Double-robot laser arc hybrid welding real-time regulation and control method based on digital twinning

Abstract

The invention belongs to the technical field of intelligent manufacturing and laser arc hybrid welding, and discloses a double-robot laser arc hybrid welding real-time regulation and control method based on digital twinning. According to the method, double robots are used, a laser-electric arc composite heat source digital twinborn model is constructed, a Gaussian distribution laser heat source and a double-ellipsoid electric arc heat source are integrated, and the dynamic behaviors (molten pool flowing and keyhole stability) of a molten pool and the generation probability of defects (air holes, cracks and the like) are predicted. And a visual sensor and an acoustic emission sensor on the two sides of the welding robot are used for capturing molten pool oscillation frequency and internal defect signals. And in combination with a reinforcement learning algorithm, welding path offset self-adaptive compensation and laser arc energy ratio adjustment are achieved, and closed-loop real-time regulation and control are formed by controlling the position of a welding gun of a welding robot and technological parameters. Based on a digital twinborn model, robot welding guns on the two sides can conduct synchronous welding on the two sides and can be independently adjusted while coupling is guaranteed according to the characteristics of specific materials, and high-quality welding is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent manufacturing and laser arc hybrid welding, and in particular to a real-time control method for dual-robot laser arc hybrid welding based on digital twins. Background Art

[0002] Laser-arc hybrid welding technology combines the high energy density and high precision of laser welding with the deep melting ability and adaptability of arc welding. It has technical advantages such as high speed, high efficiency, low deformation, single-sided welding and double-sided forming, and can solve the problems of large heat-affected zone and high residual stress in traditional arc welding. Compared with a single welding method, laser arc hybrid welding has special advantages. Its core mechanism lies in the synergy of laser and arc, and the control of the welding process is achieved through thermal field coupling and regulation of the dynamic behavior of the molten pool. However, in the traditional laser-arc hybrid welding process, the energy ratio of laser and arc is fixed, and digital twins are not combined to achieve dynamic coordination of the entire process. It cannot be adjusted in real time according to the dynamic behavior of the molten pool, and there are problems such as low laser energy utilization and high defect rate.

[0003] Double-sided welding balances the welding heat input through the synchronous action of double-sided heat sources, reducing the deformation and residual stress caused by heat accumulation on one side. For example, the lateral shrinkage deformation and longitudinal inherent deformation of T-joint double-sided synchronous welding are more evenly distributed, and the stress concentration problem in the weld area is significantly improved. However, the synchronization and coupling of double-sided welding has always been a difficulty.

[0004] Digital twins achieve process optimization, predictive maintenance and other functions by establishing a virtual mapping of the physical system and combining real-time data with simulation models. In the field of welding, its applications focus on welding process simulation, defect prediction, parameter optimization and other directions. At present, the combination of existing digital twin technology and welding technology is mainly reflected in welding quality inspection and facilitating workers to find appropriate process parameters to achieve efficient welding, but it still has a certain degree of non-autonomy, and cannot quickly and effectively adjust the welding process based on digital twin models and algorithms in real time, and cannot achieve dynamic collaborative control of the entire process. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a real-time collaborative control method for laser-arc hybrid welding process based on digital twin, which belongs to the field of intelligent manufacturing and laser-arc hybrid welding technology. This method aims at the problems of insufficient energy coordination, process parameter fragmentation and high defect rate in traditional laser-arc hybrid welding. By constructing a cross-scale digital twin model, the composite heat source characteristics of laser and arc are deeply integrated to realize adaptive control and process parameter optimization of the welding process. The system includes a composite heat source digital twin modeling module, a multi-physics field coupling simulation module and a real-time collaborative control module. Among them, the composite heat source modeling adopts Gaussian distribution laser heat source and double ellipsoid arc heat source, and dynamically optimizes the energy ratio of laser and arc in combination with quantum annealing algorithm to match the heat demand of molten pool in real time. Multi-physics field simulation establishes a heat-force-flow coupling model based on MARC software, captures the oscillation frequency and internal defect signals of the molten pool through acoustic emission sensors and terahertz imagers, analyzes the flow of the molten pool, keyhole stability and solidification behavior, and predicts the probability of pore and crack defect generation. Based on the above signal feedback, the real-time control module combines the reinforcement learning algorithm to dynamically adjust the welding path offset and energy parameters to form a closed-loop real-time control. The process stability and quality control capability of laser-arc hybrid welding are significantly improved, and it is suitable for complex working conditions such as thin-walled aerospace aluminum alloys and thick high-strength steel plates for ships. To achieve the above purpose, the present invention adopts the following technical solutions: A real-time control method for dual-robot laser arc hybrid welding based on digital twins is characterized by being implemented by the following steps: Step 1: Build a cross-scale digital twin model to predict the microstructure evolution and macro deformation of the weld through molecular dynamics and thermal-mechanical coupled finite element analysis; Step 2: During the welding process, the visual sensor and acoustic emission sensor of the welding robot are used to obtain the internal defect signals of the molten pool and the phase change characteristics of the material in real time, and the data are fed back to the computer in real time; Step 3: Based on the federated learning framework and multi-site database, the computer forms a digital twin simulation comparison matching relationship for the defect signal and the digital twin prediction model, and optimizes the process parameter combination through the quantum annealing algorithm; Step 4: The computer control system dynamically allocates welding tasks through the intelligent control cabinet, controls the welding laser, welding motor and welding robot to execute coordinated control instructions, and adopts the best welding process parameters in real time to achieve high-quality welding.

