A method for designing ring scanning laser welding processes

Abstract

This invention discloses a method for designing a ring-scanning laser welding process, belonging to the field of ring-scanning laser welding technology. This method addresses the problems of poor assembly gap tolerance, high experimental costs, and low production efficiency in traditional laser welding by optimizing experimental design and simulation calculations. The steps include: determining process design parameters and levels, selecting experimental design types, determining weld formation characteristic parameters, designing an experimental matrix, conducting process benchmark experiments and simulation benchmark experiments, correcting the energy coefficient of the simulation model, completing simulation calculations for the remaining experimental schemes, and optimizing parameter combinations through data analysis. Compared to traditional methods, this invention reduces the number of experiments and costs, improves production efficiency, possesses good versatility and adaptability, can quickly design welding process parameters, and is suitable for welding thin-walled workpieces of different materials and thicknesses.

Description

Technical Field

[0001] This invention relates to a process design method for ring scanning laser welding, which belongs to the field of ring scanning laser welding technology. Background Technology

[0002] Laser welding, with its concentrated energy density, can significantly improve the heat-affected zone and joint microstructure of thin-walled workpieces. However, it suffers from poor tolerance for assembly gaps, long pre-weld preparation time, and low production efficiency. Circular scanning laser welding, through the coordinated operation of the Z-axis focusing axis and the XY-axis motion, achieves rapid welding formation over a wide range, effectively solving the problems of poor assembly gap tolerance and low production efficiency inherent in traditional laser welding. It is a highly promising welding method. Compared to traditional laser welding, circular scanning laser welding involves numerous process parameters, such as laser power, welding speed, defocusing amount, laser scanning frequency, scanning amplitude, and scanning direction, posing challenges to its process design. Inappropriate process parameter design can lead to poor weld melting ratios and weld pool morphology, resulting in various forming defects. Due to the numerous process parameters in circular scanning laser welding, traditional experimental methods require significant manpower, material resources, and financial investment, increasing process costs and reducing production efficiency. These issues limit the large-scale application of circular scanning laser welding in thin-walled workpieces. Summary of the Invention

[0003] To address the issues of high experimental costs and low production efficiency in ring scanning laser welding, this invention proposes a process design method for ring scanning laser welding seams.

[0004] This invention adopts the following technical solution: a method for designing a ring-scanning laser welding process, comprising the following steps:

[0005] Step 1: Determine process design parameters and levels

[0006] The process design parameters and levels are determined based on the factors that affect weld formation;

[0007] Step 2: Select the experimental design type

[0008] Choose the appropriate experimental design type based on the number and complexity of the process design parameters selected by the user;

[0009] Step 3: Determine the response variable

[0010] Determine the response variable, which is a characteristic parameter of weld formation;

[0011] Step 4: Design the experimental matrix

[0012] Based on the experimental design type described in step two and the process design parameters and levels described in step one, an experimental matrix is ​​generated.

[0013] Step 5: Conduct process benchmark experiments

[0014] In the experimental matrix, 2 to 4 sets of experimental schemes are randomly selected to carry out process benchmark experiments for correcting the simulation calculation model, and the characteristic parameters of weld formation are obtained.

[0015] Step 6: Conduct simulation benchmark experiments

[0016] The simulation benchmark experiment process is as follows:

[0017] (1) Based on the same experimental scheme as the process benchmark experiment, simulation calculations were performed to obtain new characteristic parameters of weld formation;

[0018] The simulation calculation process is as follows:

[0019] 1) Input the process parameters in the experimental design;

[0020] 2) Calculate the trajectory of the laser beam center;

[0021]

[0022] Where x0 and y0 are the coordinates of the beam starting point, D is the scanning amplitude, f is the scanning frequency, and v W Let t be the welding speed and t be the welding time.

[0023] 3) Calculate the instantaneous laser energy distribution on the workpiece.

[0024]

[0025] Where η is the energy coefficient, P is the laser power, r0 is the spot radius, x(t) and y(t) are the laser beam centers, and Δt is the time step;

[0026] 4) Update the temperature of the workpiece surface and interior.

