Titanium-aluminum dissimilar metal laser bias melt-brazing heat transfer-flow-component transportation operation and IMC formation high-precision visualization system

Abstract

The invention belongs to the technical field of welding, and discloses a titanium-aluminum dissimilar metal laser bias melt-brazing heat transfer-flow-component transmission operation and IMC (in-mold control) formation high-precision visualization system. Through multi-physics field coupling and dynamic modeling, heat transfer, flow, component diffusion and reaction kinetics are combined; and the forming conditions, the content, the distribution and the morphology of TiAl3, TiAl and Ti3Al in the Ti / Al laser welding process are comprehensively predicted. Gibbs free energy criterion is utilized to accurately judge generation conditions, the dynamic mass fraction and spatial distribution of a generated phase are calculated based on a reaction rate and a diffusion equation, and a cooling rate is utilized to drive a grain size formula to predict morphological characteristics. Compared with a traditional steady-state hypothesis and empirical formula method, the method has the advantages that the prediction precision is remarkably improved, unbalanced behaviors and local non-uniformity in the welding process can be captured, and scientific basis and actual value are provided for optimizing the welding process and improving the joint performance.

Description

Technical Field

[0001] This invention belongs to the field of welding technology, and particularly relates to a high-precision visualization system for heat transfer, flow, and component transport behavior in titanium-aluminum dissimilar metal laser bias brazing welding and the formation of IMC. Background Technology

[0002] Welding is a crucial joining process in the fabrication of Ti / Al composite structures. However, the weld interface is prone to forming a large amount of brittle intermetallic compounds (IMCs), making it difficult for the joint's mechanical properties to meet service requirements. Ti / Al laser-biased brazing utilizes the melting point difference between Ti and Al, biasing the laser beam towards Al to melt the Al side while the Ti side remains slightly melted or solid. By controlling the molten state of Ti / Al, excessive mixing of the two liquid metal phases avoids the formation of a large amount of brittle IMCs. This method provides an effective means of suppressing brittle IMCs in Ti / Al welding and is now widely used in actual production. However, the process parameters of laser-biased brazing are mutually constrained. Too small a bias distance leads to unsatisfactory IMC suppression, while too large a bias distance can prevent complete Ti / Al joint connection. Extensive process experiments driven by manual experience are necessary to determine a good process that simultaneously achieves global metallurgical bonding and brittle IMC suppression, resulting in high labor and material costs that cannot meet actual production needs. Numerical simulation technology can provide a deeper understanding of the macro- and micro-forming processes of Ti / Al laser bias welding, making it an important theoretical tool for process optimization and mechanism research.

[0003] During Ti / Al laser bias brazing, the formation of IMCs is determined by both kinetics and thermodynamics. For example, whether IMCs such as TiAl3, TiAl, and Ti3Al can exist stably is determined by the free energy change, Ti-Al phase diagram, and reaction temperature in the thermodynamic realm. The distribution, content, and morphology of IMCs are determined by the Ti / Al diffusion behavior, molten pool flow, and cooling rate in the kinetic realm.

[0004] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0005] (1) It is difficult to handle the independent transport process of Ti and Al phases, resulting in inaccurate prediction of the temperature field, velocity field and pore morphology fluctuation of the molten pool;

[0006] (2) It is difficult to handle the mutual diffusion process of Ti and Al phases in the molten pool, and it is impossible to obtain a high-precision molten pool concentration field;

[0007] (3) The generation of multiple IMCs (such as TiAl3, TiAl, Ti3Al, etc.) is not considered at the same time, so it is impossible to predict the distribution and content of multiple IMCs in the weld joint after welding. Summary of the Invention

[0008] To address the problems existing in the prior art, this invention provides a high-precision visualization system for the heat transfer, flow, and component transport behavior of titanium-aluminum dissimilar metal laser bias brazing and IMC formation.

[0009] This invention is achieved as follows: a high-precision visualization system for the heat transfer, flow, and component transport behavior of titanium-aluminum dissimilar metal laser bias brazing and IMC formation includes:

[0010] Heat source model module, diffusion model module, driving force solution module, welding boundary condition solution module, prediction module;

[0011] The heat source model module, connected to the driving force solution module, is used for laser heat source models with differences in Ti and Al.

[0012] The diffusion model module, connected to the driving force solution module, is used for Ti / Al laser bias brazing Ti-Al mixed flow and component diffusion model;

[0013] The driving force solution module is connected to the heat source model module, diffusion model module, welding boundary condition solution module, and prediction module, and is used to solve the driving force for Ti / Al laser bias brazing welding.

[0014] The welding boundary condition solution module is connected to the driving force solution module and is used for the boundary conditions and solution algorithm of Ti / Al laser bias brazing welding.

[0015] The prediction module, connected to the driving force solution module, is used for IMC prediction based on the heat transfer-flow-component transport model of Ti / Al laser bias brazing.

[0016] Furthermore, the heat source model module:

[0017] (1) Differences in laser energy absorption between Ti and Al

[0018] The laser absorption rates of Ti and Al in solid and liquid states and their dependence on temperature are significantly different:

[0019] The effective absorption rate of laser light is determined by the absorption rates of Ti and Al and their compositional distribution:

[0020]

[0021] The absorption rates of Ti and Al as a function of temperature can be expressed as follows:

[0022]

[0023]

[0024] Parameter notes: Ti and Al at initial temperature ( Absorption rate (dimensionless) at ) ); Temperature sensitivity coefficient of absorptivity (unit: ); Ambient temperature (unit: );

[0025] η eff (f Ti ,f Al (T): Effective laser absorption rate, which is related to the mass fraction f of titanium (Ti). Ti The mass fraction of aluminum (Al), fAl, is a function of temperature T, representing the effective absorption ratio of laser energy under given titanium and aluminum ratios and temperature conditions.

[0026] f Ti and f Al : These represent the mass fractions of titanium (Ti) and aluminum (Al), respectively, used to reflect the proportion of titanium and aluminum in the mixed material.

[0027] T: The current temperature of the environment or materials.

[0028] η Ti (T) and η Al (T): These are the absorption rates of titanium (Ti) and aluminum (Al) at temperature T, respectively, reflecting the ability of titanium and aluminum to absorb laser energy at a specific temperature.

[0029] and : The absorption rate (dimensionless) of titanium (Ti) and aluminum (Al) at the initial temperature T0, that is, the ratio of the absorption of laser energy by titanium and aluminum at the reference temperature.

[0030] α Ti and α Al : Temperature sensitivity coefficient of absorptivity for titanium (Ti) and aluminum (Al), expressed in 1 / K. This parameter describes how sensitive the absorptivity is to temperature changes, and its value reflects how quickly the absorptivity responds to temperature changes.

[0031] T0: Ambient temperature, in K, serves as a reference temperature for temperature changes.

[0032] (2) Gaussian heat source model and calculation of dynamic absorption rates of Ti and Al

[0033] 1) Power density distribution: The effective power density of the laser on the material surface is:

[0034]

[0035] Incident laser power density (unit: W / m²) 2 ), defined by the Gaussian distribution:

[0036]

[0037] The mixed absorption rate of Ti and Al depends on their mass fraction. , and temperature ;

[0038] 2) Calculation of dynamic absorption rate: Mixed absorption rate expression:

[0039]

[0040] The absorption rates of Ti and Al are respectively:

[0041] Solid State:

[0042]

[0043]

[0044] Liquid:

[0045]

[0046]

[0047] The melting point temperature of the material:

[0048] The melting point of Ti (approximately 1941 K);

[0049] The melting point of Al (approximately 933 K).

[0050] 3) Spatial absorption distribution:

[0051] The energy distribution of laser light within a material is described by Beer-Lambert's law:

[0052]

[0053] Absorbed energy density inside the material (unit: W / m³) 3 );

[0054] Effective power density on the surface;

[0055] The absorption coefficient of the material (unit: 1 / m) is calculated by weighting the absorption coefficients of Ti and Al.

