An interface momentum processing method suitable for numerical simulation of arc-droplet-molten pool coupling in aluminum alloys

Abstract

The present invention discloses an interface momentum processing method suitable for numerical simulation of aluminum alloy arc-droplet-molten pool coupling, which aims to solve the problem of arc plasma-droplet metal phase interface morphology distortion and non-convergence of calculation that are easy to appear in the integrated numerical simulation process of aluminum alloy arc-droplet-molten pool. The interface momentum processing method is based on the fluid volume fraction method (VOF) tracking arc plasma-metal phase interface as an application premise, and is mainly implemented in the construction and solution of the momentum equation of the integrated model. Specifically: based on the determination of the unit metal VOF value and the VOF gradient value, the outermost unit layer L1 of the phase interface close to the arc domain and the outermost unit layer L2 of the arc domain adjacent to the phase interface are extracted in real time, and the abnormal fluid velocity trimming threshold is defined to perform real-time velocity correction on the velocity abnormal unit in the unit layer L1, and the real-time momentum compensation is performed on the internal unit of the droplet by statistically analyzing the axial fluid velocity distribution and the droplet size of the unit layer L2.

Description

Technical Field

[0001] The invention belongs to the technical field of numerical simulation of welding process. Background Art

[0002] The arc welding process involves complex physical phenomena, including arc plasma combustion, heat and momentum transfer of molten metal droplets, flow of liquid metal in the molten pool, and evolution of solidified dendrites. These complex physical phenomena are coupled with each other and jointly determine the final mechanical properties and service life of the welded structure. Therefore, it is of great significance to understand the internal generation mechanism and action mechanism of these physical phenomena in order to control and improve the quality of welded structures. Numerical simulation technology is an effective means to study such problems. It constructs various physical control equations, based on the actual state equivalent initial boundary conditions, and uses numerical discrete methods to solve and calculate. It can intuitively reproduce the evolution of various physical fields in the welding process, thereby revealing the generation mechanism of various physical phenomena and the internal interaction mechanism.

[0003] Aluminum alloy is an engineering material widely used in aviation, aerospace, rail transportation and other fields, and arc welding is its main processing and manufacturing method. At present, the numerical model of heat transfer and metal flow in the arc welding process of aluminum alloy is mainly based on a simplified method, that is, the heat source model (such as the double ellipsoid heat source model) is used to equate the heating process of the arc to the molten pool, the arc pressure with Gaussian distribution is used to describe the disturbance process of the arc to the molten pool, and the empirical formula is used to describe the dragging effect of the arc plasma on the molten droplet and the electromagnetic compression effect on the molten droplet itself. This simplified method can describe most of the physical behaviors of its welding process to a certain extent, but the accuracy of the model generally requires a large number of experiments to obtain key input parameters, as well as repeated debugging and correction of some empirical parameters to ensure, which is labor-intensive and inefficient. Compared with the simplified method, the arc-droplet-molten pool integrated coupling modeling and analysis method provides a new idea. By establishing a unified heat and mass transfer control equation in different calculation domains of the arc, droplet and molten pool, setting reasonable boundary conditions, and solving it in a unified way, the interaction process of the arc, droplet and molten pool can be directly reproduced. This fully coupled method not only has high calculation accuracy, but also eliminates the tedious model parameter verification process. However, due to the characteristics of aluminum alloy such as high conductivity, low viscosity, and low surface tension, when the arc-droplet-molten pool fully coupled approach is used to directly model and solve the welding process, problems such as distortion of the arc plasma-droplet liquid metal interface morphology and numerical divergence occur, resulting in calculation interruption. Therefore, in order to realize the integrated coupled numerical simulation of arc-droplet-molten pool in the aluminum alloy welding process, it is necessary to explore a reasonable coupling model interface correction method. Summary of the invention

[0004] Objective of the Invention: To solve the problems of distorted interface morphology and divergent calculation between the arc plasma and the molten droplet liquid metal in the integrated coupling simulation of aluminum alloy welding arc-droplet-pool, and to more realistically reproduce the phenomena of arc combustion, molten metal transfer, and molten pool liquid metal flow. The present invention provides an interface momentum processing method applicable to the coupled numerical simulation of aluminum alloy welding arc-droplet-pool.