[0027] T n+1 =f S (f D (f A (T n ))) (3)

[0028] Among them, T n T represents the temperature at the current moment. n+1 f represents the temperature at the next moment. A Representative convection item update:

[0029]

[0030] in, The flow velocity of the molten pool is obtained from the following equation:

[0031]

[0032] in, For partial differential operators, T is the transpose operator. For divergence operators, Here, t is the gradient operator, and t is the welding time. Let ρ be the molten pool flow velocity, ρ and μ be the material density and viscosity, respectively, and p be the molten pool pressure. As an internal source of power;

[0033] f D Representative diffusion item update:

[0034]

[0035] Where ρ is the density of the material, C is the specific heat capacity of the material, and λ is the thermal conductivity of the material;

[0036] f S Representative source item update:

[0037]

[0038] 6) Extract new characteristic parameters of weld formation

[0039] Based on the isotherm of the material's melting point, new weld formation characteristic parameters are extracted;

[0040] (2) Correct the energy coefficient in the simulation model

[0041] The characteristic parameters of weld formation obtained from the process benchmark experiment are compared with the new characteristic parameters of weld formation obtained from the simulation benchmark experiment. Based on the comparison results, the energy coefficient in formula (2) is corrected so that the simulation benchmark experiment results match the process benchmark experiment results. Finally, the average value of the corrected energy coefficients of each group of the simulation benchmark experiment is taken as the final energy coefficient.

[0042] Step 7: Execute the remaining experimental schemes in the experimental matrix.

[0043] Based on the process parameters and corrected energy coefficients in the remaining experimental schemes in the experimental matrix, repeat the simulation calculation in step six and record the response variable values ​​for each experimental scheme.

[0044] Step 8: Data Analysis and Optimization

[0045] Experimental data were analyzed using statistical analysis methods to establish a relationship model between process design parameters and response variables. Based on the established model, parameter optimization was performed to find the optimal parameter combination and complete the process design.

[0046] Furthermore, in the first step, the process design parameters include laser power, welding speed, defocusing amount, laser scanning frequency, and scanning amplitude; the level refers to the reasonable design range of the process design parameters, which is related to the material, laser, and workpiece thickness.

[0047] Furthermore, in the second step, the experimental design types include full factorial design, partial factorial design, and Taguchi design.

[0048] Furthermore, in steps three and five, the characteristic parameters for weld formation include weld width and weld depth.

[0049] Furthermore, in the fourth step, the experimental matrix is ​​a series of experimental schemes composed of different combinations of process design parameters, and each experimental scheme provides the process parameters required to carry out ring scanning laser welding.

[0050] Furthermore, in the sixth step, the new characteristic parameters for weld formation include weld width and weld depth.

[0051] Furthermore, in step six, The internal force source includes the thermal buoyancy, thermocapillary force, and recoil pressure experienced by the molten pool.

[0052] Furthermore, in step eight, the statistical analysis methods include ANOVA and regression analysis.

[0053] Compared with existing methods, the advantages of the present invention are as follows:

[0054] (1) Low experimental cost. Traditional methods require a large number of process experiments, which consume a lot of manpower, material resources and financial resources. This method only requires a small number of experiments to correct the prediction model to complete the process design;

[0055] (2) High production efficiency. Traditional methods require many experiments and have long experimental cycles, which reduces production efficiency. In contrast, this method requires only a small number of experiments, has a short experimental cycle, and has high production efficiency.

[0056] (3) Good versatility. Traditional methods rely on material and workpiece thickness. If the material and workpiece thickness change, the relevant experiments need to be carried out again. However, this method only needs to modify the input parameters to complete the process design, which has good versatility. Attached Figure Description

[0057] Figure 1 This is a flowchart of a ring-scanning laser welding process design method.

[0058] Figure 2 This is a schematic diagram of the trajectory of the center of the laser beam.

[0059] Figure 3 This is a schematic diagram of the instantaneous laser energy distribution on a flat plate.

[0060] Figure 4 This is a schematic diagram of extracting new weld formation characteristic parameters from the isotherm of the material's melting point.

[0061] Figure 5 A schematic diagram of the weld depth-to-width ratio response surface obtained based on the response surface method under different scanning amplitudes and scanning frequencies.