[0056]

[0057] (3) Ray tracing and interface tracing methods

[0058] 1) Laser Reflection and Refraction at Interfaces: When laser light propagates from one medium to another, some of its energy is reflected, and some is transmitted into the next medium. The reflectivity and refractive index are given by the Fresnel formula:

[0059] Reflectivity:

[0060]

[0061] Refractive index:

[0062]

[0063] The reflectivity varies with the dynamic changes (solid-liquid, liquid-gas) of the Ti and Al interface, and the relevant parameters include:

[0064] Refractive index of solid Ti and Al ;

[0065] Refractive index of liquid Ti and Al ;

[0066] 2) Recursive tracing of the laser path

[0067] The path of light propagation at the interface is calculated using Snell's Law:

[0068]

[0069] For complex interfaces (such as pinholes or molten pool surfaces), the ray path needs to be recursively traced, and the incident angle dynamically adjusted. and angle of refraction ;

[0070] 3) Interface tracking

[0071] By volume fraction Indicates the interface location:

[0072]

[0073] (4) Coupled solution of the heat source model for Ti / Al laser biased brazing

[0074] Combining ray tracing, interface tracing, and the absorption differences between Ti and AI, the final expression for laser power density is:

[0075]

[0076] : Interface shape (updated in real time by the interface tracking method); Incident laser power density; Dynamic absorption rate; Absorption coefficient.

[0077] Furthermore, the diffusion model module:

[0078] (1) Mass conservation equation for Ti / Al laser bias brazing

[0079] Overall mass conservation equation (for multi-component fluids):

[0080]

[0081] Parameter description:

[0082] The effective density of the molten pool (unit: kg / m3) is determined by the mass fractions of Ti and Al.

[0083]

[0084] v: Velocity of fluid in the molten pool (unit: m / s);

[0085] During the laser brazing process of Ti / Al dissimilar alloys, Al and Ti will melt and evaporate, resulting in efflorescence in the molten pool;

[0086] (2) Momentum conservation equation for Ti / Al laser bias brazing

[0087] The momentum conservation equation for molten pool flow is:

[0088]

[0089] Parameter description: Pressure field (unit: Pa); Effective viscosity of the molten pool (unit: Pa) s), calculated by viscosity weighting of Ti and Al:

[0090]

[0091] Physical force, including buoyancy and surface tension gradient driving force:

[0092]

[0093] The first term is the buoyancy driving force, where: Gravitational acceleration (unit: m / s²) ); Coefficient of thermal expansion (unit: 1 / K); : Ambient temperature (unit: K); The second term is the surface tension gradient driving force (Marangoni effect), where: surface tension With temperature and components change;

[0094] (3) Energy conservation equation for Ti / Al laser bias brazing

[0095]

[0096] ρ: Density (kg / m³) 3 ); cp: specific heat capacity (J / (kg·K)); T: temperature (K); v: fluid velocity (m / s); keff: effective thermal conductivity (W / (m·K)); qlaser: laser heat input (W / m 3 );

[0097] (4) Diffusion equation of Ti / Al laser bias brazing

[0098] The mass fractions of Ti and Al in the molten pool vary with time and space, and the component diffusion equations are as follows:

[0099]

[0100] Parameter description: Mass fraction of Ti (dimensionless, between 0 and 1); : Fluid velocity within the molten pool (unit: m / s); Concentration diffusion coefficients of Ti and Al (unit: m) 2 ( / s), calculated using the Arrhenius formula:

[0101]

[0102] Pre-diffusion coefficient (unit: m) 2 / s); Activation energy for diffusion (unit: J / mol); Gas constant (unit: 8.314 J / mol·K); Temperature (unit: K); The Soret effect (thermal diffusion) term describes the influence of the temperature gradient on diffusion.

[0103]

[0104] Soret coefficient (unit: 1 / K);

[0105] (5) Coupling effect of thermal diffusion and concentration gradient in Ti / Al laser bias brazing

[0106] Within the molten pool, the coupling effect of concentration and temperature gradients further influences component diffusion behavior; the effective diffusion coefficient... Revised to:

[0107] .

[0108] Furthermore, the driving force solving module:

[0109] (1) Surface tension gradient driving force (Marangoni effect) in Ti / Al laser bias brazing

[0110] The surface tension gradient is one of the main driving forces of molten pool flow, originating from changes in temperature and composition gradients; the driving force caused by the surface tension gradient is:

[0111]

[0112] 1) Contribution of temperature gradient: Rate of change of surface tension with temperature (unit: N / m·K):

[0113]

[0114] in, Negative values ​​(surface tension typically decreases with increasing temperature); - Temperature gradient It can be obtained through the energy conservation equation, and the temperature gradient is usually the largest near the edge of the molten pool.

[0115] 2) Contribution of component gradient: Rate of change of surface tension with Ti concentration (unit: N / m):

[0116]

[0117] in, Reflects the changes in surface tension caused by the components, concentration gradient The size is usually larger at the edge of the molten pool or at the interface of different materials; Comprehensive formula:

[0118]

[0119] (2) Buoyancy driving force of Ti / Al laser bias brazing

[0120] The buoyancy driving force is caused by the density change within the molten pool; the density change of the molten pool mainly comes from temperature differences and composition differences.

[0121]

[0122] Parameter description: Molten pool density (unit: kg / m³) 3 The mass fraction of the multi-component mixture is determined by its mass fraction.

[0123]

[0124] Gravitational acceleration (unit: m / s²) 2 ); Coefficient of thermal expansion (unit: 1 / K); Local temperature of the molten pool (unit: K); Ambient temperature (unit: K); Buoyancy driving force mainly acts on the inside of the molten pool, usually pushing upward in the high-temperature zone (center of the molten pool) and downward in the low-temperature zone;

[0125] (3) Steam pressure and recoil pressure of Ti / Al laser bias brazing

[0126] vapor pressure It is a function of temperature, described by the Clausius-Clapeyron equation:

[0127]

[0128] Vapor pressure at reference temperature (unit: Pa); Latent heat of vaporization (unit: J / mol); Gas constant (8.314 J / mol·K);

[0129] When a high-energy laser strikes a material surface, a localized high-pressure gas (plasma or vapor) is formed. This high-pressure gas exerts a shock force on the surface of the molten pool; the shock wave driving force can be expressed as:

[0130]

[0131] Local gas pressure induced by laser (unit: Pa); Surface area affected by laser (unit: m2).

[0132] Furthermore, the welding boundary condition solving module:

[0133] Boundary conditions and solution algorithm for Ti / Al laser bias brazing

[0134] The computational domain of this model is mainly divided into three parts: the Al, Ti, and air domains. The upper surface of the air domain and the lower surface of the Ti / Al domain are then designated as pressure outlets, while the remaining interfaces are all designated as walls (the upper and lower surfaces are the main surfaces in contact with air during welding). It is worth noting that the model incorporates convective heat transfer between different interfaces. Radiative heat transfer Evaporative heat transfer These are expressed by the following formulas:

[0135]

[0136]

[0137] In the formula, The convective heat transfer coefficient, For ambient temperature, This is the Stefan–Boltzmann constant. For emissivity, For interface density, For the specific heat capacity of the interface, It is the specific heat capacity of the metal. Specific heat capacity of the gas;

[0138] Finally, the present invention employs the following solution method to improve the accuracy and efficiency of the model:

[0139] For spatial discretization, the finite volume method (FVM) combined with the high-order WENO scheme is used to accurately handle the drastic changes in the interface and temperature gradient. For temporal discretization, implicit time integration methods (such as the Backward Euler method) are used to improve stability, and the fluid velocity and pressure are solved step by step using the PISO algorithm. In terms of interface tracking, the VOF method and level set method are combined to accurately capture the dynamic changes of the Ti / Al interface. During the calculation process, GPU parallel computing and the OpenFOAM framework are used to significantly improve computational efficiency.

[0140] Furthermore, the prediction module:

[0141] (1) Prediction of the formation types of multiple IMCs

[0142] The thermodynamic criteria for the formation of TiAl3, TiAl and Ti3Al can be determined by calculating the Gibbs free energy change (ΔG): ΔG < 0: the reaction proceeds spontaneously and the formed phase exists stably;

[0143]

[0144] The enthalpy of reaction (ΔH) and entropy (ΔS) can be obtained from materials thermodynamic databases (such as Thermo-Calc) or experimentally determined;

[0145] Based on the molten pool temperature field T(x,y,z,t), determine whether each region satisfies the formation condition ΔG<0 and predict the generated phase;

[0146] (2) Prediction of the content of multiple IMC formations

[0147] The reaction rate for phase formation is described by the Arrhenius equation:

[0148]

[0149] : Formation phase The reaction rate (unit: mol / m3.s)

[0150]

[0151]

[0152] :Reaction frequency factor (unit: ;

[0153] Activation energy of reaction (unit: J / mol);

[0154] Gas constant (8.314 J / moldotpK);

[0155] Local temperature (unit: K);

[0156] Local mass fractions of Ti and Al;

[0157] The reaction order is determined experimentally.

