Simulation prediction and parameter optimization method for welding residual deformation of structural welding part

Abstract

The invention provides a welding residual deformation simulation prediction and parameter optimization method for a structural welding part, which comprises the following steps of: firstly, establishing a three-dimensional model of the structural welding part, and performing simulation analysis on a welding temperature field, welding residual stress and welding deformation of the three-dimensional model by using a professional simulation analysis tool; secondly, the influence of clamping conditions, heat source forms and welding parameters on welding deformation of the structural welding part is researched, and simulation and physical verification is achieved. And then, by taking the minimum welding residual deformation as an optimization target, optimizing the welding sequence by adopting a genetic algorithm and taking the welding sequence and the welding direction as design variables to obtain a group of optimal solutions, and obtaining the welding sequence with the minimum welding deformation through simulation. And finally, on the basis of a welding sequence optimization result, performing simulation optimization by utilizing tightening position adjustment and reversible deformation design in sequence to obtain a process design scheme with minimum welding deformation.

Description

Technical Field

[0001] The invention relates to the field of welding technology, and more particularly to a method for simulating and predicting welding residual deformation and optimizing parameters of a structural welded part. Background Art

[0002] In today's engineering technology, welding technology, as an efficient and economical metal connection method, has been applied in many key industries such as energy development, ocean exploration, aerospace, and petrochemicals, greatly promoting technological progress and industrial upgrading in these fields. Welded structures have become an indispensable part of building complex engineering systems with their good mechanical properties, structural flexibility and cost-effectiveness.

[0003] However, with the expansion of the application scope of welded structures and increasingly stringent engineering requirements, the deformation problem generated during welding has become increasingly prominent and has become one of the key factors restricting the performance and reliability of welded structures. Welding deformation, including angular deformation, bending deformation, shrinkage deformation and twisting deformation, not only directly affects the overall shape and dimensional accuracy of the structural parts, leading to difficulties in subsequent assembly, but may also induce serious defects such as cracks and fractures due to stress concentration, greatly weakening the bearing capacity and service life of the structure and increasing the risk of failure.

[0004] In addition, welding deformation will accumulate and propagate with errors in other links such as manufacturing and processing during the assembly process, forming a complex assembly error chain, further reducing the overall quality and performance stability of the product. Therefore, how to fully consider and effectively control welding deformation in the design stage has become an important research topic to improve the comprehensive performance of welded structures and ensure engineering safety. Summary of the invention In view of the problems existing in the prior art, the purpose of the present invention is to provide a method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts. The finite element method is used to perform residual deformation prediction simulation and parameter optimization of structural welded parts to control welding deformation and achieve optimal welding accuracy, thereby establishing a technical route and process for welding deformation simulation and parameter optimization of structural welded parts, which can effectively shorten the production cycle of parts, reduce the cost of trial and error, and improve the welding quality of parts.

[0005] In order to achieve the above object, the present invention is implemented through the following technical solutions: A method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts, comprising: Establish a three-dimensional model of structural welded parts for welding residual deformation simulation; Input the preset welding scheme into the three-dimensional model of the structural weldment, generate simulation results, and determine the total deformation of the simulation; Use the preset welding scheme to make sample structural welded parts and determine the actual total deformation; Verify the preset welding scheme based on the actual total deformation and the simulated total deformation; For the preset welding scheme that has been verified, the optimization goal is to minimize the welding residual deformation. The genetic algorithm is used to optimize the welding sequence with the welding sequence and welding direction as design variables to obtain the optimal welding sequence of the preset welding scheme. Based on the optimal welding sequence, the tightening position adjustment and anti-deformation design are performed to update the preset welding plan; The updated preset welding scheme is input into the 3D model of the structural welded part to generate simulation results to verify the optimization effect.

[0006] Furthermore, the establishment of a three-dimensional model of a structural welded part for simulating welding residual deformation includes: Determine a simulation object of a structural welded part, divide a hexahedral mesh model according to the simulation object, and refine the mesh of the weld part to generate a three-dimensional model of the structural welded part; Establish a structural support platform according to the structural dimensions of the structural welded parts, and simulate the welding operation platform in the actual welding operation to support the structural welded parts; Create a clamping constraint unit to apply load constraints to the clamping constraint unit; Create a double ellipsoid heat source model to set heat source parameters; Create a welding information management unit, which is used to create a weld trajectory according to a given welding sequence, generate a weld, specify a weld refinement level, and create a pre-weld dwell time, weld interval time, post-weld dwell time, and cooling time for the welding operation; Create a material and ambient temperature setting unit to select the material of the structural part according to the actual material characteristics, select the welding material according to the actual production, and set the ambient operating temperature; Create a calculation unit to calculate the simulation results based on the input welding plan.