[0062] Figure 6 This is a schematic diagram comparing the weld morphology using unoptimized and optimized process parameters. Detailed Implementation

[0063] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0064] Combination Figure 1 The ring scanning laser welding process design method of the present invention includes the following steps:

[0065] Step 1: Determine process design parameters and levels

[0066] The process design parameters and level are determined based on the factors that affect weld formation. The process design parameters may include laser power, welding speed, defocusing amount, laser scanning frequency, and scanning amplitude; the level refers to the reasonable design range of the process design parameters, which is related to the material, laser, and workpiece thickness.

[0067] Step 2: Select the experimental design type

[0068] Based on the number and complexity of the process design parameters selected by the user, an appropriate experimental design type is selected, including full factorial design, partial factorial design, Taguchi design, etc.

[0069] Step 3: Determine the response variable

[0070] In this invention, the response variables are characteristic parameters of weld formation, including weld width and weld depth.

[0071] Step 4: Design the experimental matrix

[0072] Based on the experimental design type selected by the user in step two and the process design parameters and levels in step one, an experimental matrix is ​​generated. The experimental matrix consists of a series of experimental schemes composed of different combinations of process design parameters, and each experimental scheme provides the process parameters required to carry out ring scanning laser welding.

[0073] Step 5: Conduct process benchmark experiments

[0074] In the experimental matrix, 2 to 4 sets of experimental schemes are randomly selected to carry out process benchmark experiments (to correct the simulation calculation model) and obtain the characteristic parameters of weld formation, including weld width and depth.

[0075] Step 6: Conduct simulation benchmark experiments

[0076] The simulation benchmark experiment process is as follows:

[0077] (1) Based on the same experimental scheme as the process benchmark experiment, simulation calculations were performed to obtain new characteristic parameters of weld formation, including weld width and depth.

[0078] The simulation calculation process is as follows:

[0079] 1) Input the process parameters in the experimental design;

[0080] 2) Calculate the trajectory of the laser beam center;

[0081]

[0082] Where x0 and y0 are the coordinates of the beam starting point, D is the scanning amplitude, f is the scanning frequency, and v W Let t be the welding speed and t be the welding time.

[0083] 3) Calculate the instantaneous laser energy distribution on the workpiece.

[0084]

[0085] Where η is the energy coefficient (initial value is 1), P is the laser power, r0 is the spot radius, x(t) and y(t) are the laser beam centers, and Δt is the time step.

[0086] 4) Update the temperature of the workpiece surface and interior.

[0087] T n+1 =f S (f D (f A (T n ))) (3)

[0088] Among them, T n T represents the temperature at the current moment. n+1 f represents the temperature at the next moment. A Representative convection item update:

[0089]

[0090] in, The flow velocity of the molten pool is obtained from the following equation:

[0091]

[0092] in, For partial differential operators, T is the transpose operator. For divergence operators, Here, t is the gradient operator, and t is the welding time. Let ρ be the molten pool flow velocity, ρ and μ be the material density and viscosity, respectively, and p be the molten pool pressure. Internal force sources, such as thermal buoyancy, thermocapillary force, and recoil pressure experienced by the molten pool.

[0093] f D Representative diffusion item update:

[0094]

[0095] Where ρ is the density of the material, C is the specific heat capacity of the material, and λ is the thermal conductivity of the material.

[0096] f S Representative source item update:

[0097]

[0098] 7) Extract the characteristic parameters of new weld formation

[0099] Based on the isotherm of the material's melting point, new weld formation characteristic parameters are extracted, including weld width and depth.

[0100] (2) Correct the energy coefficient in the simulation model

[0101] The characteristic parameters of weld formation obtained from the process benchmark experiment are compared with the new characteristic parameters of weld formation obtained from the simulation benchmark experiment. Based on the comparison results, the energy coefficient in formula (2) is corrected so that the simulation benchmark experiment results match the process benchmark experiment results. Finally, the average value of the corrected energy coefficients of each group in the simulation benchmark experiment is taken as the final energy coefficient.

[0102] Step 7: Execute the remaining experimental schemes in the experimental matrix.

[0103] Based on the process parameters and corrected energy coefficients in the remaining experimental schemes in the experimental matrix, repeat the simulation calculations in step six and record the response variable values ​​for each experimental scheme.

[0104] Step 8: Data Analysis and Optimization

[0105] Statistical analysis methods such as ANOVA and regression analysis are used to analyze experimental data, establish a relationship model between process design parameters and response variables, optimize parameters based on the established model, find the optimal parameter combination, and complete the process design.