[0158] mass fraction of the generated phase It can be calculated by integrating the reaction rate:

[0159]

[0160] Local density, varying with temperature and composition:

[0161]

[0162] Calculate temperature at each grid point and components Calculate the reaction rate constant. and reaction rate Time integration yields the mass fraction of each generated phase. ;

[0163] (3) Prediction of the distribution of multiple IMCs

[0164] The diffusion of Ti and Al components within the molten pool is described by Fick's second law:

[0165]

[0166] Local mass fraction of Ti; : Fluid velocity within the molten pool (unit: m / s); The diffusion coefficients of Ti and Al change with temperature:

[0167]

[0168] fluid velocity Calculated by the momentum equation, it significantly affects the component distribution; buoyancy and surface tension: buoyancy and the Marangoni effect within the molten pool affect molten pool flow: buoyancy: Surface tension gradient:

[0169] Solving the component diffusion equation yields... and The distribution of the three generated phases in the weld region is predicted by combining the reaction rate equation. ;

[0170] (4) Prediction of the morphology of multiple IMC formations

[0171] Cooling rate The grain size and morphology are determined by rapid cooling to produce fine TiAl3 phase; slow cooling produces coarse Ti3Al and TiAl phases; cooling rate calculation:

[0172]

[0173] The time derivative is calculated using the energy conservation equation; grain size With cooling rate Relationship:

[0174]

[0175] , : Constants determined by experimental fitting;

[0176] The cooling rate was calculated using the preceding numerical simulation model. By combining the grain size formula, the grain size distribution of each generated phase is predicted. .

[0177] Another objective of this invention is to provide a high-precision visualization method for heat transfer, flow, and component transport behavior in titanium-aluminum dissimilar metal laser bias brazing and IMC formation, including:

[0178] Step 1, consider the laser heat source model with differences in Ti and Al;

[0179] Step 2, Ti / Al laser bias brazing Ti-Al mixed flow and component diffusion model;

[0180] Step 3, Solving for the driving force of Ti / Al laser bias brazing;

[0181] Step 4, Boundary conditions and solution algorithm for Ti / Al laser bias brazing;

[0182] Step 5: IMC prediction based on the heat transfer-flow-component transport model of Ti / Al laser bias brazing.

[0183] Another object of the present invention is to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, the computer program being executed by the processor causing the processor to perform the steps of the heat transfer-flow-component transport behavior of titanium-aluminum dissimilar metal laser bias brazing and the high-precision visualization method for IMC formation.

[0184] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the titanium-aluminum dissimilar metal laser bias brazing heat transfer-flow-component transport behavior and the high-precision visualization method for IMC formation.

[0185] Another objective of this invention is to provide an information data processing terminal, which is used to realize the heat transfer-flow-component transport behavior of the titanium-aluminum dissimilar metal laser bias brazing and the high-precision visualization system for IMC formation.

[0186] Based on the above technical solutions and the technical problems solved, please analyze the advantages and positive effects of the technical solution to be protected by this invention from the following aspects:

[0187] This invention develops a numerical simulation model for laser biased brazing of titanium and aluminum dissimilar metals, aiming to achieve high-precision visualization of welding heat transfer, flow, component transport behavior and post-weld IMC formation.

[0188] Traditional laser welding models typically employ a fixed absorptivity assumption, neglecting the dynamic characteristics of material (such as Ti and Al) absorptivity varying with temperature, phase state, and laser wavelength. In contrast, this invention introduces a dynamic absorptivity formula, incorporating the mass fraction of Ti and Al, temperature variations, and differences in their physical properties to accurately calculate the laser heat input. This method reflects the energy absorption characteristics and dynamic changes of different materials, improving the accuracy of laser energy distribution in multiphase flow fields and providing a more reliable description of the heat source for welding complex materials.

[0189] Traditional models mostly consider only the fixed physical properties of a single component or a simple binary fluid, neglecting the dynamic influence of component concentration on density, viscosity, and diffusion coefficient in a multi-component molten pool. The model of this invention dynamically calculates the physical properties of the Ti and Al molten pools, combining component diffusion and temperature fields to realistically reflect the complex flow and diffusion behavior of the mixed molten pool. This method is suitable for describing component migration and interfacial behavior, and has significant practical implications in the field of multi-component welding.

[0190] Traditional laser welding models often employ single-interface tracking methods (such as VOF or level set methods), which struggle to simultaneously capture the accuracy of interface shape and topological changes. This invention combines the advantages of both VOF and level set methods, accurately describing interface shape while handling complex interface topological changes. This approach is more stable and efficient in describing the dynamic behavior of multiphase interfaces, making it particularly suitable for describing the keyhole dynamics and molten pool evolution in Ti / Al welding.

[0191] Traditional models typically simplify laser heat input to a constant heat source, neglecting the dynamic balance between laser input and heat loss from the molten pool surface. This invention introduces an energy conservation condition, comprehensively considering the coupling of laser heat input, heat conduction, convection, and radiation. The improved heat input model more realistically reflects the dynamic distribution of heat in the molten pool, improving the accuracy of temperature field simulation and representing a significant innovation in high-power laser welding research.

[0192] Traditional models only consider the Marangoni effect, which is driven by temperature gradients, neglecting the influence of compositional gradients on surface tension in multi-component molten pools. This invention refines the surface tension driving force into the combined effect of temperature and compositional gradients, comprehensively reflecting the surface flow mechanism in Ti / Al mixed molten pools. This improvement enables a more accurate description of the flow and stability of the molten pool surface, which is of great significance for the prediction and control of welding quality.

[0193] Traditional models typically employ simple numerical solution methods (such as the finite difference method or the finite volume method), which are prone to numerical errors in regions with drastic interface changes or large temperature gradients. The model in this invention employs a high-order WENO scheme, implicit time integration, and the PISO algorithm, and significantly improves computational efficiency and stability through GPU parallel computing. This cutting-edge numerical solution framework enables the model to efficiently handle complex multiphase flow field calculations, meeting the requirements of high-precision laser welding simulation.

[0194] This method comprehensively predicts the formation conditions, content, distribution, and morphology of TiAl3, TiAl, and Ti3Al during Ti / Al laser welding by combining multiphysics coupling and dynamic modeling with heat transfer, flow, component diffusion, and reaction kinetics. It accurately determines the formation conditions using the Gibbs free energy criterion, calculates the dynamic mass fraction and spatial distribution of the formed phases based on the reaction rate and diffusion equation, and predicts morphological characteristics using a cooling rate-driven grain size formula. Compared to traditional steady-state assumptions and empirical formulas, this method significantly improves prediction accuracy, captures non-equilibrium behavior and local inhomogeneities during welding, and provides a scientific basis and practical value for optimizing welding processes and improving joint performance.

[0195] (1) The expected benefits and commercial value of the technical solution of this invention after transformation are as follows:

[0196] The visualization system constructed in this invention achieves high-precision visualization of the temperature field of the molten pool and keyhole, the flow field of the molten pool, the fluctuation behavior of the keyhole morphology, and the transport behavior of the Ti-Al dual-component system throughout the welding process. This guides process control, significantly reduces the number of experiments, and reduces manpower and material costs.

[0197] (2) The technical solution of this invention fills a technical gap in the industry both domestically and internationally:

[0198] The visualization system constructed in this invention can accurately acquire theoretical data that cannot be obtained or is difficult to obtain accurately in experiments, such as the temperature field of the molten pool and the keyhole, the flow field of the molten pool, the evolution behavior of the keyhole morphology, and the transport behavior of the Ti-Al bicomponent during the laser bias welding process of dissimilar metals. Therefore, this invention lays an irreplaceable theoretical foundation for the study of the macro- and micro-forming mechanisms of typical defects in Ti / Al welding and the nucleation and growth of brittle IMC, and has important academic value.

