence algorithm to solve the quality optimization function, solve the quality optimization model, and obtain the multi-field control parameters.

[0081] Among them, the construction process of the electromagnetic-thermal-mechanical coupling model includes:

[0082] Construct an electromagnetic field model according to Maxwell's equations;

[0083] Construct a heat conduction model according to the heat conduction control equation;

[0084] Construct a fluid flow field model according to the energy, momentum, and mass conservation control equations;

[0085] The electromagnetic field model, heat conduction model, and fluid flow field model together constitute the electromagnetic-thermal-mechanical coupling model.

[0086] Specifically, the method for constructing the electromagnetic field model is:

[0087] Describe the electromagnetic field distribution with Maxwell's equations:

[0088] Ampere's Circuital Theorem: ×H = J (Ampere's Circuital Theorem);

[0089] Maxwell's equations: ×E = - ;

[0090] No magnetic charge: ·B = 0;

[0091] Current continuity: ·J = 0; where, represents the gradient or differential operator, H is the magnetic field strength vector, J is the electromagnetic field distribution, E is the electric field strength vector, B is the magnetic induction intensity vector; t is the time variable;

[0092] By giving boundary conditions, such as incident laser power, frequency, polarization, etc., and setting the electromagnetic parameters of the material, the electromagnetic field distribution J is calculated;

[0093] Specifically, the method for constructing the heat conduction model is:

[0094] Describing the heat conduction model with the heat conduction control equation: ρ×Cp×( ) = ·(k T) + Q; where, Q is the energy coupling term, ρ is the density, Cp is the specific heat, k is the thermal conductivity, and the energy coupling term mainly includes:

[0095] Q = Q_res + Q_ltd; Q_res is the Joule heat source, Q_res = J·E; Q_ltd is the latent heat phase change term;

[0096] By giving boundary conditions, such as heat radiation, and setting the heat conduction parameters of the material, the temperature field distribution is calculated;

[0097] The method for constructing the flow field model is:

[0098] Constructing the flow field model with the energy, momentum, and mass conservation control equations:

[0099] + ·(ρ×u) = 0;

[0100] ρ×( + u· u) = - p + ·τ + f_v;

[0101] ρ×Cp×( + u· T) = - ·q + Q; where u is the velocity field distribution, p is the pressure, τ is the viscous stress tensor, q is the heat flux vector; f_v is the body force per unit volume;

[0102] By giving boundary conditions such as inlet flow velocity, pressure, etc., and setting material flow parameters such as volume expansion coefficient, dynamic viscosity coefficient, etc., the flow field distribution is calculated.

[0103] Furthermore, the construction method of the mass diffusion model is as follows:

[0104] Use the diffusion equation to describe the mass diffusion process;

[0105] Given initial conditions, convective boundary conditions, and convective / diffusive boundary conditions;

[0106] Introduce phase change conditions;

[0107] Based on the diffusion equation and phase change conditions, establish a mass control equation set, and use the finite element or finite difference method to solve the mass control equation set, and iteratively update the temperature field and velocity field until convergence to obtain the converged molten pool composition field distribution.

[0108] Specifically, the diffusion equation is:

[0109] = u· C = ·(D(T)× C) + Q_m;

[0110] Where C is the molten pool composition field distribution, which is a function of time t and spatial position (x, y, z), u is the velocity field distribution, D(T) is the diffusion coefficient, which is a function related to temperature or composition, Q_m is the source term such as phase change precipitation, etc., n represents the normal vector;

[0111] The initial condition is the given initial molten pool composition field distribution C0, and the convective boundary condition is D(T)×( )= h_m×(C - C_inf), h_m is the mass transfer coefficient, C_inf is the far-field concentration; the convective / diffusive boundary condition is D(T)_l×( ) = D(T)_s( ) = J_m, J_m is the mass flow through the interface; where l represents the liquid phase, s represents the solid phase; D(T)_l represents the diffusion coefficient in the liquid phase, D(T)_s represents the diffusion coefficient in the solid phase, represents the molten pool composition field distribution in the liquid phase, s represents the molten pool composition field distribution in the solid phase;

[0112] The phase change condition is: (C_l - ) = (1 - )×ΔC_0×exp( ), where is the equilibrium distribution coefficient, thermodynamic data, ΔC_0 is the maximum non-equilibrium concentration, v is the solid-liquid interface growth rate, and ΔV_d is the diffusion potential.