[0007] Further, the preset welding scheme is input into the three-dimensional model of the structural welded part, the simulation result is generated, and the total deformation of the simulation is determined, including: The preset welding scheme is input into the three-dimensional model of the structural weldment for finite element simulation to determine the position with the maximum welding deformation, which is recorded as position W; Determine the deformation amount a at the first end and the deformation amount b at the second end of position W; Based on the conservation of energy, the total deformation of the simulation is calculated as c=a+b.

[0008] Further, the method of making a sample of a structural welded part by using a preset welding scheme and determining an actual total deformation includes: A structural welded component sample is manufactured using a preset welding scheme, and a deformation amount A at a first end and a deformation amount B at a second end of a position W of the structural welded component sample are measured; The actual total deformation C is calculated according to the formula C=A+B.

[0009] Further, the verification of the preset welding scheme according to the actual total deformation and the simulated total deformation includes: It is determined whether the difference between the simulated total deformation c and the actual total deformation C is less than a preset threshold; if so, the verification passes; if not, the verification fails.

[0010] Furthermore, the genetic algorithm is used to optimize the welding sequence with the welding sequence and welding direction as design variables to obtain the optimal welding sequence of the preset welding scheme, including: The welding sequence and welding direction of the preset welding scheme are set as design variables; Using a preset coding method, design variables are generated according to a preset welding scheme, and an objective function is set; Based on the preset fitness value, genetic operator, genetic algorithm operating parameters and optimization termination conditions, the NSGA2 genetic algorithm is used to optimize the welding sequence according to the generated generative design variables to obtain the optimal welding sequence of the preset welding scheme.

[0011] Further, the method of using a preset coding method to generate design variables according to a preset welding scheme and setting an objective function includes: Multi-parameter cascade coding is used to identify the welding sequence and welding direction respectively, generating a welding sequence coding combination S and a welding direction coding combination F; The welding scheme is represented as Xi =SF; where 1≤i≤N, N is the number of welding schemes; The objective function is set to the minimum difference between the maximum and minimum Z-direction displacements of the six reference points P1, P2, P3, P4, P5, and P6, denoted as min{U}, U=|max(Ui)-min(Ui)|, i=1,2,3,4,5,6; where max(Ui) represents the maximum value of the Z-direction displacement in the reference point, and min(Ui) represents the minimum value of the radial displacement in the reference point.

[0012] Furthermore, the preset fitness value is calculated by the following formula: ; in, Represents the maximum Z-direction displacement height difference between two reference points of the weldment under each welding sequence.

[0013] Furthermore, the genetic operator ; where n is the population size, F irepresents the fitness of individual i; Furthermore, the operating parameters of the algorithm are set to {M, T, P1, P2}; wherein M is the population size, T is the number of iterations, P1 is the crossover probability, and P2 is the mutation probability.

[0014] Furthermore, the optimization termination condition is specifically as follows: when the number of iterations reaches T, the optimization is terminated, and the current maximum evolutionary number is taken as the optimal solution.

[0015] Compared with the prior art, the beneficial effect of the present invention is as follows: the present invention provides a method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts. First, a three-dimensional model of the structural welded parts is established, and a professional simulation analysis tool is used to simulate and analyze its welding temperature field, welding residual stress and welding deformation. Secondly, the influence of clamping conditions, heat source form and welding parameters on the welding deformation of structural welded parts is studied to achieve simulation and physical verification. Then, with the minimum welding residual deformation as the optimization goal, a genetic algorithm is used, and the welding sequence and welding direction are used as design variables to optimize the welding sequence, obtain a set of optimal solutions, and obtain the welding sequence with the minimum welding deformation through simulation. Finally, based on the results of the welding sequence optimization, the clamping position adjustment and anti-deformation design are used to perform simulation optimization to obtain a process design scheme with the minimum welding deformation.