[0005] Technical Solution: The present invention provides an interface momentum processing method applicable to the coupled numerical simulation of aluminum alloy welding arc-droplet-pool. This method is premised on the use of the volume of fluid (VOF) method for dynamic tracking of the arc plasma-metal phase interface, and is mainly implemented in the construction and solution of the momentum conservation equations during the coupled numerical simulation of aluminum alloy welding arc-droplet-pool, without affecting other modeling processes. The present invention specifically includes the following steps:

[0006] Step 1: Based on the dynamic evolution of the arc plasma-metal phase interface described by the volume of fluid (VOF) method, calculate the metal VOF values and metal VOF gradient values of the cells in the entire domain, and real-time determine the cell layer L1 closest to the arc plasma domain on the arc plasma-droplet metal phase interface.

[0007] Step 2: Define the interface abnormal fluid velocity trimming threshold, and through real-time traversal of the marked interface cell layer L1, perform momentum correction on its fluid velocity abnormal cells.

[0008] Step 3: Based on the metal VOF values and metal VOF gradient values of the cells in the entire domain, real-time mark the cell layer L2 on the side of the arc plasma domain close to the arc plasma-droplet metal phase interface.

[0009] Step 4: Considering the interference of the interface momentum correction on the momentum transfer of the arc plasma to the droplet metal, perform real-time momentum correction on the inside of the droplet based on the axial velocity distribution of the plasma in the cell layer L2 and the size and morphology of the droplet.

[0010] Further, in the above Step 1: The volume of fluid VOF describing the dynamic evolution of the arc plasma-droplet metal phase interface satisfies:

[0011]

[0012] where F is the metal volume fraction in the cell, F = 1 indicates that the cell is a metal cell, F = 0 indicates that the cell is an arc plasma cell, and 0 < F < 1 indicates that the cell is an interface cell, t is time, is the fluid velocity;

[0013] Extract the metal VOF values F of all cells in the computational domain, and calculate the metal VOF gradient values of all cells

[0014]

[0015] Among them, F x , F y , F z are the gradient components of the metal volume fraction in the unit along the x, y, and z directions respectively;

[0016] By simultaneously determining the unit metal VOF value F and VOF gradient value The outermost unit layer L1 of the arc plasma domain near the interface is extracted, and the judgment conditions are met:

[0017]

[0018] Among them, F limt It is the upper limit of the volume fraction detection of the outermost unit of the interface. Its value is closely related to the grid unit size and is generally taken as 0.01. If the interface layer profile captured by this value is incomplete or multi-layer superposition, this value can be appropriately increased or decreased.

[0019] Furthermore, in step 2, the interface unit abnormal speed trimming threshold V limt The value of should not be less than the maximum velocity of the fluid in the molten liquid metal; for short-circuit transition of aluminum alloy arc welding, the maximum velocity of the molten liquid metal during the transition process generally does not exceed 2.0m / s. Therefore, the abnormal velocity trimming threshold V of the interface unit can be taken as limt =2.0m / s; for droplet transition or droplet transition modes, the abnormal velocity trimming threshold V limt Can be appropriately enlarged.

[0020] In each solution iteration step, the units in the interface unit layer L1 are traversed in a loop. If the interface unit velocity vector is not within the abnormal velocity pruning threshold, it is pruned:

[0021]

[0022] Among them, V x , V y , V z are the components of the unit's current velocity vector along the x, y, and z directions, respectively. x ′,V y ′,V z′ is the corrected velocity component, and α is the pruning factor. In order not to reduce the pruning speed and cause over-pruning, the pruning factor α can be between 0.90 and 0.95. In most cases, the velocity of the unit after one pruning cannot meet the pruning threshold limit, so it is necessary to perform multiple velocity pruning on the velocity abnormal unit in one iteration step until the threshold requirement is met. The velocity V obtained by pruning x ′,V y ′,V z 'Directly to the unit current abnormal speed V x , V y , V z Replace and update, and participate in the calculation of the next iteration step, thereby completing the momentum correction in the unit layer L1.