[0113] Furthermore, the stress-strain model is constructed as follows:

[0114] Construct a generalized deflection constitutive equation and introduce the strain-displacement relationship condition and the stress equilibrium equation as constraint conditions to form a stress-strain model.

[0115] Among them, the generalized deflection constitutive equation is: σ = D: (ε - εth - εp - εφ), where: σ is the Cauchy stress tensor, D is the stiffness tensor, reflecting the elastic properties of the material, ε is the total strain tensor, εth is the thermal strain tensor, εp is the plastic strain tensor, and εφ is the phase change strain tensor; among them, εth = α×(T - T0), where α is the thermal expansion coefficient, T is the temperature field distribution, and T0 is the reference temperature; dεp = dλ× , where f is the yield criterion function, depending on the stress state, λ is the plastic coefficient, and plastic flow occurs when λ>0; among them, εφ = βφ×(φ - φ0), βφ is the phase change expansion coefficient, φ is the phase change degree field obtained from the diffusion model, and φ0 is the reference phase change degree; d is the derivative symbol, is the partial derivative symbol;

[0116] The strain-displacement relationship condition is: ε = ( u + uT) / 2; The stress equilibrium equation: ·σ + f = 0, where f is the body force.

[0117] Furthermore, the method for collecting the multi-field parameter set input into the current field is as follows:

[0118] Pre-collect the laser parameters, workpiece material parameters, environmental parameters, and process parameters required for the electromagnetic-thermal-mechanical coupling model to form an electromagnetic parameter set; specifically, the laser parameters include incident laser power, wavelength, frequency, polarization state, spot size, etc., the workpiece material parameters include density, specific heat, thermal conductivity, absorptivity, conductivity, permeability, etc., the environmental parameters include heat transfer coefficient, radiation emissivity, etc., and the process parameters include scanning speed, power factor, shielding gas flow rate, initial conditions of the molten pool, etc.

[0119] Pre-collect the component and phase diagram data, diffusion coefficients, partition coefficients, and flow field and temperature field distributions required for the mass diffusion model to form a diffusion parameter set; among them, the component and phase diagram data include the initial components of each element of the raw material, phase diagram parameters, etc., the diffusion coefficients include the diffusion coefficients of each element of the raw material, effective diffusion coefficients, the partition coefficients include the solid-liquid interface partition coefficient, non-equilibrium partition coefficient, etc., and the flow field and temperature field distributions are field quantities obtained from the electromagnetic-thermal-mechanical coupling model;

[0120] Pre-collect the mechanical property parameters, thermophysical parameters, boundary and loading conditions required for the stress-strain model to form a stress-strain parameter set; the mechanical property parameters include elastic modulus, Poisson's ratio, yield strength, plastic parameters, etc., the thermophysical parameters include linear expansion coefficient, volume change caused by phase change, etc., and the boundary and loading conditions include several preset constraints, preloading, etc.;

[0121] The electromagnetic parameter set, the diffusion parameter set, and the stress-strain parameter set form a multi-field parameter set.

[0122] Furthermore, the method of inputting the multi-field parameter set into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool area is as follows:

[0123] Input the laser parameters and workpiece material parameters in the electromagnetic parameter set of the multi-field parameter set into the electromagnetic field model, and use the finite element or boundary element method to solve the Maxwell equations to obtain the electromagnetic field distribution J of the current density in the molten pool area;

[0124] Input the environmental parameters, workpiece material parameters, and electromagnetic field distribution J in the electromagnetic parameter set of the multi-field parameter set into the heat conduction model, and use the finite element or boundary element method to solve the heat conduction control equations to obtain the temperature field distribution T in the molten pool area;

[0125] Input the process parameters in the electromagnetic parameter set of the multi-field parameter set and the temperature field distribution T into the flow field model, and use the finite element or boundary element method to solve the energy, momentum, and mass conservation control equations to obtain the velocity field distribution u in the molten pool area.

