Method and device for optimizing laser welding parameters based on part features

Abstract

The application discloses a kind of laser welding parameter optimization method and device based on part feature, it is related to laser welding technical field, the method includes: the part feature data of target part to be welded is collected;Based on the laser welding data record of similar part, the constraint condition of laser welding parameter is set;Laser welding function for optimizing laser welding parameter is constructed;Select multiple historical laser welding parameters that distribution discreteness meets the discreteness requirement as historical population;According to historical population and part feature data, laser welding parameter is optimized, and optimal laser welding parameter is obtained as optimization result.It solves the technical problems that existing laser welding parameter optimization exists based on experience laser welding parameter setting, lack of accurate adjustment optimization for part feature, leading to welding defects, and then affecting product quality and performance, achieves the technical effects of accurately controlling welding parameter, improving welding quality and product performance.

Description

Technical Field

[0001] This application relates to the field of laser welding technology, specifically to a method and apparatus for optimizing laser welding parameters based on part features. Background Technology

[0002] Laser welding technology, with its advantages of high precision, high efficiency, and low heat-affected zone, occupies a pivotal position in modern industrial manufacturing. Whether in automobile manufacturing, aerospace, electronics industry, or precision instrument manufacturing, laser welding plays an irreplaceable role. Especially in the field of parts manufacturing, laser welding can achieve precise connection of parts with various complex shapes and materials, greatly improving product quality and production efficiency. However, traditional welding parameter setting methods are often based on experience or experimentation, lacking precise adjustment and optimization for the characteristics of parts, making it difficult to meet the high standards of welding quality required by modern manufacturing. Improper parameter settings may lead to welding defects such as porosity, slag inclusions, and lack of fusion, thereby affecting the overall performance and service life of products. Traditional laser welding parameter optimization is often unable to adapt to such diverse needs.

[0003] Therefore, current laser welding technologies suffer from technical problems such as setting laser welding parameters based on experience and lacking precise adjustments and optimizations tailored to the characteristics of the parts, leading to welding defects and consequently affecting product quality and performance. Summary of the Invention

[0004] This application provides a laser welding parameter optimization method and apparatus based on part features. By using technical means such as setting welding constraints, constructing laser welding functions, and optimizing welding parameters, it solves the technical problems of existing laser welding parameter optimization, which rely on experience to set laser welding parameters and lack precise adjustment and optimization based on part features, resulting in welding defects and affecting product quality and performance. The application achieves the technical effect of precisely controlling welding parameters and improving welding quality and product performance.

[0005] This application provides a method for optimizing laser welding parameters based on part features. The method includes: collecting part feature data of the target part to be welded, wherein the part feature data includes galvanizing feature data and composition feature data; setting constraints on laser welding parameters based on laser welding data records of similar parts of the target part, wherein the constraints include welding power constraints and welding speed constraints; constructing a laser welding function to optimize the laser welding parameters with the aim of reducing the scale of spatter, porosity, and weld buildup during the laser process; selecting multiple historical laser welding parameters whose distribution discreteness meets the discreteness requirements based on the laser welding data records of the similar parts, as a historical population; optimizing the laser welding parameters based on the constraints and the laser welding function, and based on the historical population and part feature data, obtaining the optimal laser welding parameters as the optimization result, wherein the optimization process includes cross-updating, and the cross-updating is selected based on the influence of welding power and welding speed on spatter, porosity, and weld within the laser welding parameters.

[0006] In a possible implementation, the component feature data of the target component to be welded is collected, and the following processing is also performed: the zinc coating thickness of the target component is collected to obtain zinc coating feature data; the composition information of the target component is collected to obtain composition feature data, and combined with the zinc coating feature data, it is used as component feature data.

[0007] In a possible implementation, based on laser welding data records of similar parts to the target part, constraints on laser welding parameters are set, and the following processing is performed: based on laser welding data records of similar parts, a welding power range and a welding speed range are constructed; the welding power in the laser welding parameters falls into the welding power range as a welding power constraint, and the welding speed falls into the welding speed range as a welding speed constraint, and the constraint conditions are set.