[0006] Furthermore, the digital twin model is established using SolidWorks professional 3D design software to construct a lightweight 3D model of the welded workpiece. The thermal-mechanical coupling finite element analysis technology uses a multi-physics field simulation module, integrates the thermal-mechanical coupling model and the material library, and calculates the dynamic behavior of the molten pool and welding deformation.

[0007] Furthermore, the multi-physics simulation module establishes a laser-arc composite heat source model, wherein the laser heat source model adopts a Gaussian distribution heat source, and the arc heat source model adopts a double ellipsoid heat source model. In addition, the actual welding process can be simulated as perfectly as possible by setting up boundary conditions, including heat dissipation, displacement constraints and other boundary conditions. The welding process parameters include laser focus diameter, arc effective heating radius, etc., and variables such as molten pool flow and residual stress distribution are calculated by a nonlinear solver and output to the digital twin database.

[0008] Furthermore, the establishment of the digital twin model includes the following small steps: SolidWorks was used to build a one-to-one model of the weldment and to mesh the model.

[0009] Add material properties. Material properties include mass density, Young's modulus, Poisson's ratio, specific heat capacity, thermal conductivity and other characteristics. The more detailed the material properties are, the higher the accuracy of the digital twin model.

[0010] Set the welding path and weld, including arc starting point, arc ending point, welding direction, etc.

[0011] Set the initial conditions and boundary conditions. The initial conditions include ambient temperature, weldment position, initial stress, etc. The boundary conditions need to be set according to the specific welding process. The heat source model of laser arc hybrid welding usually adopts a Gaussian distribution heat source and a double ellipsoid heat source dual heat source composite model; in addition to the heat source, boundary conditions that need to be added include heat dissipation coefficient, displacement constraint, welding speed, etc., which need to be flexibly set according to the actual situation.

[0012] The above conditions are checked and if everything is correct, simulation is carried out to predict the microstructure evolution and macro deformation of the weld through molecular dynamics and thermal-mechanical coupling finite element analysis.

[0013] Furthermore, the visual sensor can detect the temperature distribution and thermal changes in the welding area in real time, and feed back the dynamic changes of the depth, shape, flow conditions, etc. of the molten pool to the computer in real time; the acoustic emission sensor can analyze the acoustic emission signals during the welding process, and can detect and feed back potential defects such as pores, slag inclusions, cracks, etc.

[0014] Furthermore, after receiving the signal from the sensor, the computer can match the feedback defect signal with the digital twin model, and adjust the process parameters and the robot motion trajectory based on the prediction of the digital twin model, combined with the database and reinforcement learning algorithm.