[0126] It can be understood that the temperature field distribution, velocity field distribution, and electromagnetic field distribution are mutually coupled and correlated, so they need to be considered comprehensively as the entire electromagnetic-thermal-mechanical coupling model.

[0127] Furthermore, the method of inputting the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model to solve the molten pool composition field distribution is as follows:

[0128] Input the diffusion parameter set in the multi-field parameter set and the temperature field distribution T into the mass diffusion model, and use the finite element or boundary element method to solve the mass control equations to obtain the molten pool composition field distribution C within the molten pool region.

[0129] Furthermore, the method of inputting the multi-field parameter set and the temperature field distribution into the stress-strain model to solve the stress field distribution and the strain field distribution is as follows:

[0130] By inputting the temperature field distribution and the stress-strain parameter set in the multi-field parameter set into the stress-strain model, and using the finite element or finite difference method to solve the stress-strain equations, the stress field distribution and the strain field distribution are obtained.

[0131] Furthermore, the method of constructing the pore evaluation equation, the crack evaluation equation, and the microstructure evaluation equation through the temperature field distribution, the velocity field distribution, the electromagnetic field distribution, the composition field distribution, the stress field distribution, and the strain field distribution is as follows:

[0132] The construction of the pore evaluation equation includes:

[0133] Collect the temperature field distribution, the velocity field distribution, and the electromagnetic field distribution of the sample workpieces as the first input features, and correspondingly mark whether each workpiece has pore defects as the first label data to construct the first training data set. Use an artificial neural network to fit the mapping relationship between the first input features and the first label in the first training data set as the pore evaluation equation;

[0134] The construction of the crack evaluation equation includes:

[0135] Collect the temperature field distribution, the stress field distribution, and the strain field distribution data of the workpieces with cracks and without cracks as the second input features, and correspondingly mark the number of cracks of each workpiece as the second label data to construct the second training data set. Use an artificial neural network to fit the mapping relationship between the second input features and the second label in the second training data set as the crack evaluation equation;

[0136] The construction of the microstructure evaluation equation includes:

[0137] Collect the temperature field distribution and the molten pool composition field distribution data of the sample workpieces as the third input features, collect the corresponding microstructure images or data, and label the grain size as the third label data to construct the third training data set. Use an artificial neural network to fit the mapping relationship between the third input features and the third label in the third training data set as the microstructure evaluation equation.

[0138] Through the above big data fitting method, the internal laws between the input field quantities and defects or microstructures are explored, replacing the establishment of complex analytical equations based on theoretical derivations, thus achieving the goal of constructing various evaluation equations in the case of high dimensions and high complexity.

[0139] Further, the method of constructing a quality optimization function based on the pore evaluation equation, crack evaluation equation, and tissue evaluation equation and solving the quality optimization function using a swarm intelligence algorithm is as follows:

[0140] Perform weighted summation on the pore evaluation equation, crack evaluation equation, and tissue evaluation equation to obtain a quality optimization function.

[0141] It should be noted that the weights of the various evaluation equations in the quality optimization function are set according to actual needs. For example, if more importance is attached to the influence of bubbles, the weight of the pore evaluation equation is set larger.

[0142] The method of solving the quality optimization model to obtain multi-field control parameters is as follows:

[0143] Through the use of the gradient descent algorithm for backpropagation of the error of the quality optimization function and continuous iteration, obtain the proposed distribution values of the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution, and then based on the proposed distribution values of each field distribution and the corresponding models, inversely deduce the proposed boundary conditions of the various inputs that need to be controlled as multi-field control parameters.

[0144] For example, when obtaining the proposed distribution value of the temperature field distribution, that is, when the temperature field distribution T is used as a constant, inversely deduce the corresponding boundary condition of thermal radiation through the heat conduction model, and then control the thermal radiation in the molten pool to reach the inversely deduced value of thermal radiation, that is, complete the control of the thermal field distribution. Similarly, inversely deduce the boundary conditions of other fields to achieve the degree of controlling the proposed distribution values of multiple field distributions to the proposed level, so as to maximize the quality optimization function, that is, achieve the purpose of optimizing the quality of the product as much as possible.