[0008] In possible implementations, to reduce spatter, porosity, and weld buildup during the laser process, a laser welding function optimizing the laser welding parameters is constructed, and the following processing is performed: The laser welding function is as follows: ; Where LAW represents weldability, , and These are spatter weight, porosity weight, and weld weight, respectively. , and These are information on spatter size, porosity size, and weld buildup size obtained by welding according to laser welding parameters. , and These are preset splash size information, preset porosity size information, and preset weld deposit size information, respectively.

[0009] In a possible implementation, based on laser welding data records of similar parts, multiple historical laser welding parameters whose distribution dispersion meets the dispersion requirement are selected as a historical population. The following processing is also performed: Based on laser welding data records of similar parts, a set of sample laser welding parameters, as well as a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample weld accumulation scale information are obtained. The sample laser welding parameter set includes a set of sample welding power and a set of sample welding speed. Based on the sample welding power and sample welding speed sets, welding power dispersion parameters and welding speed dispersion parameters are calculated. Based on the sample spatter scale information, sample porosity scale information, and sample weld accumulation scale information sets, a set of sample welding fitness is calculated using the laser welding function, and welding fitness dispersion parameters are calculated. Based on the welding power dispersion parameters, welding speed dispersion parameters, and welding fitness dispersion parameters, historical dispersion parameters are calculated. It is determined whether these parameters exceed a dispersion parameter threshold. If so, the sample laser welding parameter set is used as the historical population; otherwise, a new set of sample laser welding parameters is obtained.

[0010] In a possible implementation, based on the constraints and the laser welding function, and according to the historical population and part feature data, the laser welding parameters are optimized to obtain the optimal laser welding parameters. As the optimization result, the following processing is also performed: Based on the constraints and the number of laser welding parameters in the historical population, multiple new laser welding parameters are randomly generated to construct a basic population, and an index relationship with the historical population is randomly constructed; using the historical population as the adjustment direction, and according to the index relationship, multiple laser welding parameters in the basic population are adjusted and updated to obtain an updated population; based on the part feature data, multiple laser welding parameters in the updated population are simulated to obtain multiple simulated spatter scale information, multiple simulated porosity scale, and multiple simulated weld accumulation scale information; based on the multiple... Multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information are used to predict the impact on the sealing and smoothness of the target part, obtaining multiple impact magnitudes. Based on the magnitude of the impact magnitude, several laser welding parameters in the updated population are randomly selected and cross-updated with several laser welding parameters in the historical population to obtain a cross-population. Simulated laser welding is performed based on multiple laser welding parameters in the cross-population and the historical population, and the welding fitness is calculated by combining the laser welding function. Laser welding parameters with higher welding fitness are retained to obtain an updated historical population. Based on the updated historical population, iterative optimization is performed until convergence, and the laser welding parameters with the highest welding fitness in the final historical population are output to obtain the optimal laser welding parameters.

[0011] In a possible implementation, based on the part feature data, simulated laser welding is performed on multiple laser welding parameters within the updated population to obtain multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information. The following processing is also performed: based on laser welding data records of similar parts, a set of sample galvanizing feature data, a set of sample composition feature data, a set of sample laser welding parameters, a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample simulated weld seam accumulation scale information are obtained as welding simulation construction data. Using the welding simulation construction data, a laser welding simulator is constructed, and simulated laser welding is performed on multiple laser welding parameters within the updated population combined with the part feature data to obtain multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information.

[0012] In a possible implementation, based on multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information, the impact on the sealing and smoothness of the target part is predicted, resulting in multiple impact magnitudes. Based on the magnitude of the impact magnitude, several laser welding parameters within the updated population are randomly selected and cross-updated with several laser welding parameters from the historical population using either welding power or welding speed. The following processing is also performed: based on finished product processing data records of similar parts to the target part, a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample simulated weld seam accumulation scale information are obtained, along with a set of sample sealing impact magnitudes and a set of sample smoothness impact magnitudes. The set of sample spatter scale information, the set of sample porosity scale information, and the set of sample simulated weld seam accumulation scale information are then used to... A part finished product influence analyzer is constructed using a set of information on stacking scale, a set of influence amplitudes on sample sealing performance, and a set of influence amplitudes on sample smoothness. Based on multiple sets of simulated spatter scale, simulated porosity scale, and simulated weld stacking scale information, multiple sealing performance influence amplitudes and multiple smoothness influence amplitudes are analyzed and predicted, and weighted to obtain multiple influence amplitudes. Multiple cross-probabilities are generated based on these influence amplitudes, where the magnitude of the influence amplitude is positively correlated with the magnitude of the cross-probability. According to these multiple cross-probabilities, several laser welding parameters within the updated population are randomly selected and their welding power or welding speed is cross-updated with several laser welding parameters within the historical population to obtain a cross-population, where the number of selected laser welding parameters is less than the total number of individuals in the updated population.