[0015] Furthermore, the reinforcement learning algorithm adjusts the process parameters such as laser power, arc current and welding speed, and selects the optimal parameter combination through the search algorithm and welding database to match the heat input demand of the molten pool in real time. For T-joints, the molten pools formed on both sides of the double-sided robot during the welding process are not completely consistent. Therefore, the process parameters used by the robot are not the same for different molten pool morphologies. It is necessary to compare with the digital twin model and use the optimization algorithm to obtain the process parameters of the robots on both sides, and adjust the process in real time, which can effectively suppress the formation of weld defects and promote the effective fusion of materials.

[0016] Furthermore, the establishment of the process parameters and the adjustment of the motion trajectory of the double-sided welding robot are specifically achieved in the following manner.

[0017] 1) Extraction of dynamic features of the molten pool. Based on the infrared thermal image sequence collected by the visual sensor, the molten pool depth-to-width ratio λ=W / L and the temperature gradient ∇T are extracted; the acoustic emission signal extracts the energy entropy E as a defect probability indicator; 2) Detection of matching degree between digital twin model and actual process.

[0018] Define the difference function: D = α | λ 1 -λ 0 |+β|∇T 1 -∇T 0 |+γE Among them. 1 is the depth-to-width ratio of the molten pool in the theoretical model, λ 0 is the depth-to-width ratio of the molten pool during the actual welding process, ∇T 1 is the theoretical temperature gradient, ∇T 0 is the actual temperature gradient. E is used as a defect probability indicator. For example, when E=1, there must be defects in the welding process. α, β, γ are weight coefficients, α+β+γ=1. The weight ratios need to be determined by material properties, principal component analysis, and actual welding process requirements. In general, 0.04≤α≤0.05, β≤0.01, γ≥0.95. Definition D normal ≤0.4, that is, when the difference function D≤0.4, it can be determined that the actual welding process is basically consistent with the prediction of the digital twin model.

[0019] 3) Trigger parameter adjustment.

[0020] When the difference function D>0.4, it is determined that there are obvious deficiencies in the actual welding process and the process parameters need to be adjusted in time. At this time, the balance function is defined as R = k 1 D + k 2 ∣Δ v ∣+ k3 ∣Δ P 1 ∣+ k 4 ∣Δ P 2 ∣ Where D is the difference value, Δ v is the difference between welding speed before and after adjustment, Δ P 1 is the difference between laser power before and after adjustment, Δ P 1 It is the difference between arc power before and after adjustment. Arc power P=UI, which is the product of arc voltage and welding current. k 1 , k 2 , k 3 , k 4 is the weight coefficient, k 1 + k 2 + k 3 + k 4 =1, the weight proportions need to be determined by material properties, principal component analysis and actual welding process requirements. Even for the same defect or to improve the molten pool, there are many ways to adjust, and the process parameters can change in coordination. Therefore, the reference to the balance function can limit the adjustment range of each process parameter, thereby avoiding new welding problems caused by excessive parameter changes. In the process of parameter adjustment, a variety of improvement schemes are first provided based on the welding database through the reinforcement learning algorithm, and then the parameter fluctuation values ​​in each scheme are introduced into the balance function, and a variety of R values ​​can be obtained. When the R value of the balance function is the smallest, the corresponding scheme can be used as the optimal solution.

[0021] 4) Robotic arm trajectory correction.

[0022] Robot arm motion trajectory compensation formula: Δ(x, y, z) = δ·cosθ(x, y, z) + K p ∫δdt Where δ is the real-time offset between the center of the molten pool and the theoretical position of the weld (measured by the visual sensor), θ(x, y, z) is the angle between the welding direction and the X, Y, and Z axes of the robot coordinate system, and K p is the integral gain coefficient, which determines the correction strength of the historical error, and ∫δdt represents the cumulative effect of the offset over time, which is used to quantify the long-term systematic error. δ⋅cosθ(x, y, z) can calculate the lateral compensation (quick response to instantaneous deviation) based on the current molten pool offset δ and the welding direction angle θ.

[0023] Furthermore, the computer control system combines the digital twin simulation model and prediction results, detection feedback signals, and algorithm-optimized process parameters, welding trajectory correction parameters, etc., to issue control instructions to systems such as the laser generator, welding power supply, welding robot, and wire feeder. The systems work together and adjust in real time to ultimately achieve closed-loop control of the welding process and complete high-quality and efficient adaptive welding.