[0145] Embodiment 2

[0146] As Figure 2 shown, the multi-field synchronous control system in laser additive manufacturing includes a multi-field model construction module, a parameter collection module, a distribution solution module, and a condition control module; among them, each module is connected electrically;

[0147] The multi-field model construction module constructs an electromagnetic-thermal-mechanical coupling model, a mass diffusion model, and a stress-strain model, and sends the electromagnetic-thermal-mechanical coupling model, mass diffusion model, and stress-strain model to the distribution solution module;

[0148] The parameter collection module collects the multi-field parameter set input into the current field and sends the multi-field parameter set to the distribution solution module;

[0149] The distribution solving module inputs a multi-field parameter set into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool region; inputs the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model to solve the composition field distribution of the molten pool; inputs the multi-field parameter set and temperature field distribution into the stress-strain model to solve the stress field distribution and strain field distribution, and sends the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution to the condition control module;

[0150] The condition control module constructs a quality optimization function based on the pore evaluation equation, crack evaluation equation, and microstructure evaluation equation, uses a swarm intelligence algorithm to solve the quality optimization function, solves the quality optimization model, and obtains multi-field control parameters.

[0151] Embodiment 3

[0152] Figure 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As Figure 3 shown, the present application also provides an electronic device 100. The electronic device 100 may include one or more processors and one or more memories. Among them, computer-readable code is stored in the memory, and when the computer-readable code is run by one or more processors, it can execute the multi-field synchronous control method in laser additive manufacturing as described above.

[0153] The method or device according to the embodiment of the present application can also be implemented by means of Figure 3 the architecture of the electronic device shown. As Figure 3 shown, the electronic device 100 may include a bus 101, one or more CPUs 102, a ROM 103, a RAM 104, a communication port 105 connected to the network, an input / output component 106, a hard disk 107, etc. The storage device in the electronic device 100, such as the ROM 103 or the hard disk 107, can store the multi-field synchronous control method in laser additive manufacturing provided by the present application.

[0154] Furthermore, the electronic device 100 may further include a user interface 108. Of course, Figure 3 the architecture shown is only exemplary, and when implementing different devices, one or more components shown in the electronic device may be omitted according to actual needs. Figure 3

[0155] Embodiment 4

[0156] Figure 4 is a schematic structural diagram of a computer-readable storage medium provided by an embodiment of the present application. As Figure 4 ​As shown, there is a computer-readable storage medium 200 provided by this application. Computer-readable instructions are stored on the computer-readable storage medium 200. When the computer-readable instructions are run by a processor, the multi-field synchronous control method in laser additive manufacturing according to the embodiments of this application described with reference to the above drawings can be executed. The computer-readable storage medium 200 includes but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory, etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disks, flash memory, etc.

[0157] In addition, according to the embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be run by a processor to execute instructions corresponding to the method steps provided by this application. When this computer program is executed by a central processing unit (CPU), the above functions defined in the method of this application are executed.

[0158] The methods, apparatuses, and devices of this application can be implemented in many ways. For example, the methods, apparatuses, and devices of this application can be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is only for illustration. The steps of the method of this application are not limited to the specific order described above, unless otherwise specifically stated. In addition, in some embodiments, this application can also be implemented as a program recorded in a recording medium, and these programs include machine-readable instructions for implementing the method according to this application. Therefore, this application also covers the recording medium storing the program for executing the method according to this application.

[0159] In addition, in the above technical solutions provided by the embodiments of this application, the parts that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive elaboration.

[0160] As described in the above specific embodiments, the purpose, technical solutions, and beneficial effects of the present invention are further described in detail. It should be understood that the above is only the specific embodiment of the present invention and is not used to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0161] The above preset parameters or preset thresholds are all set by those skilled in the art according to the actual situation or obtained through a large amount of data simulation.