[0013] This application also provides a laser welding parameter optimization device based on part features, including: The part feature data acquisition module is used to acquire part feature data of the target part to be welded, wherein the part feature data includes galvanizing feature data and composition feature data. The constraint setting module is used to set constraints on laser welding parameters based on laser welding data records of similar parts to the target part. The constraints include welding power constraints and welding speed constraints. A laser welding function construction module is used to construct a laser welding function that optimizes laser welding parameters in order to reduce the spatter, porosity and weld buildup during the laser process. The historical laser welding parameter selection module is used to select multiple historical laser welding parameters whose distribution dispersion meets the discreteness requirements based on the laser welding data records of the same type of parts, as a historical population. A laser welding parameter optimization module is used to optimize laser welding parameters based on the constraints and laser welding function, according to the historical population and part feature data, to obtain the optimal laser welding parameters as the optimization result. The optimization process includes cross-updating, which is selected based on the influence of welding power and welding speed on spatter, porosity and weld seam within the laser welding parameters.

[0014] This application proposes a laser welding parameter optimization method and apparatus based on part characteristics. The method involves collecting part characteristic data of the target part to be welded; setting constraints on laser welding parameters based on laser welding data records of similar parts; constructing a laser welding function to optimize the parameters, aiming to reduce spatter, porosity, and weld buildup during the laser process; selecting multiple historical laser welding parameters whose distribution discreteness meets the discreteness requirements as a historical population; and optimizing the laser welding parameters based on the historical population and part characteristic data to obtain the optimal laser welding parameters as the optimization result. This method solves the technical problem of existing laser welding parameter optimization methods that rely on experience-based parameter settings and lack precise adjustment and optimization based on part characteristics, leading to welding defects and affecting product quality and performance. It achieves the technical effect of precisely controlling welding parameters and improving welding quality and product performance. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the apparatus according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0016] Figure 1 A schematic diagram of the laser welding parameter optimization method based on part features provided in this application embodiment; Figure 2 A schematic diagram of the structure of the laser welding parameter optimization device based on part features provided in this application embodiment.

[0017] Explanation of reference numerals in the attached drawings: Part feature data acquisition module 10, constraint setting module 20, laser welding function construction module 30, historical laser welding parameter selection module 40, and laser welding parameter optimization module 50. Detailed Implementation

[0018] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application.

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same or different subsets of all possible embodiments and may be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or apparatuses. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0021] This application provides a method for optimizing laser welding parameters based on part features, such as... Figure 1 As shown, the method includes: Step S100: Collect part feature data of the target part to be welded. This part feature data includes galvanization feature data and composition feature data. The target part to be welded is pre-selected as a galvanized part, such as a galvanized steel pipe. Specifically, galvanization feature data refers to the condition information of the galvanized layer on the part's surface. Galvanization is a common anti-corrosion treatment method, preventing oxidation and corrosion by covering the part's surface with a layer of zinc. Galvanization feature data typically includes parameters such as the thickness, uniformity, and adhesion of the galvanized layer. For example, an excessively thick galvanized layer may cause excessive smoke and spatter during welding, affecting the appearance and performance of the weld joint; while poor adhesion may cause the zinc layer to peel off during welding, forming welding defects. Composition feature data refers to the chemical composition information of the part's material. Different material compositions have different adaptability to laser welding. This typically includes the material's elemental composition, alloy content, and impurity content, which determine the material's melting point, thermal conductivity, coefficient of thermal expansion, and other physical properties, thus affecting heat conduction, molten pool formation, and solidification processes during laser welding.