[0024] The advantages and benefits of the present invention are as follows: The present invention provides a real-time control method for dual-robot laser arc hybrid welding based on digital twins. By constructing a cross-scale digital twin model, a multi-physical field coupling simulation of microscopic molten pool flow and macroscopic weld formation is realized, and real-time feedback of visual and acoustic emission signals is combined to actively suppress typical defects of composite welding such as pores and cracks. At the same time, in order to solve the problem of mismatch of bilateral process parameters during bilateral robot welding of T-joints, digital twin technology is used to realize the interaction between welding process parameters and model simulation, and to realize real-time control of process parameters such as welding path adaptive compensation, laser-arc energy ratio adjustment, and welding speed. The present invention can realize adaptive closed-loop control, and provides a fully software-defined intelligent solution for precision welding in the fields of aerospace, new energy vehicles, etc., with broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of a real-time control method of dual-robot laser arc hybrid welding process based on digital twin; Figure 2 It is a digital twin platform for real-time control of dual-robot laser arc hybrid welding process based on digital twin; Figure 3 It is a real-time control flow chart of dual-robot laser arc hybrid welding process based on digital twin; Figure 4 This is an example diagram of double-sided robot welding of aluminum alloy T-joints. DETAILED DESCRIPTION

[0026] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific implementation cases described herein are only used to explain the present invention, rather than to limit the present invention. It is also necessary to explain that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0027] In the description of the present invention, unless otherwise clearly specified and limited, the terms "connected", "connected", and "fixed" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0028] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly below and obliquely below the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0029] In the description of this embodiment, the terms "upper", "lower", "right", etc., directions or positional relationships are based on the directions or positional relationships shown in the drawings, and are only for the convenience of description and simplification of operation, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used to distinguish in the description and have no special meaning.

[0030] The present invention is further described below by a specific embodiment. This embodiment is a welding process for laser arc hybrid welding of a frame-truss rocket tank wall plate structure, the rocket tank wall plate is a T-shaped structure, and the selected material is 2219 aluminum alloy.

[0031] A real-time control method for dual-robot laser arc hybrid welding based on digital twins, such as Figure 1 As shown, it mainly includes 1) robot welding system, 2) digital twin simulation system, 3) monitoring system (including visual sensor and acoustic emission sensor), and 4) control system.

[0032] Preferably, if Figure 2 As shown, a real-time control method for dual-robot laser arc hybrid welding based on digital twin is specifically implemented by the following steps: Step 1: Build a cross-scale digital twin model to predict the microstructure evolution and macro deformation of the weld through molecular dynamics and thermal-mechanical coupled finite element analysis; Step 2: During the welding process, the visual sensor and acoustic emission sensor of the welding robot are used to obtain the internal defect signals of the molten pool and the phase change characteristics of the material in real time, and the data are fed back to the computer in real time; Step 3: Based on the federated learning framework and multi-site database, the computer forms a digital twin simulation comparison matching relationship for the defect signal and the digital twin prediction model, and optimizes the process parameter combination through the quantum annealing algorithm; Step 4: The computer control system dynamically allocates welding tasks through the intelligent control cabinet, controls the welding laser, welding motor and welding robot to execute coordinated control instructions, and adopts the best welding process parameters in real time to achieve high-quality welding.

[0033] Preferably, the digital twin model is established using SolidWorks professional 3D design software to construct a lightweight 3D model of the rocket tank wall panel structure. The thermal-mechanical coupling finite element analysis technology uses a multi-physics field simulation module, integrates the thermal-mechanical coupling model and the material library, and calculates the dynamic behavior of the molten pool and welding deformation.

[0034] Preferably, the multi-physics simulation module establishes a laser-arc composite heat source model, wherein the laser heat source model adopts a Gaussian distribution heat source, and the arc heat source model adopts a double ellipsoid heat source model. In addition, the actual welding process can be simulated as perfectly as possible by setting up boundary conditions, including heat dissipation, displacement constraints and other boundary conditions. Welding process parameters include laser focus diameter, arc effective heating radius, etc., and variables such as molten pool flow and residual stress distribution are calculated by a nonlinear solver and output to the digital twin database.