[0199] (3) Whether the technical solution of the present invention solves the technical problem that people have long wanted to solve but have never been able to solve successfully:

[0200] This invention achieves accurate prediction of the composition, content, and distribution of brittle IMCs by obtaining key thermo-mechanical parameters through a visualization system of Ti / Al laser bias brazing process, thus solving the problem of low computational efficiency caused by using methods such as phase field method, cellular automata, and molecular dynamics to predict IMC formation. Attached Figure Description

[0201] Figure 1 This is a block diagram of a high-precision visualization system for the heat transfer, flow, and component transport behavior of titanium-aluminum dissimilar metal laser bias brazing and IMC formation provided in this embodiment of the invention.

[0202] Figure 2 This is a flowchart of a high-precision visualization method for heat transfer, flow, and component transport behavior and IMC formation in titanium-aluminum dissimilar metal laser bias brazing, provided in an embodiment of the present invention.

[0203] Figure 3 The embodiments of the present invention provide (a) a method for constructing a laser biased brazing heat source model considering the differences between Ti and Al; and (b) a Gaussian laser heat source model and energy distribution diagrams at each cross section.

[0204] Figure 4 This is a model diagram of Ti / Al mixed flow and component diffusion in Ti / Al laser bias brazing provided in an embodiment of the present invention.

[0205] Figure 5 This is a solution diagram of the driving force for Ti / Al laser bias brazing provided in an embodiment of the present invention.

[0206] Figure 6 This is the boundary condition and mesh construction diagram of the computational domain for Ti / Al laser bias brazing provided in this embodiment of the invention.

[0207] Figure 7 This is a reliability verification of the numerical simulation model for laser bias brazing of Ti / Al dissimilar alloys provided in the embodiments of the present invention. (a) Simulation results with an offset of 0.3 mm (b) Simulation results with an offset of 0.7 mm (c) Experimental results with an offset of 0.3 mm (d) Experimental results with an offset of 0.7 mm.

[0208] Figure 8 The velocity field of the molten pool / keyhole when the beam offset is 0.3mm according to the embodiment of the present invention is: (a) 0ms (b) 11ms (c) 22ms (d) 33ms; local magnification of (a)~(d) (e)~(h).

[0209] Figure 9The velocity field of the molten pool / keyhole when the beam offset is 0.7mm according to the embodiment of the present invention is: (a) 0ms (b) 11ms (c) 22ms (d) 33ms; local magnification of (a)~(d) (e)~(h).

[0210] Figure 10 The following are the evolution behavior and velocity field of the small hole morphology at the Ti / Al interface under different beam offsets provided in the embodiments of the present invention: (a) 24ms with an offset of 0.3mm (b) 28ms with an offset of 0.3mm (c) 24ms with an offset of 0.7mm (d) 28ms with an offset of 0.7mm.

[0211] Figure 11 The following are the temperature evolution and melting and mixing amount of Ti on the Ti side during the welding process provided in the embodiments of the present invention: (a)~(j) when the offset is 0.3mm; (k)~(t) when the offset is 0.7mm.

[0212] Figure 12 This is a relative concentration field diagram of Al in the molten pool at different times during the solidification process of the molten pool with an offset of 0.3 mm, provided in an embodiment of the present invention.

[0213] Figure 13 This is a relative concentration field diagram of Al in the molten pool at different times during the solidification process of the molten pool with an offset of 0.7 mm, provided in an embodiment of the present invention.

[0214] Figure 14 The following are visualization results of the phase distribution and content prediction of intermetallic compounds (IMCs) under different beam offsets provided in the embodiments of the present invention, and their comparison with the actual experimental results: (a, b) Simulation results of IMC composition and content at offset 0.3 mm; (c, d) Simulation results of IMC composition and content at offset 0.5 mm; (e, f) Simulation results of IMC composition and content at offset 0.7 mm; (g) Experimental EDS elemental analysis results at offset 0.3 mm; (h) Experimental EDS elemental analysis results at offset 0.5 mm; (i) Experimental EDS elemental analysis results at offset 0.7 mm.

[0215] Figure 15 The following are the key parameters for IMC formation obtained by this system in the embodiments of the present invention: (a) thermal cycling curves at the Ti / Al interface under different biases; (b) thermal cycling curves at the upper, middle and lower parts of the interface with a bias of 0.7 mm; (c) melting amount of Ti cross section under different biases; (d) temperature gradient G at the upper, middle and lower parts of the Ti / Al interface with time under a bias of 0.3 mm; (e) temperature gradient G at the upper, middle and lower parts of the Ti / Al interface with time under a bias of 0.7 mm; (f) distribution of cooling rate GR at different depths at the Ti / Al interface under different biases.

[0216] Figure 1The module consists of: 1. Heat source model module; 2. Diffusion model module; 3. Driving force solution module; 4. Welding boundary condition solution module; 5. Prediction module. Detailed Implementation

[0217] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0218] like Figure 1 As shown in the embodiment of the present invention, a high-precision visualization system for heat transfer, flow, and component transport behavior and IMC formation in titanium-aluminum dissimilar metal laser bias brazing includes:

[0219] Heat source model module 1, diffusion model module 2, driving force solution module 3, welding boundary condition solution module 4, prediction module 5;

[0220] Heat source model module 1, connected to driving force solution module 3, is used for laser heat source model with differences in Ti and Al;

[0221] Diffusion model module 2, connected to driving force solution module 3, is used for Ti / Al laser bias brazing Ti-Al mixed flow and component diffusion model;

[0222] The driving force solution module 3 is connected to the heat source model module 1, the diffusion model module 2, the welding boundary condition solution module 4, and the prediction module 5, and is used to solve the driving force for Ti / Al laser bias brazing welding.

[0223] The welding boundary condition solution module 4 is connected to the driving force solution module 3 and is used for the boundary conditions and solution algorithm of Ti / Al laser bias brazing welding.

[0224] Prediction module 5, connected to driving force solution module 3, is used for IMC prediction based on the heat transfer-flow-component transport model of Ti / Al laser bias brazing.

[0225] like Figure 2 As shown in the embodiment of the present invention, a method for high-precision visualization of heat transfer, flow, and component transport behavior and IMC formation in titanium-aluminum dissimilar metal laser bias brazing includes:

[0226] S101, a laser heat source model considering the differences between Ti and Al;

[0227] S102, Ti / Al laser bias brazing Ti-Al mixture flow and component diffusion model;

[0228] Solution of driving force for laser bias brazing of Ti / Al in S103;

[0229] S104, Boundary conditions and solution algorithm for Ti / Al laser bias brazing;

[0230] S105, IMC prediction based on the heat transfer-flow-component transport model of Ti / Al laser bias brazing.

[0231] Another object of the present invention is to provide a computer device, the computer device including a memory and a processor, the memory storing a computer program, the computer program being executed by the processor causing the processor to perform the steps of the heat transfer-flow-component transport behavior of titanium-aluminum dissimilar metal laser bias brazing and the high-precision visualization method for IMC formation.

[0232] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the titanium-aluminum dissimilar metal laser bias brazing heat transfer-flow-component transport behavior and the high-precision visualization method for IMC formation.

[0233] Another objective of this invention is to provide an information data processing terminal, which is used to realize the heat transfer-flow-component transport behavior of the titanium-aluminum dissimilar metal laser bias brazing and the high-precision visualization system for IMC formation.

[0234] Specific implementation of the present invention:

[0235] 1. Laser heat source model considering the differences between Ti and Al

[0236] 1.1 Differences in laser energy absorption between Ti and Al

[0237] The laser absorption rates of Ti and Al in solid and liquid states and their dependence on temperature are significantly different:

[0238] The effective absorption rate of laser light is determined by the absorption rates of Ti and Al and their compositional distribution:

[0239]

[0240] The absorption rates of Ti and Al as a function of temperature can be expressed as follows:

[0241]

[0242]

[0243] Parameter notes: Ti and Al at initial temperature ( Absorption rate (dimensionless) at ). Temperature sensitivity coefficient of absorptivity (unit: ). Ambient temperature (unit: ).

[0244] 1.2 Gaussian heat source model and calculation of dynamic absorptivity of Ti and Al

[0245] (1) Power density distribution: The effective power density of the laser on the material surface is:

[0246]

[0247] Incident laser power density (unit: W / m²) 2 ), defined by the Gaussian distribution:

[0248]

[0249] The mixed absorption rate of Ti and Al depends on their mass fraction. , and temperature .

[0250] (2) Calculation of dynamic absorption rate: Mixed absorption rate expression:

[0251]

[0252] The absorption rates of Ti and Al are respectively:

[0253] Solid State:

[0254]

[0255]

[0256] Liquid:

[0257]

[0258]

[0259] The melting point temperature of the material:

[0260] The melting point of Ti (approximately 1941 K).