[0162] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A multi-field synchronous control method in laser additive manufacturing, characterized in that It includes the following steps: Step 1: Construct an electromagnetic-thermal-mechanical coupling model, a mass diffusion model, and a stress-strain model; Step 2: Collect a set of multi-field parameters input into the current field; Step 3: Input the set of multi-field parameters into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool area; input the set of multi-field parameters, temperature field distribution, and velocity field distribution into the mass diffusion model to solve the composition field distribution of the molten pool; input the set of multi-field parameters and temperature field distribution into the stress-strain model to solve the stress field distribution and strain field distribution; Step 4: Construct a pore evaluation equation, a crack evaluation equation, and a microstructure evaluation equation based on the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution; Step 5: Construct a quality optimization function based on the pore evaluation equation, crack evaluation equation, and microstructure evaluation equation, use a swarm intelligence algorithm to solve the quality optimization function, solve the quality optimization model, and obtain a set of multi-field control parameters; The construction process of the electromagnetic-thermal-mechanical coupling model includes: Construct an electromagnetic field model based on Maxwell's equations; Construct a heat conduction model based on the heat conduction control equation; Construct a fluid flow field model based on the control equations of energy, momentum, and mass conservation; The electromagnetic field model, heat conduction model, and fluid flow field model together constitute the electromagnetic-thermal-mechanical coupling model; The construction method of the mass diffusion model is: Use the diffusion equation to describe the mass diffusion process; Specify the initial conditions, convective boundary conditions, and convective / diffusive boundary conditions; Introduce the phase change condition; Based on the diffusion equation and phase change condition, establish a set of mass control equations, and use the finite element or finite difference method to solve the set of mass control equations, and iteratively update the temperature field and velocity field until convergence to obtain the converged composition field distribution of the molten pool; The construction method of the stress-strain model is: Construct a generalized deflection constitutive equation, and introduce the strain-displacement relationship condition and stress equilibrium equation as constraint conditions to form the stress-strain model; The method of constructing the pore evaluation equation, crack evaluation equation, and microstructure evaluation equation is: The construction of the pore evaluation equation includes: Collect the temperature field distribution, velocity field distribution, and electromagnetic field distribution of the sample workpieces as the first input features, and correspondingly mark whether each workpiece has pore defects as the first label data, construct the first training dataset, and use 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 pore evaluation equation; The construction of the crack evaluation equation includes: Collect the temperature field distribution, stress field distribution, and strain field distribution data of workpieces with and without cracks as the second input features, and correspondingly mark the number of cracks of each workpiece as the second label data, construct the second training dataset, and use 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: Collect the temperature field distribution and molten pool composition field distribution data of the sample workpiece as the third input feature, collect the corresponding microstructure images or data, label the grain size as the third label data, construct the third training dataset, and use 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 tissue evaluation equation; The method for solving the quality optimization model to obtain the multi-field control parameters is as follows: By using the gradient descent algorithm to perform backpropagation of the error on the quality optimization function and continuously iterating, obtain the proposed distribution values of the temperature field distribution, velocity field distribution, electromagnetic field distribution, composition field distribution, stress field distribution, and strain field distribution, and then based on the proposed distribution values of each field distribution and the corresponding model, inversely deduce the proposed boundary conditions that need to be input for each item to be controlled as the multi-field control parameters.

2. The multi-field synchronous control method in laser additive manufacturing according to claim 1, characterized in that The method for constructing the electromagnetic field model is as follows: Describe the electromagnetic field distribution with Maxwell's equations; By giving the boundary conditions and setting the material electromagnetic parameters, calculate the electromagnetic field distribution J; The method for constructing the heat conduction model is as follows: Describe the heat conduction model with the heat conduction control equation; By giving the boundary conditions and setting the material heat conduction parameters, calculate the temperature field distribution; The method for constructing the flow field model is as follows: Construct the flow field model with the energy, momentum, and mass conservation control equations; By giving the boundary conditions and setting the material flow parameters, calculate the flow field distribution.

3. The multi-field synchronous control method in laser additive manufacturing according to claim 2, wherein The method for collecting the multi-field parameter set input into the current field is as follows: Pre-collect the laser parameters, workpiece material parameters, environmental parameters, and process parameters required by the electromagnetic-thermal-mechanical coupling model to form the electromagnetic parameter set; Pre-collect the composition and phase diagram data, diffusion coefficient, distribution coefficient, and flow field and temperature field distribution required by the mass diffusion model to form the diffusion parameter set; Pre-collect the mechanical property parameters, thermophysical parameters, boundary, and loading conditions required by the stress-strain model to form the stress-strain parameter set; The electromagnetic parameter set, the diffusion parameter set, and the stress-strain parameter set form the multi-field parameter set.