[0022] In one possible implementation, step S100 further includes step S110, acquiring the thickness of the galvanized layer of the target part to obtain galvanized layer characteristic data. The thickness of the galvanized layer on the surface of the target part is measured and recorded before laser welding, serving as galvanized layer characteristic data. It also includes step S120, acquiring the composition information of the target part to obtain composition characteristic data, which, combined with the galvanized layer characteristic data, serves as part characteristic data. Acquiring the composition information of the target part refers to determining the chemical composition of the part material through chemical analysis and spectral analysis, constituting composition characteristic data. This may include the types and proportions of elements, as well as any alloying or impurity components that may be present. Combining the obtained composition characteristic data with the galvanized layer characteristic data constitutes part characteristic data. Based on this part characteristic data, the heat conduction, molten pool formation, and solidification during the welding process can be predicted more accurately, thereby adjusting parameters such as welding power and welding speed to achieve high-quality welding of the product.

[0023] Step S200: Based on the laser welding data records of similar parts to the target part, set the constraints for the laser welding parameters, including welding power constraints and welding speed constraints. Based on laser welding data records of similar parts to the target part, we can set constraints on laser welding parameters to ensure the safety of the welding process and the stability of welding quality. These constraints mainly include welding power constraints and welding speed constraints. Specifically, welding power constraints mean that the adjustment range of welding power during laser welding must be controlled within certain limits. The welding power range is determined based on laser welding data records of similar parts, combined with factors such as the material properties, thickness, and required welding depth of the target part. Too low a welding power may lead to insufficient weld joint strength and substandard welding quality; while too high a welding power may cause overheating, excessive spatter, or even burn-through. Welding speed constraints mean that the adjustment range of welding speed during laser welding also needs to be limited. Too high a welding speed may lead to insufficient weld penetration and an unstable weld joint; while too low a welding speed may cause weld overheating and coarse grains, affecting the performance of the weld joint. Based on laser welding data records of similar parts, a suitable welding speed range is determined so that welding within this range ensures that the weld penetration and width meet requirements while avoiding welding defects.

[0024] In one possible implementation, step S200 further includes step S210, constructing a welding power range and a welding speed range based on laser welding data records of similar parts. According to historical laser welding data records of similar parts, a suitable welding power and welding speed range applicable to the target part is determined. It also includes step S220, setting the constraint conditions by ensuring that the welding power in the laser welding parameters falls within the welding power range as a welding power constraint, and ensuring that the welding speed falls within the welding speed range as a welding speed constraint. In actual laser welding, the welding power and welding speed parameters of the target part are set within these two ranges. This means that a welding power or welding speed cannot be arbitrarily selected; instead, an appropriate value must be chosen within the welding power and welding speed ranges based on the characteristics of the part and the welding requirements. Setting the welding power and welding speed within the corresponding ranges creates welding power constraints and welding speed constraints. These welding constraints ensure the rationality of the welding parameters, thereby helping to avoid problems such as overheating, overcooling, excessively fast or slow welding speeds during the welding process, ensuring the quality and stability of the weld joint. Furthermore, the welding constraints provide a basis for setting welding parameters in actual welding, reducing welding defects and quality problems caused by improper parameter settings.

[0025] Step S300: To reduce the spatter, porosity and weld buildup during the laser process, a laser welding function is constructed to optimize the laser welding parameters. During welding, excessive welding power and speed can lead to spatter and porosity, while insufficient welding power and speed can result in weld buildup. Therefore, appropriate welding power and speed are necessary to reduce these three defects. To minimize spatter, porosity, and weld buildup during laser welding, a laser welding function is designed. Specifically, to reduce spatter, the laser welding function needs to consider the effects of welding power and speed. Appropriate adjustments to welding power and speed can reduce overheating and poor penetration, thus reducing spatter generation. To reduce porosity, the laser welding function needs to focus on pre-weld surface treatment and the selection of welding process parameters. Pre-weld surface treatments such as physical-mechanical cleaning or chemical cleaning can effectively suppress the formation of weld porosity. Simultaneously, reasonable adjustments to welding process parameters, such as laser power, focal length, and welding speed, can optimize weld penetration. To reduce weld buildup, the laser welding function needs to consider the uniformity of weld filling and the control of welding speed. Appropriately increasing the laser heat input and reducing the welding speed can slow down the cooling of the molten pool, resulting in more uniform weld filling and avoiding excessive weld height and buildup.