[0016] The present invention adopts the NSGA2 genetic algorithm, takes the welding sequence and welding direction as design variables, and performs optimization based on the welding sequence with the minimum welding deformation. The technical route of finally optimizing the welding sequence based on the minimum welding deformation is feasible and very efficient. On the basis of the optimization of the welding sequence, the optimization simulation of the welding clamping position and the application simulation of the anti-deformation design can further reduce the welding deformation and improve the welding accuracy. The present invention realizes the combined application of the optimization simulation of the welding sequence parameters and the simulation of the welding clamping position, anti-deformation design, etc., which can improve the welding deformation accuracy to a greater extent and reduce the trial and error cost.

[0017] It can be seen that compared with the prior art, the present invention has outstanding substantive features and significant progress, and the beneficial effects of its implementation are also obvious. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0019] Figure 1 It is a flow chart of the method of embodiment 1 of the present invention.

[0020] Figure 2 It is a method flow chart of embodiment 2 of the present invention.

[0021] Figure 3 It is a schematic diagram of the welding sequence and welding direction parameter combination of the second embodiment of the present invention.

[0022] Figure 4 It is the iterative result and Pareto front diagram of the second embodiment of the present invention. DETAILED DESCRIPTION

[0023] The specific implementation of the present invention is described below with reference to the accompanying drawings.

[0024] Embodiment 1: like Figure 1 A method for simulation prediction and parameter optimization of welding residual deformation of a structural welded part shown in the figure comprises the following steps: S1: Establish a three-dimensional model of structural welded parts for welding residual deformation simulation.

[0025] In a specific implementation, first, a simulation object of a structural welded part is determined, a hexahedral mesh model is divided according to the simulation object, and the mesh of the weld part is refined to generate a three-dimensional model of the structural welded part.

[0026] Then, based on the created 3D model, perform the following configuration: A structural part support platform is established according to the structural dimensions of the structural welded parts, simulating the welding operation platform in actual welding operations to support the structural welded parts.

[0027] Create a clamp constraint element to apply load constraints to the clamp constraint element.

[0028] A double ellipsoid heat source model is created to set the heat source parameters.

[0029] Create a welding information management unit to create weld trajectories based on a given welding sequence, generate welds, specify weld refinement levels, and create pre-weld dwell time, weld interval time, post-weld dwell time, and cooling time for welding operations.

[0030] Create a material and ambient temperature setting unit to select the material of the structural part according to the actual material properties, select the weld material according to actual production, and set the ambient operating temperature.

[0031] Create a calculation unit to calculate the simulation results based on the input welding plan.

[0032] S2: Input the preset welding scheme into the three-dimensional model of the structural welded part, generate simulation results, and determine the total deformation of the simulation.

[0033] In a specific implementation, the preset welding scheme is first input into the three-dimensional model of the structural welded part for finite element simulation to determine the position with the largest welding deformation, which is recorded as position W; then the deformation a at the first end and the deformation b at the second end of position W are determined; finally, based on the law of conservation of energy, the total deformation c=a+b of the simulation is calculated.

[0034] S3: Use the preset welding scheme to make structural welded component samples and determine the actual total deformation.

[0035] In a specific implementation, a structural welded component sample is manufactured using a preset welding scheme, and the deformation A at the first end and the deformation B at the second end of the structural welded component sample position W are measured; at this time, the actual total deformation C is calculated according to the formula C=A+B.

[0036] S4: Verify the preset welding scheme based on the actual total deformation and the simulated total deformation.

[0037] In a specific implementation, it is determined whether the difference between the simulated total deformation c and the actual total deformation C is less than a preset threshold; if so, the verification passes; if not, the verification fails.

[0038] S5: For the preset welding scheme that has passed the verification, the welding residual deformation is minimized as the optimization goal. The genetic algorithm is used to optimize the welding sequence with the welding sequence and welding direction as the design variables to obtain the optimal welding sequence of the preset welding scheme.

[0039] In a specific implementation manner, the specific process of this step is as follows: S51: The welding sequence and welding direction of the preset welding scheme are set as design variables.

[0040] S52: using a preset coding method, generating design variables according to a preset welding scheme, and setting an objective function.

[0041] Specifically, multi-parameter cascade coding is used to identify the welding sequence and welding direction respectively, and a welding sequence coding combination S and a welding direction coding combination F are generated; The welding scheme is expressed as Xi =SF; where 1≤i≤N, N is the number of welding schemes.