4. The multi-field synchronous control method in laser additive manufacturing according to claim 3, wherein, The method for inputting the multi-field parameter set into the electromagnetic-thermal-mechanical coupling model and solving the temperature field distribution, velocity field distribution, and electromagnetic field distribution in the molten pool area is as follows: Input the laser parameters and workpiece material parameters in the electromagnetic parameter set of the multi-field parameter set into the electromagnetic field model, and use the finite element or boundary element method to solve Maxwell's equations to obtain the electromagnetic field distribution J of the current density in the molten pool area; Input the environmental parameters, workpiece material parameters, and electromagnetic field distribution J in the electromagnetic parameter set of the multi-field parameter set into the heat conduction model, and use the finite element or boundary element method to solve the heat conduction control equations to obtain the temperature field distribution T in the molten pool area; Input the process parameters in the electromagnetic parameter set of the multi-field parameter set and the temperature field distribution T into the flow field model, and use the finite element or boundary element method to solve the energy, momentum, and mass conservation control equations to obtain the velocity field distribution u in the molten pool area.

5. The multi-field synchronous control method in laser additive manufacturing according to claim 4, wherein, The method for inputting the multi-field parameter set, temperature field distribution, and velocity field distribution into the mass diffusion model and solving the molten pool composition field distribution is as follows: Input the diffusion parameter set in the multi-field parameter set and the temperature field distribution T into the mass diffusion model, and use the finite element method or the boundary element method to solve the mass control equations to obtain the molten pool composition field distribution C in the molten pool area.

6. The multi-field synchronous control method in laser additive manufacturing according to claim 5, characterized in that The method of inputting the multi-field parameter set and the temperature field distribution into the stress-strain model and solving the stress field distribution and the strain field distribution is as follows: By inputting the temperature field distribution and the stress-strain parameter set in the multi-field parameter set into the stress-strain model, and using the finite element method or the finite difference method to solve the stress-strain equations, the stress field distribution and the strain field distribution are obtained.

7. A multi-field synchronous control system in laser additive manufacturing, which is used to implement the multi-field synchronous control method in laser additive manufacturing according to any one of claims 1-6, characterized in that, It includes a multi-field model construction module, a parameter collection module, a distribution solution module, and a condition control module; among them, each module is connected electrically. The multi-field model construction module constructs an electromagnetic-thermal-mechanical coupling model, a mass diffusion model, and a stress-strain model, and sends the electromagnetic-thermal-mechanical coupling model, the mass diffusion model, and the stress-strain model to the distribution solution module. The parameter collection module collects the multi-field parameter set input into the current field and sends the multi-field parameter set to the distribution solution module. The distribution solution module inputs the multi-field parameter set into the electromagnetic-thermal-mechanical coupling model to solve the temperature field distribution, the velocity field distribution, and the electromagnetic field distribution in the molten pool area; inputs the multi-field parameter set, the temperature field distribution, and the velocity field distribution into the mass diffusion model to solve the molten pool composition field distribution; inputs the multi-field parameter set and the temperature field distribution into the stress-strain model to solve the stress field distribution and the strain field distribution, and sends the temperature field distribution, the velocity field distribution, the electromagnetic field distribution, the composition field distribution, the stress field distribution, and the strain field distribution to the condition control module. The condition control module constructs a quality optimization function based on the pore evaluation equation, the crack evaluation equation, and the microstructure evaluation equation, uses the swarm intelligence algorithm to solve the quality optimization function, solves the quality optimization model, and obtains the multi-field control parameters.

8. An electronic device, characterized in that, It includes: A processor and a memory, where: The memory stores a computer program that can be called by the processor. The processor executes the multi-field synchronous control method in laser additive manufacturing according to any one of claims 1-7 in the background by calling the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that, It stores an erasable computer program on it. When the computer program runs on the computer device, the computer device executes the multi-field synchronous control method in laser additive manufacturing according to any one of claims 1-7 in the background.

Images