[0261] The melting point of Al (approximately 933 K).

[0262] (3) Spatial absorption distribution:

[0263] The energy distribution of laser light within a material is described by Beer-Lambert's law:

[0264]

[0265] Absorbed energy density inside the material (unit: W / m³) 3 ).

[0266] Effective power density on the surface.

[0267] The absorption coefficient of the material (unit: 1 / m) is calculated by weighting the absorption coefficients of Ti and Al.

[0268]

[0269] 1.3 Ray Tracing and Interface Tracing Methods

[0270] (1) Reflection and Refraction of Laser at Interfaces: When a laser beam propagates from one medium to another, some of its energy is reflected and some is transmitted to the next medium. The reflectivity and refractive index are given by the Fresnel formula:

[0271] Reflectivity:

[0272]

[0273] Refractive index:

[0274]

[0275] The reflectivity varies with the dynamic changes (solid-liquid, liquid-gas) of the Ti and Al interface, and the relevant parameters include:

[0276] Refractive index of solid Ti and Al .

[0277] Refractive index of liquid Ti and Al .

[0278] (2) Recursive tracing of laser path

[0279] The path of light propagation at the interface is calculated using Snell's Law:

[0280]

[0281] For complex interfaces (such as pinholes or molten pool surfaces), the ray path needs to be recursively traced, and the incident angle dynamically adjusted. and angle of refraction .

[0282] (3) Interface tracking

[0283] By volume fraction Indicates the interface location:

[0284]

[0285] 1.4 Coupling Solution of Heat Source Model for Ti / Al Laser Biased Brazing

[0286] Combining ray tracing, interface tracing, and the absorption differences between Ti and AI, the final expression for laser power density is:

[0287]

[0288] Interface shape (updated in real time by the interface tracking method). Incident laser power density. Dynamic absorption rate. Absorption coefficient.

[0289] Figure 3 (a) A method for constructing a heat source model for laser-biased brazing considering the differences between Ti and Al; Figure 3 (b) Model of Gaussian laser heat source and energy distribution at each cross section.

[0290] 2. Model of Ti-Al Mixture Flow and Component Diffusion in Ti / Al Laser-Biased Brazing

[0291] 2.1 Mass Conservation Equation for Ti / Al Laser Bias Brazing

[0292] Overall mass conservation equation (for multi-component fluids):

[0293]

[0294] Parameter description:

[0295] The effective density of the molten pool (unit: kg / m3) is determined by the mass fractions of Ti and Al.

[0296]

[0297] v: Velocity of fluid in the molten pool (unit: m / s).

[0298] During the laser brazing process of Ti / Al dissimilar alloys, Al and Ti undergo melting and evaporation, resulting in efflorescence in the molten pool.

[0299] 2.2 Momentum Conservation Equation for Ti / Al Laser Bias Brazing

[0300] The momentum conservation equation for molten pool flow is:

[0301]

[0302] Parameter description: Pressure field (unit: Pa). Effective viscosity of the molten pool (unit: Pa) s), calculated by viscosity weighting of Ti and Al:

[0303]

[0304] Physical force, including buoyancy and surface tension gradient driving force:

[0305]

[0306] The first term is the buoyancy driving force, where: Gravitational acceleration (unit: m / s²) ). Coefficient of thermal expansion (unit: 1 / K). Ambient temperature (unit: K). The second term is the surface tension gradient driving force (Marangoni effect), where: surface tension With temperature and components change.

[0307] 2.3 Energy Conservation Equation for Ti / Al Laser Bias Brazing

[0308]

[0309] ρ: Density (kg / m³) 3 cp: Specific heat capacity (J / (kg·K)). T: Temperature (K). v: Fluid velocity (m / s). keff: Effective thermal conductivity (W / (m·K)). qlaser: Laser thermal input (W / m²) 3 ).

[0310] 2.4 Diffusion equation of Ti / Al laser bias brazing

[0311] The mass fractions of Ti and Al in the molten pool vary with time and space, and the component diffusion equations are as follows:

[0312]

[0313] Parameter description: : The mass fraction of Ti (dimensionless, between 0 and 1). : Fluid velocity in the molten pool (unit: m / s). Concentration diffusion coefficients of Ti and Al (unit: m) 2 ( / s), calculated using the Arrhenius formula:

[0314]

[0315] - Pre-diffusion coefficient (unit: m) 2 / s). - Activation energy for diffusion (unit: J / mol). Gas constant (unit: 8.314 J / mol·K). Temperature (unit: K). The Soret effect (thermal diffusion) term describes the influence of the temperature gradient on diffusion.

[0316]

[0317] Soret coefficient (unit: 1 / K).

[0318] Coupling effect of thermal diffusion and concentration gradient in 2.5Ti / Al laser bias brazing

[0319] Within the molten pool, the coupling effect of concentration and temperature gradients further influences component diffusion behavior. Effective diffusion coefficient. Revised to:

[0320]

[0321] Figure 4 Ti / Al laser bias brazing Ti-Al mixing flow and component diffusion model

[0322] 3. Solution of driving force for Ti / Al laser bias brazing

[0323] 3.1 Surface tension gradient driving force (Marangoni effect) in Ti / Al laser bias brazing

[0324] The surface tension gradient is one of the main driving forces of molten pool flow, originating from changes in temperature and composition gradients. The driving force caused by the surface tension gradient is:

[0325]

[0326] (1) Contribution of temperature gradient: Rate of change of surface tension with temperature (unit: N / m·K):

[0327]

[0328] in, It is a negative value (surface tension typically decreases with increasing temperature). Temperature gradient. It can be obtained through the energy conservation equation, and the temperature gradient is usually the largest near the edge of the molten pool.

[0329] (2) Contribution of component gradient: Rate of change of surface tension with Ti concentration (unit: N / m):

[0330]

[0331] in, Reflects the changes in surface tension caused by the components, concentration gradient The size is usually larger at the edge of the molten pool or at the interface of different materials. Comprehensive formula:

[0332]

[0333] 3.2Ti / Al laser bias brazing buoyancy driving force

[0334] The buoyancy driving force is caused by the density change within the molten pool. The density change of the molten pool mainly comes from temperature differences and composition differences.

[0335]

[0336] Parameter description: Molten pool density (unit: kg / m³) 3 The mass fraction of the multi-component mixture is determined by its mass fraction.

[0337]

[0338] Gravitational acceleration (unit: m / s²) 2 ). Coefficient of thermal expansion (unit: 1 / K). Local temperature of the molten pool (unit: K). Ambient temperature (unit: K). The buoyancy driving force mainly acts inside the molten pool, usually pushing upwards in the high-temperature zone (center of the molten pool) and downwards in the low-temperature zone.

[0339] 3.3Ti / Al laser bias brazing steam pressure and back pressure

[0340] vapor pressure It is a function of temperature, described by the Clausius-Clapeyron equation:

[0341]

[0342] Vapor pressure at reference temperature (unit: Pa). Latent heat of vaporization (unit: J / mol). Gas constant (8.314 J / mol·K).

[0343] When a high-energy laser strikes a material surface, a localized high-pressure gas (plasma or vapor) is formed. This high-pressure gas exerts a shock force on the molten pool surface. The shock wave driving force can be expressed as:

[0344]

[0345] Local gas pressure caused by laser (unit: Pa). Surface area affected by laser (unit: m2).

[0346] Figure 5 Solution of driving force for Ti / Al laser bias brazing

[0347] 4. Boundary conditions and solution algorithm for Ti / Al laser bias brazing

[0348] The computational domain of this model is mainly divided into three parts: the Al, Ti, and air domains. The upper surface of the air domain and the lower surface of the Ti / Al domain are then designated as pressure outlets, while the remaining interfaces are all designated as walls (the upper and lower surfaces are the primary surfaces in contact with air during welding). It is worth noting that the model incorporates convective heat transfer between different interfaces. Radiative heat transfer Evaporative heat transfer These are expressed by the following formulas:

[0349]

[0350]

[0351] In the formula, The convective heat transfer coefficient, For ambient temperature, This is the Stefan–Boltzmann constant. For emissivity, For interface density, For the specific heat capacity of the interface, It is the specific heat capacity of the metal. This represents the specific heat capacity of the gas.