[0026] In one possible implementation, step S300 further includes constructing a laser welding function that optimizes the laser welding parameters, as follows: ; Where LAW represents weldability, , and These are spatter weight, porosity weight, and weld weight, respectively. , and These are information on spatter size, porosity size, and weld buildup size obtained by welding according to laser welding parameters. , and These are preset spatter size information, preset porosity size information, and preset weld buildup size information, respectively. The laser welding function is a mathematical model used to evaluate the welding quality under different combinations of laser welding parameters. It comprehensively considers factors such as spatter size, porosity size, and weld buildup size during the welding process, quantifying these factors into a comparable value, namely welding adaptability. Among these, the smaller the spatter size, porosity size, and weld buildup size, the better, and the higher the welding quality.

[0027] Step S400: Based on the laser welding data records of similar parts, select multiple historical laser welding parameters whose distribution dispersion meets the dispersion requirement as a historical population. Analyze the laser welding data records of similar parts to select combinations of historical welding parameters that have a certain degree of dispersion in distribution and meet the dispersion requirement, forming a set of historical parameters, i.e., the historical population. Specifically, calculate the variance of welding power and welding speed within multiple laser welding parameters as dispersion parameters, and calculate the variance of multiple stress and fitness as fitness dispersion parameters. The larger the variance, the more dispersed the parameters, and the better the global optimization. The dispersion requirement means that the selected welding parameters should have a certain degree of variation and difference, rather than being too concentrated or similar. By selecting historical welding parameters with dispersion as the historical population, further optimization analysis can be performed based on these welding parameters. Iteration and search can be carried out on the basis of the historical population to find laser welding parameters more suitable for the target part.

[0028] In one possible implementation, step S400 further includes step S410, which involves obtaining a set of sample laser welding parameters, a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample weld accumulation scale information based on laser welding data records of similar parts. The set of sample laser welding parameters includes a set of sample welding power and a set of sample welding speed. A certain number of sample data are extracted from existing laser welding data to form a sample laser welding parameter set, a sample spatter scale information set, a sample porosity scale information set, and a sample weld accumulation scale information set. Specifically, the sample laser welding parameter set includes welding power and welding speed data for multiple samples. The sample welding power set includes welding power values ​​for different samples, reflecting the amount of energy input during welding. The sample welding speed set includes welding speed values ​​for different samples, reflecting the speed of the welding process. The sample spatter scale information set records the degree of spatter generated by different sample welds, reflecting the stability and quality of the welding process. The sample porosity scale information set records the number and size of pores present in the welds of different samples. The sample weld accumulation scale information set describes the morphology and accumulation of the weld, reflecting the filling effect and uniformity of the weld. The process also includes step S420, calculating welding power dispersion parameters and welding speed dispersion parameters based on the sample welding power set and sample welding speed set. The variances of welding power and welding speed within multiple laser welding parameters are calculated as welding power dispersion parameters and welding speed dispersion parameters, respectively reflecting the fluctuation or dispersion of welding power and the degree of variation in welding speed among different samples. The method also includes step S430, which calculates a set of sample welding fitness based on the laser welding function, using the set of sample spatter scale information, sample porosity scale information, and sample weld seam accumulation scale information. These sets are then input into the laser welding function to obtain a set of sample welding fitness, and multiple fitness variances are calculated to obtain the welding fitness dispersion parameters. The method further includes step S440, which calculates historical dispersion parameters based on the welding power dispersion parameters, welding speed dispersion parameters, and welding fitness dispersion parameters. It determines whether these parameters exceed a dispersion parameter threshold; if so, the set of sample laser welding parameters is used as the historical population; otherwise, a new set of sample laser welding parameters is obtained.By comprehensively analyzing the dispersion parameters of welding power, welding speed, and welding adaptability, historical discreteness parameters are calculated. These parameters are used to evaluate the discreteness of the entire sample laser welding parameter set. Specifically, the discreteness parameter threshold is set based on actual application requirements and optimization objectives. It is used to determine whether the discreteness of the sample laser welding parameter set meets the requirements. If the historical discreteness parameter is greater than the discreteness parameter threshold, it indicates that the sample laser welding parameter set has sufficient discreteness, providing a rich search space and possibilities for subsequent parameter optimization. This sample laser welding parameter set is then considered as the historical population. If the historical discreteness parameter is not greater than the discreteness parameter threshold, it indicates that the discreteness of the sample laser welding parameter set is insufficient, which may lead to the optimization process getting stuck in local optima or the search space being too narrow. In this case, it is necessary to obtain a new sample laser welding parameter set.