[0035] Preferably, the digital twin model establishment includes the following small steps: 1) Use SolidWorks to build a one-to-one model of the weldment and divide the model into meshes.

[0036] 2) Add material properties. Material properties include mass density, Young's modulus, Poisson's ratio, specific heat capacity, thermal conductivity and other characteristics. The more detailed the material properties are, the higher the accuracy of the digital twin model.

[0037] 3) Set the welding path and weld, including arc starting point, arc ending point, welding direction, etc. For the double-sided welding of the T-shaped structure of the rocket tank wall, two welds and two welding paths need to be set.

[0038] 4) Set the initial conditions and boundary conditions. The initial conditions include ambient temperature, weldment position, initial stress, etc. Boundary conditions need to be set according to the specific welding process. The heat source model of laser arc hybrid welding usually adopts a Gaussian distribution heat source and a double ellipsoid heat source dual heat source composite model; in addition to the heat source, boundary conditions that need to be added include heat dissipation coefficient, displacement constraint, welding speed, etc., which need to be flexibly set according to actual conditions.

[0039] 5) Check the above conditions. If everything is correct, perform simulation and predict the microstructure evolution and macro deformation of the weld through molecular dynamics and thermal-mechanical coupling finite element analysis.

[0040] Preferably, the visual sensor can detect the temperature distribution and thermal changes in the welding area in real time, and feed back the dynamic changes of the depth, shape, flow conditions, etc. of the molten pool to the computer in real time; the acoustic emission sensor can analyze the acoustic emission signals during the welding process, and can detect and feed back potential defects such as pores, slag inclusions, cracks, etc.

[0041] Preferably, after receiving the signal from the sensor, the computer can match the feedback defect signal with the digital twin model, and adjust the process parameters and the robot motion trajectory based on the prediction of the digital twin model in combination with the database and reinforcement learning algorithm.

[0042] Preferably, the reinforcement learning algorithm adjusts the process parameters such as laser power, arc current and welding speed, selects the optimal parameter combination through the search algorithm and the welding database, and matches the heat input demand of the molten pool in real time. For T-joints, the molten pools formed on both sides of the double-sided robot during the welding process on both sides are not completely consistent. Therefore, for different molten pool morphologies, the process parameters used by the robots are also different. It is necessary to compare with the digital twin model and use the optimization algorithm to obtain the process parameters of the robots on both sides, and adjust the process in real time, which can effectively suppress the formation of weld defects and promote the effective fusion of materials.

[0043] Preferably, the establishment of the process parameters and the adjustment of the motion trajectory of the double-sided welding robots are specifically achieved in the following manner.

[0044] 1) Extraction of dynamic features of the molten pool. Based on the infrared thermal image sequence collected by the visual sensor, the molten pool depth-to-width ratio λ=W / L and the temperature gradient ∇T are extracted; the acoustic emission signal extracts the energy entropy E as a defect probability indicator; 2) Detection of matching degree between digital twin model and actual process.

[0045] Define the difference function: D = α | λ 1 -λ 0 |+β|∇T 1 -∇T 0 |+γE Among them. 1 is the depth-to-width ratio of the molten pool in the theoretical model, λ 0 is the depth-to-width ratio of the molten pool during the actual welding process, ∇T 1 is the theoretical temperature gradient, ∇T 0 is the actual temperature gradient. E is used as a defect probability indicator. For example, when E=1, there must be defects in the welding process. α, β, γ are weight coefficients, α+β+γ=1. The weight ratios need to be determined by material properties, principal component analysis, and actual welding process requirements. In general, 0.04≤α≤0.05, β≤0.01, γ≥0.95. Definition D normal ≤0.4, that is, when the difference function D≤0.4, it can be determined that the actual welding process is basically consistent with the prediction of the digital twin model.

[0046] 3) Trigger parameter adjustment.