[0352] Figure 6 Computational Domain Boundary Conditions and Mesh Construction for Ti / Al Laser Bias Brazing

[0353] Finally, the present invention employs the following solution method to improve the accuracy and efficiency of the model:

[0354] For spatial discretization, the finite volume method (FVM) combined with the high-order WENO scheme is used to accurately handle the drastic changes in the interface and temperature gradient. For temporal discretization, implicit time integration methods (such as the Backward Euler method) are used to improve stability, and the fluid velocity and pressure are solved step by step using the PISO algorithm. In terms of interface tracking, the VOF method and level set method are combined to accurately capture the dynamic changes of the Ti / Al interface. During the calculation process, GPU parallel computing and the OpenFOAM framework are used to significantly improve computational efficiency.

[0355] 5. IMC Prediction Method Based on the Heat Transfer-Flow-Component Transport Model of Ti / Al Laser Biased Brazing

[0356] 5.1 Prediction of Various IMC Formation Types

[0357] The thermodynamic criteria for the formation of TiAl3, TiAl, and Ti3Al can be determined by calculating the Gibbs free energy change (ΔG): -ΔG<0: the reaction proceeds spontaneously, and the formed phase exists stably.

[0358]

[0359] Enthalpy of reaction (ΔH) and entropy (ΔS) can be obtained from materials thermodynamic databases (such as Thermo-Calc) or experimentally determined.

[0360] Based on the molten pool temperature field T(x,y,z,t), determine whether each region satisfies the formation condition ΔG<0 and predict the generated phase.

[0361] 5.2 Prediction of the content of various IMC formations

[0362] The reaction rate for phase formation is described by the Arrhenius equation:

[0363]

[0364] : Formation phase The reaction rate (unit: mol / m3.s)

[0365]

[0366]

[0367] :Reaction frequency factor (unit: .

[0368] Activation energy of reaction (unit: J / mol).

[0369] Gas constant (8.314 J / moldotpK).

[0370] Local temperature (unit: K).

[0371] Local mass fractions of Ti and Al.

[0372] The reaction order is determined experimentally.

[0373] mass fraction of the generated phase It can be calculated by integrating the reaction rate:

[0374]

[0375] Local density, varying with temperature and composition:

[0376]

[0377] Calculate temperature at each grid point and components Calculate the reaction rate constant. and reaction rate By integrating over time, the mass fraction of each generated phase is obtained. .

[0378] 5.3 Prediction of the Distribution of Multiple IMC Formations

[0379] The diffusion of Ti and Al components within the molten pool is described by Fick's second law:

[0380]

[0381] Local mass fraction of Ti. : Fluid velocity within the molten pool (unit: m / s). The diffusion coefficients of Ti and Al change with temperature:

[0382]

[0383] fluid velocity Calculated from the momentum equation, it significantly affects the component distribution. Buoyancy and surface tension: Buoyancy and the Marangoni effect within the molten pool affect molten pool flow: Buoyancy: Surface tension gradient:

[0384] Solving the component diffusion equation yields... and The spatial distribution of the three generated phases in the weld region is predicted by combining the reaction rate equation. .

[0385] 5.3. Morphological Prediction of Multiple IMC Formations

[0386] Cooling rate The grain size and morphology are determined by rapid cooling, which produces fine TiAl3 phase; slow cooling produces coarse Ti3Al and TiAl phases. Cooling rate calculation:

[0387]

[0388] The time derivative is calculated using the energy conservation equation. Grain size. With cooling rate Relationship:

[0389]

[0390] , : A constant determined by experimental fitting.

[0391] The cooling rate was calculated using the preceding numerical simulation model. By combining the grain size formula, the grain size distribution of each generated phase is predicted. .

[0392] IV. Specific Implementation Cases and Results

[0393] 1. Reliability Verification of a Visualized System for the Entire Ti / Al Laser-Biased Brazing Process

[0394] To verify the effectiveness and reliability of this visualization system, the process parameters for Ti / Al laser quality brazing were set to an offset of 0.3 mm (with the Ti / Al butt joint seam as the initial position, the laser beam is offset 0.3 mm towards the Al side) and an offset of 0.7 mm, respectively. Process experiments were conducted, and the visualization system developed in this invention was used to simulate the surface flow behavior and temperature field of the molten pool. As shown in the figure below, the width and length of the molten pool in the numerical simulation results are consistent with the results of the high-speed camera in the actual experiment, indicating that the visualization system has extremely high reliability.

[0395] Figure 7 Reliability verification of numerical simulation model for laser bias brazing of Ti / Al dissimilar alloys. (a) Simulation results with bias of 0.3mm (b) Simulation results with bias of 0.7mm (c) Experimental results with bias of 0.3mm (d) Experimental results with bias of 0.7mm.

[0396] 2. Visualization of Ti / Al laser-biased brazing weld pool flow behavior and Ti-Al component transport behavior

[0397] By setting the laser bias to 0.3 mm and 0.7 mm respectively, and performing post-processing operations in the visualization system developed in this invention, the flow behavior of the Ti / Al laser bias brazing weld pool and the Ti-Al component transport behavior can be obtained.

[0398] It can be observed that when the offset is 0.3 mm, because the laser is closer to Ti, more Ti melts. At this time, the diffusion direction of Ti and Al is opposing / lateral, and the component transport phenomenon is strong. Figure 8 As shown.

[0399] It can be observed that when the offset is 0.7 mm, because the laser is further away from Ti, the amount of Ti melted is less than that at 0.3 mm. At this time, the diffusion direction of Ti and Al is tangential / longitudinal, and the component transport phenomenon is significantly weakened. Figure 9 As shown.

[0400] Figure 8 Figure 9 This is clearly consistent with the experimental results.

[0401] 3. Visualization of the evolution of keyhole morphology in Ti / Al laser bias brazing

[0402] By setting laser offsets of 0.3 mm and 0.7 mm, and performing post-processing operations in the visualization system developed in this invention, the dynamic evolution behavior of the small hole morphology in Ti / Al laser-biased brazing can be obtained, such as... Figure 10 As shown.

[0403] It can be observed that when the offset is 0.3 mm, the orifice is closer to the Ti side, and the flow rate at the Al / Ti interface is significantly faster; Ti has a higher melting point, therefore the overall volume of the molten pool is smaller (as shown in a and b). Conversely, when the offset is 0.7 mm, the orifice is closer to the Al side and farther from the Ti side, the flow rate at the Al / Ti interface is significantly reduced, and the molten pool is almost entirely composed of Al, thus its volume is larger. These phenomena are clearly consistent with the experimental results.

[0404] 4. Visualization of the temperature field of the molten pool and the amount of Ti melting and mixing in Ti / Al laser-biased brazing.

[0405] By setting the laser offset to 0.3 mm and 0.7 mm, and performing post-processing operations in the visualization system developed in this invention, the temperature field of the Ti / Al laser-biased brazing weld pool and its melting and solidification behavior can be obtained, such as... Figure 11 As shown.

[0406] It can be observed that when the bias is 0.3 mm, the temperature on the Ti side is higher and the amount of melting and mixing is greater because the laser beam is more biased towards the Ti side (as shown in j). Conversely, when the bias is 0.7 mm, the temperature on the Ti side is significantly lower and the amount of melting and mixing is less because the laser beam is farther away from the Ti side (as shown in t).

[0407] 5. Visualization of molten pool concentration / solute field in Ti / Al laser biased brazing

[0408] By setting the laser bias to 0.3 mm and 0.7 mm respectively and performing post-processing operations in the visualization system developed in this invention, the concentration field of the Ti / Al laser bias brazing weld pool can be obtained.

[0409] It can be observed that when the offset is 0.3mm ( Figure 12 Since Ti melts and mixes in greater quantities, the relative concentration of Al is lower, and there is uneven distribution in the early stage, Al and Ti are mixed evenly at the end of solidification.

[0410] When the offset is 0.7mm ( Figure 13 Because the amount of Ti that melts and mixes is relatively small, the relative concentration of Al is significantly higher, and there is still uneven distribution in the early stage. However, Al and Ti are evenly mixed at the end of solidification.

[0411] The above phenomena are clearly consistent with the actual experimental process.

[0412] 6. Visualization of the composition and distribution of brittle IMC at the Ti / Al laser-biased brazing interface

[0413] With laser offsets set to 0.3 mm, 0.5 mm, and 0.7 mm, post-processing operations are performed in the visualization system developed in this invention to obtain predicted results of the brittle IMC composition and content of the Ti / Al interface in laser-biased brazing. The simulation results are then compared with experimental EDS results. Figure 14 As shown.