[0042] The objective function is set to the minimum difference between the maximum and minimum Z-direction displacements of the six reference points P1, P2, P3, P4, P5, and P6, denoted as min{U}, U=|max(Ui)-min(Ui)|, i=1,2,3,4,5,6; where max(Ui) represents the maximum value of the Z-direction displacement in the reference point, and min(Ui) represents the minimum value of the radial displacement in the reference point.

[0043] S53: Based on the preset fitness value, genetic operator, operation parameters of the genetic algorithm and optimization termination conditions, the NSGA2 genetic algorithm is used to optimize the welding sequence according to the generated generation design variables to obtain the optimal welding sequence of the preset welding scheme.

[0044] In this step, the preset fitness value is calculated by the following formula: ; in, Represents the maximum Z-direction displacement height difference between two reference points of the weldment under each welding sequence.

[0045] Genetic Operator ; where n is the population size, F i represents the fitness of individual i.

[0046] The operating parameters of the algorithm are set as {M, T, P1, P2}; where M is the population size, T is the number of iterations, P1 is the crossover probability, and P2 is the mutation probability.

[0047] The optimization termination condition is as follows: when the number of iterations reaches T, the optimization is terminated, and the current maximum evolutionary generation is taken as the optimal solution.

[0048] S6: Based on the optimal welding sequence, the tightening position adjustment and anti-deformation design are performed to update the preset welding plan.

[0049] In a specific implementation, first, according to the results of the welding sequence parameter optimization, the model welding sequence is adjusted accordingly, and clamping devices are newly added at both ends of the longitudinal beam to adjust the clamping gap between the clamping device and the parts. Then, according to the results of the welding sequence parameter optimization, the model welding sequence is adjusted accordingly, and clamping devices are newly added at both ends of the longitudinal beam to adjust the clamping gap between the clamping device and the parts.

[0050] S7: Input the updated preset welding scheme into the three-dimensional model of the structural welded part to generate simulation results to verify the optimization effect.

[0051] The present embodiment discloses a method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts. The finite element method is used to perform residual deformation prediction simulation and parameter optimization of structural welded parts to control welding deformation and achieve optimal welding accuracy, thereby establishing a technical route and process for simulation and parameter optimization of welding deformation of structural welded parts, which can effectively shorten the production cycle of parts, reduce the cost of trial and error, and improve the welding quality of parts.

[0052] Embodiment 2: Based on Example 1, Figure 2As shown in the figure, this embodiment also discloses a method for simulating and predicting welding residual deformation and optimizing parameters of a structural welded part, which specifically includes the following steps: 1. Simulation of welding residual deformation of the finite element model of the structural welded part: 1.1. Establish a finite element model: Determine the simulation object, divide the hexahedron mesh model, and refine the mesh of the weld part.

[0053] 1.2. Boundary support clamping constraint: Establish a support platform for the structural part according to the structural dimensions of the structural welded part, simulate the welding operation platform in the actual welding operation occasion, and use it to support the structural welded part. Create a clamping constraint unit and apply a load constraint to the clamping constraint unit.

[0054] 1.3. Create a heat source: Adopt a double-ellipsoid heat source model, and set the heat source parameters: current, voltage, efficiency, speed, and the heat source adopts a double-ellipsoid heat source.

[0055] 1.4. Weld generation and welding time management: Create a weld track according to the welding sequence given in the preset welding plan, generate a weld, and specify the weld refinement level. Create the pre-weld dwell time, weld interval time, post-weld dwell time, and cooling time of the welding operation, and generate weld management information.

[0056] 1.5. Material and environmental temperature setting: Select Q235A material as the main body material of the structural part according to the actual material characteristics, select the weld material according to the actual production, and set the environmental operating temperature at 20°C.

[0057] 1.6. Submit the calculation, view and compare the simulation results. The specific simulation results are as follows: Through finite element simulation, the maximum welding deformation at both ends of the long longitudinal beam is 8.7 mm on the right side and 6.2 mm on the left side respectively. According to the law of conservation of energy, the total deformation is 14.9 mm.

[0058] 2. Fabrication of the structural welded part sample: 2.1. Fabrication and error detection of components: Prepare materials according to the simulation "丰" character welding structure model, and measure the material error of the components. Measurement of blanking accuracy: The maximum deformation is 3.1 mm. 2.2. Welding of the structural part: Weld according to the parameters such as welding voltage, current, welding sequence, and welding speed set in the finite element model of the structural welded part. The weld interval time, clamping position and quantity, and post-weld cooling time are executed according to the values set in the simulation model. After welding according to the simulation model parameters and cooling for the required time, measure the deformation error.