[0023] Furthermore, in step 3, the unit metal VOF value F and VOF gradient value The extraction and calculation of are the same as step 1; when the arc plasma calculation domain is close to the unit layer L2 on the interface side, the unit metal VOF value F and VOF gradient value are obtained. The following criteria must be met simultaneously:

[0024]

[0025] Among them, δ is the upper limit of the detection of the cell volume fraction gradient on the side of the arc plasma domain closest to the interface. Its value is closely related to the grid cell size and is generally taken as 0.001. If the cell layer contour captured by this value is incomplete or multi-layer superposition, this value can be appropriately increased or decreased.

[0026] Furthermore, in step 4, considering the interference of the arc plasma's momentum transfer to the molten metal due to the interface momentum correction, momentum compensation is performed on the molten metal domain unit; the momentum compensation amount is added to the axial momentum conservation equation in the form of a volume force source term:

[0027]

[0028] Among them, M d is the momentum compensation body force, β d is the momentum compensation coefficient, ρ p is the arc plasma density, A d is the axial projection area of ​​the droplet, is the average axial velocity value in unit layer L2, V d is the droplet volume;

[0029] Momentum compensation coefficient β d The value is generally in the range of 0.1-1.0, and can be determined by calculation and verification based on the droplet transition frequency obtained from the experiment;

[0030] Loop through the unit layer L2 and extract the axial velocity component v of each unit in real time zi , calculate the average axial velocity

[0031]

[0032] Wherein, N is the total number of units contained in the unit layer L2;

[0033] Taking the unit metal VOF value F = 0.5 in the calculation domain as the statistical segmentation interface, the molten droplet size and shape are obtained in real time, and the axial projection area A of the molten droplet is calculated. d and the volume V occupied by the droplet d :

[0034]

[0035] Among them, ΔL r is the maximum radial size difference of the droplet, N′ is the total number of units contained in the droplet in the calculation domain, Δx i , Δy i , Δz i It is the unit size of the droplet unit in the x, y, and z directions.

[0036] Beneficial effects: The present invention solves the problems of interface morphology distortion and calculation divergence that occur in the integrated coupled simulation process of aluminum alloy welding arc-droplet-molten pool by correcting the momentum of the arc plasma-molten droplet metal interface in real time; the correction method adopted is simple and easy to implement, and can assist in truly reproducing the evolution of temperature field, flow field, electric field, induced magnetic field and other processes in the aluminum alloy arc welding process, thereby helping to deepen the understanding of the essence of various physical phenomena in the aluminum alloy arc welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] FIG. 1 is an implementation flow of the present invention. picture;

[0038] Figure 2 Calculation domain of aluminum-magnesium alloy MIG cylindrical coupling simulation ;

[0039] Fig. 3 Distribution diagram of unit layer L1 extracted at time t = 0.044s on the YOZ section;

[0040] Fig. 4 The cells with abnormal velocity and trimmed in the unit layer L1 on the YOZ section at time t = 0.044s;

[0041] Fig. 5 Distribution of unit layer L2 extracted at time t = 0.044s on the YOZ section;

[0042] Figure 6 Comparison of droplet morphology captured in the new transition period 0.044s test and the calculation results of the coupling model ;

[0043] Fig.7 Temperature field distribution of YOZ section at 0.044s calculation time of aluminum-magnesium alloy coupling model . DETAILED DESCRIPTION

[0044] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0045] The implementation process of the present invention is described below in conjunction with the numerical simulation process of arc-droplet-molten pool coupling in metal arc welding (MIG) of aluminum-magnesium alloys.

[0046] As shown in FIG1 , it is a flowchart of the implementation of the present invention. The interface momentum processing method thereof is based on the premise of using the volume fraction method (VOF) to track the arc plasma-metal interface, and is mainly used in the construction and solution of the momentum conservation equation.