[0106] In the manufacturing of conduits for launch vehicle propulsion systems, such as bellows and connecting rings, tungsten inert gas (TIG) welding is commonly used. This method results in a large heat-affected zone when welding thin-walled conduits, easily leading to coarse microstructure and decreased joint mechanical properties. Laser welding, with its concentrated energy density, can significantly improve the heat-affected zone and joint microstructure of conduit welding, but its tolerance for assembly gaps is poor. Circular scanning laser welding (SLS) can achieve rapid welding formation over a large area, effectively solving problems such as poor assembly gap tolerance and a large heat-affected zone. The following example of SLS welding a 0.4mm thick 1Cr18Ni9Ti stainless steel plate illustrates the implementation process of this invention.

[0107] Example

[0108] Step 1: Determine process design parameters and levels

[0109] Based on process experience and relevant literature, the design parameters and levels of the scanning laser welding process for 1Cr18Ni9Ti stainless steel flat plate were determined, as shown in Table 1.

[0110] Table 1. Design parameters and level of flatbed scanning laser welding process for 1Cr18Ni9Ti stainless steel

[0111] Serial Number Process design parameters (units) level 1 Laser power (kW) 0.5～2.0 2 Welding speed (m / min) 0.5～2.0 3 Defocusing amount (mm) -10～10 4 Scan width (mm) 0.5～3.0 5 Scan frequency (Hz) 50～200

[0112] Step 2: Select the experimental design type

[0113] To better examine the impact of different process parameters, a 4-level design is used in this example. As shown in Table 1, the experimental design in this example consists of 5 factors and 4 levels. To reduce the number of experiments and improve design efficiency, a Taguchi design is used in this example, and the experimental design table is shown in Table 2.

[0114] Table 21. Experimental Design for Flatbed Scanning Laser Welding of Cr18Ni9Ti Stainless Steel

[0115] factor name level A laser power 0.5，1.0，1.5，2.0 B Welding speed 0.5，1.0，1.5，2.0 C Defocus -10，-3，3，10 D Scan Amplitude 0.5，1.0，2.0，3.0 E Scan frequency 50，100，150，200

[0116] Step 3: Determine the response variable

[0117] In this invention, the response variables are characteristic parameters of weld formation, including weld width and weld depth.

[0118] Step 4: Design the experimental matrix

[0119] Based on the corresponding variables in Table 2 of the second step and the third step, the experimental matrix for this example is shown in Table 3.

[0120] Table 31 Cr18Ni9Ti Stainless Steel Flatbed Scanning Laser Welding Experimental Matrix

[0121] Experimental protocol laser power Welding speed Defocus Scan Amplitude Scan frequency Weld width Melting depth 1 0.5 0.5 -10 0.5 50 2 0.5 1.0 -3 1.0 100 3 0.5 1.5 3 2.0 150 4 0.5 2.0 10 3.0 200 5 1.0 0.5 -3 2.0 200 6 1.0 1.0 -10 3.0 150 7 1.0 1.5 10 0.5 100 8 1.0 2.0 3 1.0 50 9 1.5 0.5 3 3.0 100 10 1.5 1.0 10 2.0 50 11 1.5 1.5 -10 1.0 200 12 1.5 2.0 -3 0.5 150 13 2.0 0.5 10 1.0 150 14 2.0 1.0 3 0.5 200 15 2.0 1.5 -3 3.0 50 16 2.0 2.0 -10 2.0 100

[0122] Step 5: Conduct process benchmark experiments

[0123] In the experimental matrix (Table 3), 2-4 groups were randomly selected to conduct process benchmark experiments to correct the simulation calculation model and obtain characteristic parameters of weld formation, including weld width and depth. Test plates were cut using a shearing machine, with the burrs facing upwards during welding. The test plate dimensions were 0.4mm × 100mm × 300mm. Before welding, the material surface was cleaned with alcohol, and the material was clamped using a piano key fixture. A RAYCUS RFL-C6000W continuous fiber laser and a KUKAKR30HA robot were used to conduct scanning laser welding process experiments. The laser output wavelength was 1060-1070nm, the rated output power was 6000W, and the actual laser output power was controlled using analog signals. The system was equipped with a SCANLAB HURRYSCAN 30 two-dimensional scanning welding head, which was equipped with a transverse side-blowing air curtain, enabling conventional one-dimensional and two-dimensional scanning paths. The welding process used a dual-path shielding gas system: a side-axis tail shield and a back shield. The protective gas is argon with a purity of 99.99%, with a tail protective gas flow rate of 40-50 L / min and a back protective gas flow rate of 3-5 L / min.