[0414] It can be observed that when the bias is 0.3 mm, the simulated IMC from Al to Ti side is arranged in the order of TiAl3, TiAl, and Ti3Al, with a total thickness of approximately 15 micrometers. The above results are in clear agreement with the experimental results in (g).

[0415] It can be observed that when the bias is 0.5 mm, the simulated IMC from Al to Ti side is arranged in the order of TiAl3, TiAl, and Ti3Al, with a total thickness of approximately 10 micrometers. The above results are in clear agreement with the experimental results in (gh).

[0416] It can be observed that when the bias is 0.7 mm, the simulated IMC from Al to Ti side is arranged in the order of TiAl3, TiAl, and Ti3Al, with a total thickness of approximately 5 micrometers. The above results are in clear agreement with the experimental results in (i).

[0417] As the bias increases, the laser beam moves away from the Ti side, resulting in a decrease in the amount of melting on the Ti side. The component transport behavior of Al and Ti weakens, and thus the overall IMC content shows a decreasing trend, which is consistent with the previous simulation and experimental results.

[0418] Example 1: Visual Implementation of Basic Coupling

[0419] This embodiment constructs a high-precision visualization system for laser biased brazing of titanium and aluminum dissimilar metals, comprising five functional units: dynamic absorption modeling of laser energy, multi-field coupled transport modeling of the molten pool, driving force modeling, interface and boundary behavior modeling, and prediction of intermetallic compound formation. Inputs include laser power, spot radius, scanning speed, bias distance, initial geometric dimensions of titanium and aluminum, initial temperature, ambient temperature, temperature-dependent thermophysical properties of materials, titanium-aluminum diffusion-related parameters, and reaction kinetic parameters. Outputs include temperature field, velocity field, pressure field, titanium mass fraction field, effective absorbed power density field, and content distribution and morphology of intermetallic compound phases.

[0420] During operation, the mixing absorption capacity is first calculated based on local temperature and composition. Then, the absorbed power density is written into the energy input term and iterated synchronously with the multi-field equation system. The volume force term of the momentum equation is simultaneously superimposed with the surface tension driving force caused by the temperature gradient and composition gradient, the buoyancy driving force caused by the density difference, and the recoil pressure driving force caused by evaporation. The interface is characterized by a volume fraction field, and the boundary heat transfer includes convection, radiation, and evaporation heat transfer terms. The time progression adopts an implicit strategy to ensure stability, thereby obtaining a spatiotemporal field evolution sequence and intermetallic compound formation prediction results that can be reproduced under experimental conditions.

[0421] Example 2: Comparison of absorption differences and molten pool behavior under varying bias.

[0422] This embodiment sets different laser bias values ​​under the same plate thickness and overlap type, forming three working conditions where the laser mainly acts on the aluminum side, the interface, and the titanium side. The system input remains consistent except for the bias value, and the output focuses on recording the effective absorption distribution, the free surface morphology of the molten pool, the peak position of the component gradient at the interface, the main flow circulation structure, and the continuity and thickness distribution of the intermetallic compound phase near the interface.

[0423] During the calculation, the mixing absorption capacity is dynamically updated with temperature and composition, causing the laser energy input to exhibit different spatial projections when the bias changes. This results in the migration of the peak temperature position of the molten pool, the main Marangoni driving direction, and the location of the reflow zone. By comparing the component transport and reaction rate integral results of the three sets of conditions, visual evidence can be obtained of the transformation of intermetallic compounds from a continuous layer to a local island-like distribution at the interface, and the differences in interface structure caused by changes in bias can be reproduced.

[0424] Example 3: The Influence of Phase Switching Absorption Model on the Accuracy of Results

[0425] This embodiment uses the same welding conditions and employs dynamic absorption calculation schemes that include both solid-state and liquid-state absorption temperature-sensitive models, comparing them with a scheme using only a single absorption model. Inputs include the initial absorption capacity of titanium and aluminum in the solid and liquid states, their temperature sensitivity coefficients, and their respective melting temperatures. Outputs include the absorption power density over time curves, the peak temperature error of the molten pool, the molten pool aspect ratio, the evaporation area, and the predicted thickness of intermetallic compounds.

[0426] During runtime, the model determines the phase state based on local temperature at each time step and switches the corresponding absorption capacity calculation accordingly, ensuring consistency between heat input and phase transformation process. Comparative results show that during the transition phase between molten pool formation and collapse, the dynamic phase absorption model significantly reduces temperature field jumps and stabilizes the recoil pressure term, thereby improving the stability of free surface morphology prediction. This, in turn, makes the integral of intermetallic compound formation rate closer to the physical process, demonstrating the supporting role of the phase coupling mechanism in improving model accuracy.

[0427] Example 4: Energy Deposition Correction for Ray Tracing and Interface Tracing

[0428] This embodiment addresses conditions with a tendency for pinholes or significant undulations on the free surface of the molten pool by employing an energy deposition correction scheme based on real-time updates of the interface morphology. Inputs include the relationship between interface refractive index parameters and phase changes, the incident angle range, the initial free surface morphology, and environmental medium parameters. Outputs include a diagram showing the difference in energy deposition distribution before and after correction, the proportion of interface reflection loss, the depth of heat source distribution within the molten pool, the peak molten pool flow velocity, and the spatial offset of intermetallic compound distribution.

[0429] The interface position is determined using a volume fraction field during calculation. The local incident angle is then updated based on the interface normal direction, and the ratio of reflected to transmitted energy is iteratively calculated to obtain the power density input that evolves with the interface. This correction prevents the heat source from being fixed to the initial plane, but rather dynamically distributed along the surface of the molten pool. This allows for a stable reproduction of the co-evolution of the high-temperature columnar region in the center of the molten pool and the surrounding reflow region, and provides a visual prediction of how intermetallic compounds are more easily enriched in high-temperature, high-mixing regions.

[0430] Example 5: Verification of the separation of Marangoni temperature and component terms

[0431] This embodiment decomposes the surface tension gradient driving force into two parts: the temperature gradient contribution and the composition gradient contribution, and performs comparative calculations on individual and combined activation. The input includes two types of parameters: the rate of change of surface tension with temperature and the rate of change of surface tension with titanium mass fraction, as well as the same thermal input conditions; the output records the velocity vector distribution on the molten pool surface, the number of recirculation rings, the thickness of the mixed layer at the interface, and the continuity index of the intermetallic compound phase.

[0432] During operation, temperature-driven and composition-driven flows were applied separately to the surface shear term of the momentum equation, while ensuring that other conditions remained consistent. The results were reproducible: under temperature-driven flow alone, the flow structure was dominated by a single ring from the center outward or from the outside to the center; under composition-driven flow alone, shear flow enhanced on the bias side appeared near the interface; and under combined driving flow, stronger interfacial mixing and steeper composition gradient peaks were formed, resulting in a significant increase in the formation rate of intermetallic compounds near the interface, thus supporting the necessity of a synergistic driving force mechanism.

[0433] Example 6: Contribution of buoyancy-driven convection to the interior of the molten pool

[0434] This embodiment demonstrates the effect of buoyancy-driven force caused by density changes on convection within the molten pool. Different gravity direction conditions are set up, corresponding to the normal gravity direction and the reverse gravity direction, respectively. Inputs include the coefficient of thermal expansion, the relationship between density and temperature and composition, and identical heat sources and geometric conditions. Outputs include the vertical velocity component distribution within the molten pool, temperature stratification, composition stratification, and the offset of intermetallic compounds in the height direction.

[0435] When solving the problem, the buoyancy term is incorporated into the body force term based on the local density difference and the direction of gravity, while keeping the other driving forces constant. A comparison reveals that under normal gravity, the high-temperature, low-density region floats upwards, forming an updraft, while under the opposite gravity direction, reverse convection occurs, causing a positional reversal between the interfacial mixing layer and the intermetallic compound enrichment region. This mechanistically demonstrates that the coupling of buoyancy driving force and temperature field has a considerable impact on the transport process.