[0059] 2.3. Welding deformation measurement: The maximum welding deformation is 5.3mm, which occurs at the end edge of the longitudinal beam. The welding deformation at the left end is 3mm, and the welding deformation in the middle is the smallest, close to 0. Taking into account the deformation error of the blanking, the actual welding structure has a deformation of 8.4mm on the right side, 6.1mm on the left side, and a total deformation of 14.5mm.

[0060] 3. Simulation physical verification: The simulated deformations on both sides of the longitudinal beam were 6.2mm and 8.7mm respectively, totaling 14.9mm. The actual deformations on both sides of the long beam after welding were 6.1mm and 8.4mm respectively, totaling 14.5mm. The overall difference was less than 0.5mm. The simulated deformation and actual deformation trends and values ​​were highly consistent and passed the verification.

[0061] 4. Welding deformation simulation optimization: 4.1 Genetic algorithm optimization of welding sequence: Establish a finite element model, determine the model of the genetic algorithm, and then use the optimization software to optimize the welding process. This welding sequence optimization process is a double-loop process. Each time a new generation of population is generated using the genetic algorithm, the finite element software will call the NSGA2 subroutine to simulate the entire process of thermal-mechanical coupling welding for each individual, and obtain the maximum weld deformation (that is, the adaptation value) under each process. On this basis, the optimal conditions are judged. If the optimal standard is met, the optimal welding sequence can be obtained. If this standard cannot be met, selection, crossover, mutation and other operations are required to form a new generation of populations, and then the next cycle of operations is performed. The specific steps can be divided into the following parts: (1) Determine the design variables: For the welding sequence optimization problem of the longitudinal beam and cross beam of the battery pack bracket in the preset welding scheme, the influence of the welding sequence and welding direction of the six welds on the welding deformation is studied when the welding heat input, clamping positioning method, welding process parameters, etc. do not change. Therefore, the welding sequence and welding direction of the welds are set as the optimization variables of this problem.

[0062] (2) Determine the objective function: The objective function is set to the minimum difference between the maximum and minimum Z displacements of the six reference points P1, P2, P3, P4, P5, and P6, which is recorded as U. The optimal welding sequence is solved, that is, the objective function is min{U}, U=|max(Ui)-min(Ui)| i=1,2,3,4,5,6, where max(Ui) represents the maximum value of the Z displacement in the reference point, and min(Ui) represents the minimum value of the radial displacement in the reference point.

[0063] (3) Determine the encoding method: The properties of each weld should include two parameters: the welding sequence and the welding direction. For this type of genetic coding problem containing multiple parameters, the coding method adopts multi-parameter cascade coding. The two parameters of each weld (sequence number, direction) are represented by numbers in different code positions in the S and F combinations. The combination of welding sequence and welding direction is as follows: Figure 3 As shown. Then a welding scheme can be expressed as Xi = SF, 1≤i≤N, N is the number of welding schemes, and after the welding sequence and welding direction of the weld are encoded separately, these two encoding methods are combined to form a double-parameter cascade encoding, and it is used as the input parameter of the genetic algorithm, such as 1234560101.

[0064] (4) Determine the individual evaluation method: Because the optimal welding process is determined by the minimum welding deformation, and the optimal result after optimization is the individual with the largest fitness value f, the objective function must be transformed as follows: ;in, Represents the maximum Z-direction displacement height difference between two reference points of the weldment under each welding sequence.

[0065] (5) Design of genetic operators: The selection operation adopts proportional selection operation. The percentage of individual fitness in the population determines whether the gene can be inherited to its offspring, that is, In the formula, M is the population size, and Fi represents the fitness of individual i. Individuals in the population will face selection or elimination according to the size of the fitness value. The crossover process uses a single-point crossover operator. After the crossover probability is determined, a parent individual that needs to be crossed will be selected, and in this process, a crossover point position will be randomly generated. The two parent individuals will exchange genetic codes according to this position, and new individuals will be generated.

[0066] (6) Determine the operating parameters of the genetic algorithm: The operating parameters of the algorithm are set to {M, T, P1, P2} {10, 8, 0.8 0.005}. The specific settings are shown in Table 1 below.