[0047] Before applying the present invention, the basic process of constructing the aluminum-magnesium alloy MIG arc-molten droplet-molten pool coupling model is first carried out. A cylindrical coupling simulation calculation domain of aluminum-magnesium alloy MIG is created as shown in Figure 2, with a bottom radius of 10 mm and a height of 15 mm, and the calculation domain is meshed and discretized.

[0048] Construct the governing equations:

[0049] The mass conservation equation:

[0050]

[0051] Momentum conservation equation:

[0052]

[0053] Energy conservation equation:

[0054]

[0055] Current continuity equation:

[0056]

[0057] Maxwell's equations:

[0058]

[0059] in, is the fluid velocity vector, ρ is the fluid density, t is the time, μ is the dynamic viscosity, p is the pressure, S m is the momentum source term, H is the thermal enthalpy, T is the temperature, S e is the energy source term, σ is the conductivity, φ is the electric potential, A is the magnetic potential loss, μ0 is the vacuum magnetic permeability, is the current density, is the magnetic field strength vector.

[0060] The volume fraction method (VOF) is used to track the arc plasma-metal phase interface:

[0061]

[0062] Among them, F is the volume fraction of metal in the cell. F = 1 indicates that the cell is a metal cell, F = 0 indicates that the cell is an arc plasma cell, and 0 < F < 1 indicates that the cell is a phase interface cell.

[0063] A material model is established and boundary conditions are set for the computational domain:

[0064] The physical property parameters of aluminum-magnesium alloy and argon are taken from material handbooks and published literature. Except for the density being a constant, other parameters such as specific heat capacity, thermal conductivity, electrical conductivity, viscosity, etc. are all functions of temperature; the welding current used in the model is 110 A, the voltage is 15.8 V, the wire feeding speed is 0.108 m / s, the shielding gas is argon, and its flow rate is 20 L / min. Therefore, the end of the wire is set as the velocity inlet with an inlet velocity of 0.108 m / s, the bottom and side of the substrate are wall boundaries, the side of the argon domain is the pressure outlet boundary, the top of the argon domain is the argon velocity inlet with an inlet velocity of 1.76 m / s, the magnetic vector potential has a flux of 0 on other boundaries except for being 0 on the side of the argon domain and the side of the substrate; the electric potential has a flux of 3.5e7 A / m at the end of the wire 2 , 0 at the bottom of the substrate, and 0 flux on other boundaries.

[0065] Next, the present invention is applied to perform real-time momentum correction on the arc plasma-droplet metal phase interface during the solution process of the arc-droplet-pool coupling model:

[0066] Step 1: Real-time extract the metal VOF value F of all cells in the computational domain and calculate the metal VOF gradient value of all cells

[0067]

[0068] Among them, F x , F y , F z are the gradient components of the metal volume fraction in the cell along the x, y, and z directions respectively;

[0069] By simultaneously judging the cell metal VOF value F and the VOF gradient value extract the outermost cell layer L1 of the arc plasma domain adjacent to the interface, and the judgment condition is satisfied:

[0070]

[0071] Among them, F limtThe value is 0.01, as shown in FIG3 , which is the distribution of the unit layer L1 on the YOZ section extracted in real time at 0.044 s.

[0072] Step 2: Since the aluminum-magnesium alloy MIG welding droplet transition form is short-circuit transition when the welding current is 110A, the interface unit abnormal trimming threshold V is set limt =2.0m / s. In each solution iteration step, the units in the interface unit layer L1 are traversed in a loop. If the interface unit velocity vector is not within the abnormal velocity pruning threshold, it is pruned. The pruning factor is set to 0.9 to ensure the pruning speed and pruning quality:

[0073]

[0074] Among them, V x , V y , V z are the components of the unit's current velocity vector along the x, y, and z directions, respectively. x ′,V y ′,V z ′ is the corrected velocity component. For those that fail to meet the abnormal velocity threshold range after one pruning, multiple pruning is performed until the requirement is met. As shown in FIG4 , this is a real-time unit with abnormal velocity in unit layer L1 on the YOZ section at the calculation time of 0.044 s and undergoing pruning correction.