[0047] When the difference function D>0.4, it is determined that there are obvious deficiencies in the actual welding process and the process parameters need to be adjusted in time. At this time, the balance function is defined as R = k 1 D + k 2 ∣Δ v ∣+ k 3 ∣Δ P 1 ∣+ k 4 ∣Δ P 2 ∣ Where D is the difference value, Δ v is the difference between welding speed before and after adjustment, Δ P 1 is the difference between laser power before and after adjustment, Δ P 1 It is the difference between arc power before and after adjustment. Arc power P=UI, which is the product of arc voltage and welding current. k 1 , k 2 , k 3 , k 4 is the weight coefficient, k 1 + k 2 + k 3 + k 4=1, the weight proportions need to be determined by material properties, principal component analysis and actual welding process requirements. Even for the same defect or to improve the molten pool, there are many ways to adjust, and the process parameters can change in coordination. Therefore, the reference to the balance function can limit the adjustment range of each process parameter, thereby avoiding new welding problems caused by excessive parameter changes. In the process of parameter adjustment, a variety of improvement schemes are first provided based on the welding database through the reinforcement learning algorithm, and then the parameter fluctuation values ​​in each scheme are introduced into the balance function, and a variety of R values ​​can be obtained. When the R value of the balance function is the smallest, the corresponding scheme can be used as the optimal solution.

[0048] 4) Robotic arm trajectory correction.

[0049] Robot arm motion trajectory compensation formula: Δ(x, y, z) = δ·cosθ(x, y, z) + K p ∫δdt Where δ is the real-time offset between the center of the molten pool and the theoretical position of the weld (measured by the visual sensor), θ(x, y, z) is the angle between the welding direction and the X, Y, and Z axes of the robot coordinate system, and K p is the integral gain coefficient, which determines the correction strength of the historical error, and ∫δdt represents the cumulative effect of the offset over time, which is used to quantify the long-term systematic error. δ⋅cosθ(x, y, z) can calculate the lateral compensation (quick response to instantaneous deviation) based on the current molten pool offset δ and the welding direction angle θ.

[0050] Preferably, the computer control system combines the digital twin simulation model and prediction results, detection feedback signals, and algorithm-optimized process parameters, welding trajectory correction parameters, etc., to issue control instructions to systems such as the laser generator, welding power supply, welding robot, and wire feeder. The systems work together and adjust in real time to ultimately achieve closed-loop control during the welding process and complete high-quality and efficient adaptive welding.

[0051] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, various obvious changes, readjustments and substitutions can be made without departing from the protection scope of the present invention. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A real-time control method for dual-robot laser arc hybrid welding based on digital twin, characterized in that: The specific steps include: Step 1: Build a cross-scale digital twin model to predict the microstructure evolution and macro deformation of the weld through molecular dynamics and thermal-mechanical coupled finite element analysis; Step 2: During the welding process, the visual sensor and acoustic emission sensor of the welding robot are used to obtain the internal defect signals of the molten pool and the phase change characteristics of the material in real time, and the data are fed back to the computer in real time; Step 3: Based on the federated learning framework and multi-site database, the defect signals and digital twin prediction model are used to form a digital twin simulation comparison matching relationship, and the process parameter combination is optimized through the quantum annealing algorithm; Step 4: The control system dynamically allocates welding tasks through the intelligent control cabinet, controls the welding laser, welding motor and welding robot to execute coordinated control instructions, and adopts the best welding process parameters in real time to achieve high-quality welding.

2. According to a real-time control method for dual-robot laser arc hybrid welding based on digital twinning according to claim 1, it is characterized in that: The digital twin model is established using SolidWorks professional 3D design software to construct a lightweight 3D model of the welded workpiece; the thermal-mechanical coupling finite element analysis technology uses a multi-physics field simulation module, integrating the thermal-mechanical coupling model and the material library to calculate the dynamic behavior of the molten pool and welding deformation.

3. The real-time control method for dual-robot laser arc hybrid welding based on digital twin according to claim 2 is characterized in that: The multi-physics field simulation module establishes a laser-arc composite heat source model, wherein the laser heat source model adopts a Gaussian distribution heat source, and the arc heat source model adopts a double ellipsoid heat source model. In addition, by setting up boundary conditions, the actual welding process can be simulated as perfectly as possible, including a variety of boundary conditions for heat dissipation and displacement constraints; the welding process parameters include the laser focus diameter and the effective heating radius of the arc, and the molten pool flow and residual stress distribution variable parameters are calculated by a nonlinear solver and output to the digital twin database.