[0124] Step 6: Conduct simulation benchmark experiments

[0125] The simulation benchmark experiment process is as follows:

[0126] (1) Based on the same experimental scheme as the process benchmark experiment, simulation calculations were performed to obtain new characteristic parameters of weld formation, including weld width and depth.

[0127] The simulation calculation process is as follows:

[0128] 1) Input the process parameters in the experimental design;

[0129] 2) Calculate the trajectory of the laser beam center;

[0130]

[0131] Where x0 and y0 are the coordinates of the beam starting point, D is the scanning amplitude, f is the scanning frequency, and v W Let t be the welding speed and t be the welding time. The trajectory of the laser beam center is as follows: Figure 2 As shown.

[0132] 3) Calculate the instantaneous laser energy distribution on the flat plate.

[0133]

[0134] Where η is the energy coefficient (initial value is 1), P is the laser power, r0 is the spot radius, x(t) and y(t) are the laser beam centers, and Δt is the time step. The instantaneous laser energy distribution on the plate is as follows: Figure 3 As shown.

[0135] 4) Update the temperature of the workpiece surface and interior.

[0136] T n+1 =f S (f D (f A (T n ))) (3)

[0137] Among them, T n T represents the temperature at the current moment. n+1 f represents the temperature at the next moment. A Representative convection item update:

[0138]

[0139] in, The flow velocity of the molten pool is obtained from the following equation:

[0140]

[0141] in, For partial differential operators, T For transpose operation, For divergence operators, Here, t is the gradient operator, and t is the welding time. Let ρ be the molten pool flow velocity, ρ and μ be the material density and viscosity, respectively, and p be the molten pool pressure. Internal force sources, such as thermal buoyancy, thermocapillary force, and recoil pressure experienced by the molten pool.

[0142] f D Representative diffusion item update:

[0143]

[0144] Where ρ is the density of the material, C is the specific heat capacity of the material, and λ is the thermal conductivity of the material.

[0145] f S Representative source item update:

[0146]

[0147] 8) Extract new characteristic parameters of weld formation

[0148] Based on the isotherm of the material's melting point, new weld formation characteristic parameters are extracted, including weld width and depth, such as... Figure 4 As shown.

[0149] (2) Correct the energy coefficient in the simulation model

[0150] The characteristic parameters of weld formation obtained from the process benchmark experiment are compared with the new characteristic parameters of weld formation obtained from the simulation benchmark experiment. Based on the comparison results, the energy coefficient in formula (2) is corrected so that the simulation benchmark experiment results match the process benchmark experiment results. Finally, the average value of the corrected energy coefficients of each group in the simulation benchmark experiment is taken as the final energy coefficient.

[0151] Step 7: Execute the remaining experimental schemes in the experimental matrix.

[0152] Based on the process parameters and corrected energy coefficients in the remaining experimental schemes in the experimental matrix, repeat the simulation calculation in step 6 and record the response variable values ​​for each experimental scheme, as shown in Table 4.

[0153] Table 4. Experimental Matrix of Flatbed Scanning Laser Welding for 1Cr18Ni9Ti Stainless Steel

[0154]

[0155]

[0156] Step 8: Data Analysis and Optimization

[0157] Statistical analysis methods such as ANOVA, regression analysis, and response surface methodology are used to analyze experimental data, establish a relationship model between process design parameters and response variables, optimize parameters based on the established model, find the optimal parameter combination, and complete the process design. Figure 5 The figure shows a schematic diagram of the weld depth-to-width ratio (ratio of weld depth to weld width) response surface obtained based on the response surface method under different scanning amplitudes and scanning frequencies. Figure 5 The scanning amplitude and scanning frequency corresponding to the highest point of the corresponding surface are the scanning amplitude and scanning frequency for obtaining the optimal aspect ratio. Figure 6 This is a schematic diagram comparing the weld morphology using unoptimized and optimized process parameters. The optimized weld has a smooth surface and a full shape.