[0075] Step 3: Loop through the calculation of the global unit. If the unit metal VOF value F and VOF gradient value When the following conditions are met at the same time, the unit is marked as unit layer L2:

[0076]

[0077] Among them, δ is 0.001. As shown in FIG5 , the distribution of the unit layer L2 extracted in real time at the calculation time of 0.044 s on the YOZ section is shown.

[0078] Step 4: The unit metal VOF value F = 0.5 in the calculation domain is used as the statistical segmentation interface to obtain the droplet size and shape in real time, and calculate the real-time axial projection area A of the droplet d and the volume V occupied by the droplet d :

[0079]

[0080] Among them, ΔL r is the maximum radial size difference of the droplet, N′ is the total number of units contained in the droplet in the calculation domain, Δx i , Δy i , Δz iIt is the unit size of the droplet unit in the x, y, and z directions.

[0081] Loop through the unit layer L2 and extract the axial velocity component v of each unit in real time zi , calculate the average axial velocity

[0082]

[0083] Wherein, N is the total number of units contained in the unit layer L2;

[0084] Based on the axial projection area of ​​the droplet A d 、Volume V d and the average axial velocity Calculate the droplet momentum compensation M d :

[0085]

[0086] Among them, β d is the momentum compensation coefficient, which is set to 5.0 after debugging, ρ p is the arc plasma density.

[0087] The momentum compensation amount M d Applied as a source term in the axial momentum equation:

[0088]

[0089] Among them, V z is the metal axial velocity component, S mz is the axial component of the body force source term including electromagnetic force, buoyancy, surface tension, etc.

[0090] By correcting the momentum of the arc plasma-molten droplet metal phase interface in the present invention, the stable solution calculation of the arc-molten droplet-molten pool integrated coupling model of aluminum-magnesium alloy arc welding can be achieved, the arc plasma-molten droplet metal interface morphology distortion and non-convergence problems can be avoided, and the dynamic evolution of physical fields such as temperature field, velocity field, electric field, induced magnetic field, etc. in the welding process can be truly reproduced. As shown in Figure 6, the molten droplet morphology photographed in the aluminum-magnesium alloy test is compared with the calculation results of the coupling model. The experimental molten droplet morphology is taken at 0.044s after the end of the previous droplet transition. It can be seen that the simulation results are in good agreement with the experimental results, indicating that the interface momentum processing method introduced in the present invention is reasonable. Figure 7 is the YOZ cross-section temperature field distribution at 0.044s calculated by the aluminum-magnesium alloy arc-molten droplet-molten pool integrated model. The temperature of the bottom of the molten droplet reaches above 2000K, which is basically consistent with the molten droplet temperature of other aluminum alloys with similar components reported in the literature.

[0091] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.

Claims

1. An interface momentum processing method suitable for numerical simulation of aluminum alloy arc-droplet-molten pool coupling, characterized in that The steps include: Step 1: Based on the dynamic evolution of the arc plasma-metal phase interface described by the fluid volume fraction method VOF, the metal VOF value and metal VOF gradient value of the unit in the whole domain are calculated and obtained, and the unit layer L1 of the arc plasma-molten droplet metal phase interface closest to the arc plasma domain is determined in real time; Step 2: define the abnormal interface fluid velocity trimming threshold, and perform momentum correction on the abnormal fluid velocity unit by traversing the marked interface unit layer L1 in real time; Step 3: Based on the metal VOF value and metal VOF gradient value of the cells in the whole domain, the cell layer L2 on the side of the arc plasma domain close to the arc plasma-molten droplet metal phase interface is marked in real time; Step 4: Considering the interference of the momentum transfer of arc plasma to the molten metal due to the interface momentum correction, the real-time momentum correction of the inside of the molten droplet is performed based on the plasma axial velocity distribution of the unit layer L2 and the size and shape of the molten droplet; The momentum correction of the molten metal droplet is achieved by adding a body force source term to the axial momentum equation. The magnitude of the body force is: Among them, M d is the momentum compensation body force, β d is the momentum compensation coefficient, which is in the range of 0.1-1.