4. The real-time control method for dual-robot laser arc hybrid welding based on digital twin according to claim 1 is characterized in that: The visual sensor can detect the temperature distribution and thermal changes in the welding area in real time, and feed back the dynamic changes of the depth, shape, flow conditions, etc. of the molten pool to the computer in real time; the acoustic emission sensor can analyze the acoustic emission signals during the welding process, and can detect and feed back potential defects such as pores, slag inclusions, and cracks.

5. The real-time control method for dual-robot laser arc hybrid welding based on digital twin according to claim 1 is characterized in that: After receiving the signal from the sensor, the computer matches the feedback defect signal with the digital twin model, and adjusts the process parameters and the robot motion trajectory based on the prediction of the digital twin model, combined with the database and reinforcement learning algorithm.

6. The real-time control method for dual-robot laser arc hybrid welding based on digital twin according to claim 5 is characterized in that: The reinforcement learning algorithm is specifically: The welding process is set for the T-joint. The bilateral robots perform welding on both sides. The process parameters of laser power, arc current and welding speed are adjusted. The optimal parameter combination is screened through the search algorithm and the welding database to obtain the process parameters of the robots on both sides.

7. The real-time control method for dual-robot laser arc hybrid welding based on digital twin according to claim 6 is characterized in that: The establishment of process parameters and the adjustment of the motion trajectory of the double-sided welding robot are specifically achieved in the following ways: 1) Based on the infrared thermal image sequence collected by the visual sensor, the molten pool depth-to-width ratio λ=W / L and the temperature gradient ∇T are extracted; the energy entropy E is extracted from the acoustic emission signal as a defect probability indicator; 2) Define the difference function: D = α|λ1-λ0|+β|∇T1-∇T0|+γE; Among them, λ1 is the molten pool depth-to-width ratio of the theoretical model, λ0 is the molten pool depth-to-width ratio in the actual welding process, ∇T1 is the theoretical temperature gradient, and ∇T0 is the actual temperature gradient; E is the defect probability index, α, β, γ are weight coefficients, α+β+γ=1, and the proportion of each weight needs to be determined by material properties, principal component analysis and actual welding process requirements; 3) When the difference function D> the set value, it is determined that there are obvious deficiencies in the actual welding process and the process parameters need to be adjusted in time, so the balance function is defined R = k 1 D + k 2|Δ v ∣+ k 3|Δ P 1|+ k 4|Δ P 2∣; Where D is the difference value, Δ v is the difference between welding speed before and after adjustment, Δ P 1 is the difference between the laser power before and after adjustment, Δ P 1 is the difference between arc power before and after adjustment. Arc power P=UI, which is the product of arc voltage and welding current. k 1, k 2, k 3. k 4 is the weight coefficient, k 1+ k 2+ k 3+ k 4=1, the weight ratios shall be determined by material properties, principal component analysis and actual welding process requirements; In the process of parameter adjustment, a variety of improvement schemes are first provided through reinforcement learning algorithm based on welding database, and then the parameter fluctuation values ​​in each scheme are introduced into the balance function to obtain a variety of balance function R values. When the balance function R value is the smallest, the corresponding scheme is taken as the optimal solution. 4) Set the robot arm motion trajectory compensation formula to: Δ(x, y, z) = δ·cosθ(x, y, z) + K p ∫δdt; Where δ is the real-time offset between the center of the molten pool and the theoretical position of the weld, θ(x, y, z) is the angle between the welding direction and the X, Y, and Z axes of the robot coordinate system, and K p is the integral gain coefficient, which determines the correction strength of the historical error. ∫δdt represents the cumulative effect of the offset over time and is used to quantify the long-term systematic error. δ⋅cosθ(x, y, z) can calculate the lateral compensation amount based on the current molten pool offset δ and the welding direction angle θ.

8. The real-time control method for dual-robot laser arc hybrid welding based on digital twin according to claim 7 is characterized in that: The control system combines the digital twin simulation model and prediction results, detection feedback signals, and algorithm-optimized process parameters and welding trajectory correction parameters to issue control instructions to systems such as the laser generator, welding power supply, welding robot, and wire feeder.

Images