Claims

1. A method for designing a ring-scanning laser welding process, characterized in that, Includes the following steps: Step 1: Determine process design parameters and levels The process design parameters and levels are determined based on the factors that affect weld formation; Step 2: Select the experimental design type Choose the appropriate experimental design type based on the number and complexity of the process design parameters selected by the user; Step 3: Determine the response variable Determine the response variable, which is a characteristic parameter of weld formation; Step 4: Design the experimental matrix Based on the experimental design type described in step two and the process design parameters and levels described in step one, an experimental matrix is ​​generated. Step 5: Conduct process benchmark experiments In the experimental matrix, 2 to 4 sets of experimental schemes are randomly selected to carry out process benchmark experiments for correcting the simulation calculation model and to obtain the characteristic parameters of weld formation. Step 6: Conduct simulation benchmark experiments The simulation benchmark experiment process is as follows: (1) Based on the same experimental scheme as the process benchmark experiment, simulation calculations were performed to obtain new characteristic parameters of weld formation; (2) Correct the energy coefficient in the simulation model The characteristic parameters of weld formation obtained from the process benchmark experiment are compared with the new characteristic parameters of weld formation obtained from the simulation benchmark experiment. Based on the comparison results, the energy coefficient in formula (2) is corrected so that the simulation benchmark experiment results match the process benchmark experiment results. Finally, the average value of the corrected energy coefficients of each group of the simulation benchmark experiment is taken as the final energy coefficient. Step 7: Execute the remaining experimental schemes in the experimental matrix. Based on the process parameters and corrected energy coefficients in the remaining experimental schemes in the experimental matrix, repeat the simulation calculation in step six and record the response variable values ​​for each experimental scheme. Step 8: Data Analysis and Optimization Statistical analysis methods are used to analyze experimental data, establish a relationship model between process design parameters and response variables, optimize parameters based on the established model, find the optimal parameter combination, and complete the process design. In the first step, the process design parameters include laser power, welding speed, defocusing amount, laser scanning frequency, and scanning amplitude; the level refers to the reasonable design range of the process design parameters, which is related to the material, laser, and workpiece thickness. In steps three and five, the characteristic parameters for weld formation include weld width and weld depth. In the sixth step, the new characteristic parameters for weld formation include weld width and weld depth; In step six, the simulation calculation process is as follows: 1) Input the process parameters in the experimental design; 2) Calculate the trajectory of the laser beam center; (1)； Where x0 and y0 are the coordinates of the beam starting point, D is the scanning amplitude, f is the scanning frequency, and v W Let t be the welding speed and t be the welding time. 3) Calculate the instantaneous laser energy distribution on the workpiece. (2)； Where η is the energy coefficient, P is the laser power, r0 is the spot radius, x(t) and y(t) are the laser beam centers, and Δt is the time step; 4) Update the temperature of the workpiece surface and interior. (3)； Among them, T n T represents the temperature at the current moment. n+1 f represents the temperature at the next moment. A Representative convection item update: (4)； in, The flow velocity of the molten pool is obtained from the following equation: (5)； in, For partial differential operators, T For transpose operation, For divergence operators, Here, t is the gradient operator, and t is the welding time. Let ρ be the molten pool flow velocity, ρ and μ be the material density and viscosity, respectively, and p be the molten pool pressure. As an internal source of power; f D Representative diffusion item update: (6)； Where ρ is the density of the material, C is the specific heat capacity of the material, and λ is the thermal conductivity of the material; f S Representative source item update: (7)； 5) Extract new characteristic parameters of weld formation New weld formation characteristic parameters are extracted based on the isotherm of the material's melting point.

2. The method as described in claim 1, characterized in that, In the second step, the experimental design types include full factorial design, partial factorial design, and Taguchi design.

3. The method as described in claim 1, characterized in that, In the fourth step, the experimental matrix is ​​a series of experimental schemes composed of different combinations of process design parameters. Each experimental scheme provides the process parameters required to carry out ring scanning laser welding.

4. The method as described in claim 1, characterized in that, The internal force source includes the thermal buoyancy, thermocapillary force, and recoil pressure experienced by the molten pool.

5. The method as described in claim 1, characterized in that, In step eight, the statistical analysis methods include ANOVA and regression analysis.

Images