0. It is calculated and verified based on the experimental droplet transition frequency. p is the arc plasma density, A d is the axial projection area of ​​the droplet, is the average axial velocity value in unit layer L2, V d is the droplet volume; By looping through the unit layer L2, the axial velocity component v of each unit is extracted in real time zi , we can calculate: Wherein, N is the total number of units contained in the unit layer L2; Axial projection area of ​​molten droplet A d and the volume V occupied by the droplet d The VOF value of the unit metal in the calculation domain is F = 0.5 as the statistical segmentation interface, and the real-time calculation is obtained: Where, ΔL r is the maximum radial size difference of the droplet, N′ is the total number of units contained in the droplet in the calculation domain, Δx i , Δy i , Δz i It is the unit size of the droplet unit in the x, y, and z directions.

2. The interface momentum processing method suitable for numerical simulation of aluminum alloy arc-droplet-molten pool coupling according to claim 1 is characterized in that: The step 1 is to dynamically track the arc plasma-metal interface based on the fluid volume fraction method VOF, and obtain the metal VOF value F and metal VOF gradient value of the unit in the whole domain by calculation. When the unit layer L1 on the side of the arc plasma domain closest to the arc plasma-droplet metal phase interface is judged, the metal VOF value F and the metal VOF gradient value of the unit are The following conditions must be met at the same time: Among them, F limt It is the upper limit of the volume fraction detection of the outermost unit of the interface. Its value is closely related to the grid unit size and is taken as 0.

01. If the interface layer contour captured by this value is incomplete or multi-layer superposition, increase or decrease this value appropriately.

3. The interface momentum processing method suitable for numerical simulation of aluminum alloy arc-droplet-molten pool coupling according to claim 1 is characterized in that: In step 2, the interface unit abnormal speed trimming threshold V limt The value of should not be less than the maximum velocity of the fluid in the molten liquid metal; for short-circuit transition of aluminum alloy arc welding, the maximum velocity of the molten liquid metal during the transition process does not exceed 2.0m / s, so the abnormal velocity trimming threshold V of the interface unit is taken as limt =2.0m / s; for droplet transition or shot droplet transition mode, the abnormal velocity trimming threshold V limt Appropriate enlargement; The trimming judgment condition and trimming amount of the fluid velocity abnormal unit in the interface unit layer L1 are: Among them, V x , V y , V z are the components of the unit's current velocity vector along the x, y, and z directions, respectively. x ′,V y ′,V z ′ is the corrected velocity component, and α is the pruning factor. In order not to reduce the pruning speed and cause over-pruning, the pruning factor α is set between 0.90 and 0.

95. When the fluid velocity abnormal unit is pruned, there is a situation where the unit velocity after one pruning cannot meet the abnormal velocity pruning threshold limit. Therefore, it is necessary to perform multiple velocity pruning on the velocity abnormal unit in one iteration step until the threshold requirement is met; the pruned velocity V x ′,V y ′,V z 'Directly to the unit current abnormal speed V x , V y , V z Replace and update, and participate in the calculation of the next iteration step, thereby completing the momentum correction in the unit layer L1.

4. The interface momentum processing method suitable for numerical simulation of aluminum alloy arc-droplet-molten pool coupling according to claim 1 is characterized in that: In step 3, when marking the unit layer L2 on the side of the arc plasma domain close to the arc plasma-molten droplet metal interface, the unit metal VOF value F and VOF gradient value The following criteria must be met simultaneously: Among them, δ is the upper limit of the volume fraction gradient detection of the unit on the side of the arc plasma domain closest to the interface. Its value is closely related to the grid unit size and is taken as 0.

001. If the unit layer contour captured by this value is incomplete or multi-layer superposition, this value can be appropriately increased or decreased.

Images