[0436] Example 7: Changes in free surface morphology caused by recoil pressure

[0437] This embodiment selects a condition with significant evaporation and uses a vapor pressure and recoil pressure driving force model to predict free surface depressions and surface fluctuations. Inputs include latent heat of vaporization, reference vapor pressure, gas constant, local temperature-to-vapor pressure mapping, and ambient pressure boundary; outputs include the free surface depression depth over time curve, local pressure peak distribution, molten pool surface velocity field, and the enrichment distribution of intermetallic compounds around the depression.

[0438] In the calculation, the vapor pressure is updated in real time by the temperature field and converted into surface normal pressure applied to the free surface region. The recoil pressure and Marangoni shear work together to form complex surface flow. The results are reproducible: when the recoil pressure increases, the depression deepens and a stronger surface backflow is formed. The composition gradient near the interface is stretched, resulting in the enrichment of intermetallic compounds in a banded pattern at the edge of the depression, which provides support for the feasibility of evaporation-driven and morphology prediction.

[0439] Example 8: Implementation of unified boundary conditions for boundary heat transfer and evaporative heat transfer

[0440] This embodiment employs a unified boundary heat transfer representation that simultaneously incorporates convective heat transfer, radiative heat transfer, and evaporative heat transfer into the free surface heat flux boundary. Inputs include convective heat transfer coefficient, emissivity, ambient temperature, Stefan constant, latent heat of vaporization, gas molecular weight, and reference vapor pressure; outputs include free surface heat flux distribution, cooling rate distribution, grain size index distribution, and predicted intermetallic compound morphology.

[0441] During runtime, the net heat flux of the free surface region is calculated at each time step and applied to the energy equation. Evaporative heat transfer and recoil pressure terms are updated synchronously by the temperature field. This process can reproduce the spatial difference in the transition of intermetallic compound morphology from fine to coarse phases due to the higher cooling rate at the edge of the free surface and the lower cooling rate in the central region. This difference can be visualized by mapping the cooling rate to the grain scale.

[0442] Example 9: Joint Prediction of Heterogeneous Formation Criteria and Reaction Kinetics

[0443] This embodiment addresses the multiphase formation of common intermetallic compounds in the titanium-aluminum system, setting up a joint prediction process that includes free energy criteria and reaction kinetics integrals. Inputs include enthalpy and entropy change data for each formation reaction, reaction frequency factor, activation energy, reaction order, and temperature and composition fields. Outputs include formation region determination maps for each phase, mass fraction variation curves over time, spatial distribution cloud maps, and interphase competition relationship diagrams.

[0444] The calculation first calculates the free energy change based on temperature at each grid point and determines the probability of generation. Then, within the region that meets the generation conditions, the rate constant is calculated and time integration is performed to obtain the phase content. This process is reproducible: different regions exhibit spatial partitioning of phase types due to differences in temperature and composition, and the generation rate decays and freezes the phase distribution during the rapid temperature decrease phase, thereby achieving synchronous prediction of the distribution of multiple phase types and supporting a thermodynamic and kinetic synergistic mechanism.

[0445] Example 10: Implementation of the Correlation between Morphology Prediction and Cooling Rate

[0446] This embodiment is used to achieve morphology prediction output, employing grain-scale mapping driven by cooling rate. Inputs include the time history of the temperature field, grain-scale mapping coefficients and exponential parameters, and initial nucleation conditions for each region; outputs include a three-dimensional distribution map of the cooling rate, a three-dimensional distribution map of the grain size, the segmentation results of fine-grained and coarse-grained regions, and a visualization of the morphological differences of intermetallic compounds.

[0447] During operation, the rate of temperature change over time is obtained from the energy equation, and the cooling rate is calculated at each spatial location and mapped to obtain the grain size index. This implementation method is reproducible: the cooling rate is lower and the grain size is larger in the heat input center region, while the cooling rate is higher and the grain size is smaller in the edge region, thus forming a visualized morphology gradient output, providing sufficient support for the feasibility and repeatability of morphology prediction.

[0448] Figure 15 The key parameters for IMC formation obtained by this system are: (a) thermal cycling curves at the Ti / Al interface under different biases; (b) thermal cycling curves at the upper, middle and lower parts of the interface with a bias of 0.7 mm; (c) melting amount of Ti cross section under different biases; (d) temperature gradient G at the upper, middle and lower parts of the Ti / Al interface with time under a bias of 0.3 mm; (e) temperature gradient G at the upper, middle and lower parts of the Ti / Al interface with time under a bias of 0.7 mm; and (f) distribution of cooling rate GR at different depths at the Ti / Al interface under different biases.

[0449] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A high-precision visualization system for the heat transfer, flow, component transport behavior, and intermetallic compound formation in laser biased brazing of titanium and aluminum dissimilar metals, characterized in that... Includes: laser energy dynamic absorption modeling unit, molten pool multi-field coupling transport modeling unit, driving force formation and action modeling unit, interface and boundary behavior modeling unit, and intermetallic compound formation prediction unit; The laser energy dynamic absorption modeling unit constructs an absorption model that considers the differences in laser energy absorption of titanium and aluminum at different temperatures and phase states, and dynamically calculates the effective absorption distribution based on the local temperature field and component field. The molten pool multi-field coupled transport modeling unit constructs a multi-field coupled model that includes mass conservation, momentum conservation, energy conservation, and component transport, used to describe the mixed flow heat transfer process and component migration process of titanium and aluminum in the molten pool. The driving force formation and action modeling unit constructs a model of surface tension gradient driving force, buoyancy driving force and recoil pressure driving force caused by temperature gradient, component gradient and density change, which is used to drive the flow behavior in the molten pool. The interface and boundary behavior modeling unit constructs a dynamic evolution model of the molten pool free surface and gas interface, as well as the titanium-aluminum interface, and constructs the corresponding heat-mass-momentum exchange boundary conditions. The intermetallic compound formation prediction unit calculates the spatial distribution of formation conditions, formation rate, and morphological evolution of intermetallic compounds based on temperature field composition field and time history. The aforementioned units are coupled through a unified spatial-temporal field, forming a synergistic closed loop between the formation of the driving force for the laser energy absorption transport process and the formation of intermetallic compounds.

2. The system as described in claim 1, characterized in that, The laser energy dynamic absorption modeling unit determines the mixed absorption capacity based on the local mass fraction and local temperature of titanium and aluminum, and couples the mixed absorption capacity as a laser input term with the energy conservation process.

3. The system as described in claim 1, characterized in that, The multi-field coupled transport modeling unit of the molten pool simultaneously introduces the changes in physical properties caused by temperature changes and the changes in physical properties caused by composition changes into the flow and heat transfer processes, so that the flow field, heat transfer field and composition field form a mutual feedback relationship.

4. The system as described in claim 1, characterized in that, The driving force formation and action modeling unit simultaneously calculates the surface tension changes caused by temperature gradient and the surface tension changes caused by composition gradient, and superimposes the two as different sources of the same surface driving force.

5. The system as described in claim 1, characterized in that, The intermetallic compound formation prediction unit determines whether an intermetallic compound has formed based on changes in free energy and calculates the formation rate based on reaction kinetics.

6. A high-precision visualization method for the heat transfer, flow, component transport behavior, and intermetallic compound formation in laser biased brazing of titanium and aluminum dissimilar metals, characterized in that... Includes the following steps: Construct a dynamic absorption model of laser energy that considers the absorption differences between titanium and aluminum; Construct a multi-field coupled model that includes flow heat transfer and component transport; Construct a driving force model based on temperature and composition and apply it to the flow process; Construct a dynamic behavior model for the molten pool interface and boundary; Predicting the formation behavior of intermetallic compounds based on temperature and composition fields; The above steps are coupled and solved within the same computational framework, so that the evolution of laser energy input to the molten pool behavior and the formation of intermetallic compounds form a causal closed loop.

7. The method as described in claim 6, characterized in that, When constructing a dynamic absorption model for laser energy, the influence of material phase changes on absorption capacity is simultaneously incorporated into the calculation.

8. The method as described in claim 6, characterized in that, When constructing a multi-field coupled model, the effects of temperature change on diffusion capacity and the effects of composition change on diffusion capacity are both incorporated into the component transport process.

9. The method as described in claim 6, characterized in that, When constructing the driving force model, the surface tension gradient driving force, buoyancy driving force, and recoil pressure driving force are all used as the flow driving source.

10. The method as described in claim 6, characterized in that, When predicting the formation behavior of intermetallic compounds, the spatial distribution of formation rate and grain morphology are used as different outputs of the same prediction process.

Images