[0067] Table 1 Genetic algorithm parameters

[0068] (7) Termination conditions: In this algorithm, when the number of iterations of the algorithm reaches the maximum, the algorithm takes the maximum evolutionary generation as the optimal solution.

[0069] (8) Genetic algorithm optimization results and analysis: During the optimization of welding sequence and direction, the initial population distribution and the iterative Pareto front after 8 iterations are shown as follows: Figure 4 As shown in the figure, the ordinate represents the deformation value of the right side of the longitudinal beam, and the abscissa represents the deformation value of the left end of the longitudinal beam. From the above population distribution, it can be seen that in the initial population, the distribution of solutions is relatively scattered. In the continuous evolution of the population, individuals with lower fitness will be gradually eliminated until the 8th generation, when individuals with higher fitness (smaller U) gradually increase and gather near the optimal point of the problem. According to the Pareto frontier, three optimal solutions, namely 546123, 526143, and 312645 welding sequences, are selected for simulation analysis, and the welding sequence with the smallest welding deformation is 526143.

[0070] 4.2 Clamping position optimization: According to the results of welding sequence parameter optimization, the model welding sequence was adjusted accordingly, and clamping devices were added at both ends of the longitudinal beam to adjust the clamping gap between the clamping device and the parts. Welding deformation was further optimized. After optimization, the deformation on the left side was 4.87mm, the deformation on the right side was 5.19mm, and the total deformation was 10.06mm. The welding accuracy was improved by 32.4% compared with the basic solution of 14.9mm.

[0071] 4.3 Anti-deformation optimization: According to the results of welding sequence and clamping position optimization, combined with the deformation trend of the basic scheme simulation results, anti-deformation design simulation is performed. The original structure welding deformation trend is that the longitudinal beam center expands to both ends of the longitudinal beam. According to the anti-deformation design concept, the middle part of the longitudinal beam is preset to deform downward by 5mm, and the welding sequence and welding voltage, welding current, welding speed and clamping inheritance parameters are optimized. Comprehensive welding deformation optimization simulation analysis shows that the deformation on the left side is 2.1mm, the deformation on the right side is 2.8mm, the deformation in the middle is 1.65mm, and the total deformation is 6.55mm. Compared with the basic scheme with a total deformation of 14.9mm, the welding accuracy is improved by 56%.

[0072] 5. Summary: Using NSGA2 genetic algorithm, taking welding sequence and welding direction as design variables, searching for the minimum welding deformation based on welding sequence, and finally optimizing the welding sequence based on the minimum welding deformation is feasible and very efficient. On the basis of welding sequence optimization, the optimization simulation of welding clamping position and the application simulation of anti-deformation design can further reduce welding deformation and improve welding accuracy.

[0073] In summary, the combined application of welding sequence parameter optimization simulation with welding clamping position, anti-deformation design and other simulations can improve the welding deformation accuracy to a greater extent and reduce the trial and error cost.

[0074] It can be seen that this embodiment discloses a method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts, and realizes a new technical route and method for simulation and optimization of welding deformation of structural welded parts that integrates simulation physical verification, NSGA2 genetic algorithm welding sequence parameter optimization, clamping device and anti-deformation design simulation optimization. This method uses NSGA2's fast non-dominated multi-objective optimization algorithm with an elite retention strategy, and can optimize the software to directly call NSGA2 subroutines to perform fast multi-objective optimization. This method includes simulation and physical verification steps, which can ensure that the simulation results are highly consistent with the actual welding deformation trend and deformation value. This method introduces the concept of anti-deformation design, and obtains the best welding accuracy by combining welding deformation parameter optimization with welding clamping position and anti-deformation simulation.

[0075] The present invention will be further described with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art may make various changes or modifications to the present invention, and these equivalent forms also fall within the scope limited by the application.

Claims

1. A method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts, characterized in that: include: Establish a three-dimensional model of structural welded parts for welding residual deformation simulation; Input the preset welding scheme into the three-dimensional model of the structural weldment, generate simulation results, and determine the total deformation of the simulation; Use the preset welding scheme to make sample structural welded parts and determine the actual total deformation; Verify the preset welding scheme based on the actual total deformation and the simulated total deformation; For the preset welding scheme that has been verified, the optimization goal is to minimize the welding residual deformation. The genetic algorithm is used to optimize the welding sequence with the welding sequence and welding direction as design variables to obtain the optimal welding sequence of the preset welding scheme. Based on the optimal welding sequence, the tightening position adjustment and anti-deformation design are performed to update the preset welding plan; The updated preset welding scheme is input into the 3D model of the structural welded part to generate simulation results to verify the optimization effect.

2. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 1, characterized in that: The method of establishing a three-dimensional model of a structural welded part for simulating welding residual deformation comprises: Determine a simulation object of a structural welded part, divide a hexahedral mesh model according to the simulation object, and refine the mesh of the weld part to generate a three-dimensional model of the structural welded part; Establish a structural support platform according to the structural dimensions of the structural welded parts, and simulate the welding operation platform in the actual welding operation to support the structural welded parts; Create a clamping constraint unit to apply load constraints to the clamping constraint unit; Create a double ellipsoid heat source model to set heat source parameters; Create a welding information management unit, which is used to create a weld trajectory according to a given welding sequence, generate a weld, specify a weld refinement level, and create a pre-weld dwell time, weld interval time, post-weld dwell time, and cooling time for the welding operation; Create a material and ambient temperature setting unit to select the material of the structural part according to the actual material characteristics, select the welding material according to the actual production, and set the ambient operating temperature; Create a calculation unit to calculate the simulation results based on the input welding plan.

3. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 2, characterized in that: The method of inputting the preset welding scheme into the three-dimensional model of the structural welded part, generating simulation results, and determining the total deformation of the simulation includes: The preset welding scheme is input into the three-dimensional model of the structural weldment for finite element simulation to determine the position with the maximum welding deformation, which is recorded as position W; Determine the deformation amount a at the first end and the deformation amount b at the second end of position W; Based on the conservation of energy, the total deformation of the simulation is calculated as c=a+b.

4. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 3 is characterized in that: The method of making a sample of a structural welded part by using a preset welding scheme and determining the actual total deformation includes: A structural welded component sample is manufactured using a preset welding scheme, and a deformation amount A at a first end and a deformation amount B at a second end of a position W of the structural welded component sample are measured; The actual total deformation C is calculated according to the formula C=A+B.

5. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 4, characterized in that: The verification of the preset welding scheme according to the actual total deformation and the simulated total deformation includes: It is determined whether the difference between the simulated total deformation c and the actual total deformation C is less than a preset threshold; if so, the verification passes; if not, the verification fails.

6. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 5, characterized in that: The genetic algorithm is used to optimize the welding sequence with the welding sequence and welding direction as design variables to obtain the optimal welding sequence of the preset welding scheme, including: The welding sequence and welding direction of the preset welding scheme are set as design variables; Using the preset coding method, the design variables are generated according to the preset welding scheme, and the objective function is set; Based on the preset fitness value, genetic operator, genetic algorithm operating parameters and optimization termination conditions, the NSGA2 genetic algorithm is used to optimize the welding sequence according to the generated generative design variables to obtain the optimal welding sequence of the preset welding scheme.

7. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 6, characterized in that: The method of using a preset coding method to generate design variables according to a preset welding scheme and setting an objective function includes: Multi-parameter cascade coding is used to identify the welding sequence and welding direction respectively, generating a welding sequence coding combination S and a welding direction coding combination F; The welding scheme is represented as Xi =SF; where 1≤i≤N, N is the number of welding schemes; The objective function is set to minimize the difference between the maximum and minimum Z-direction displacements of the six reference points P1, P2, P3, P4, P5, and P6, denoted as min{U}, U=|max(Ui)-min(Ui)|, i=1,2,3,4,5,6; Among them, max(Ui) represents the maximum value of the Z-direction displacement in the reference point, and min(Ui) represents the minimum value of the radial displacement in the reference point.

8. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 7, characterized in that , the preset fitness value is calculated by the following formula: ； in, Represents the maximum Z-direction displacement height difference between two reference points of the weldment under each welding sequence.

9. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 8, characterized in that , the genetic operator ; Where n is the population size, F i represents the fitness of individual i; The operating parameters of the algorithm are set to {M, T, P1, P2}; Among them, M is the population size, T is the number of iterations, P1 is the crossover probability, and P2 is the mutation probability.

10. The method for simulation prediction and parameter optimization of welding residual deformation of structural welded parts according to claim 9, characterized in that ,The optimization termination condition is specifically: When the number of iterations reaches T, the optimization terminates and the current maximum evolutionary number is taken as the optimal solution.

Images