[0029] Step S500: Based on the constraints and the laser welding function, and according to the historical population and part feature data, the laser welding parameters are optimized to obtain the optimal laser welding parameters as the optimization result. The optimization process includes cross-updating, which selects the influence of welding power and welding speed on spatter, porosity and weld seam based on the laser welding parameters. Based on constraints and laser welding functions, and using historical population and part characteristic data, the system optimizes laser welding parameters to find the most suitable combination of laser welding parameters for a specific part, achieving optimal welding results. Specifically, the system searches for optimal laser welding parameters using optimization algorithms (such as genetic algorithms and particle swarm optimization) based on constraints, laser welding functions, historical population data, and part characteristic data. Cross-updating simulates the genetic crossover process in nature, generating new parameter combinations by combining the advantages of different combinations, aiming to find better solutions. In this cross-updating process, welding power and welding speed are key laser welding parameters. The system analyzes the specific impact of welding power and welding speed on welding quality based on historical population data and the evaluation results of the laser welding function. This allows for a more precise balance of these factors when selecting new parameter combinations, improving welding quality and reducing problems such as spatter, porosity, and weld buildup. Through continuous iteration and optimization, the system ultimately finds a set of laser welding parameters that satisfies the constraints and optimizes the laser welding function. This set represents the optimal laser welding parameters for the current part characteristic data and is output as the optimization result for actual welding operations.

[0030] In one possible implementation, step S500 further includes step S510, which involves randomly generating multiple new laser welding parameters based on the constraints and the number of laser welding parameters in the historical population, constructing a basic population, and randomly constructing an index relationship with the historical population. By introducing randomness into the existing historical population, the search space is expanded, thereby increasing the possibility of finding better laser welding parameters. Specifically, multiple new laser welding parameters are randomly generated based on the number of laser welding parameters in the historical population to construct the basic population. Then, an index relationship is randomly constructed between the newly generated laser welding parameters and the historical population, which helps to track the source and evolution of each parameter in subsequent optimization processes. The implementation also includes step S520, which involves adjusting and updating multiple laser welding parameters in the basic population according to the index relationship, using the historical population as the adjustment direction, to obtain an updated population. The method utilizes superior genes (i.e., excellent laser welding parameter combinations) from historical populations to guide the adjustment of parameters in the base population, aiming to obtain laser welding parameter combinations with better performance. Specifically, it finds the corresponding parameters in the historical population for each parameter in the base population based on index relationships, adjusts the parameters in the base population, and finally obtains an updated population. The method also includes step S530, which simulates laser welding using multiple laser welding parameters within the updated population based on the part feature data, obtaining multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information. Based on the input laser welding parameters and part feature data, the method simulates the formation process of the weld joint, including stages such as molten pool flow, solidification, and cooling, obtaining multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information. The process also includes step S540, which involves predicting the impact on the sealing and smoothness of the target part based on multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information, thereby obtaining multiple impact magnitudes. Based on the magnitude of the impact magnitude, several laser welding parameters within the updated population are randomly selected and cross-updated with several laser welding parameters within the historical population to obtain a cross-population. By analyzing the simulation information, the impact magnitude of different laser welding parameter combinations on the sealing and smoothness of the target part can be predicted. Laser welding parameters with larger impact magnitudes are cross-updated, have a higher probability of being selected, and their welding power or welding speed is swapped with the corresponding laser welding parameters within the historical population to improve optimization accuracy. Specifically, based on the magnitude of the impact magnitude, several laser welding parameters within the updated population are randomly selected and cross-updated with several laser welding parameters within the historical population. By combining the superior genes from the updated population and the historical population, a new combination of laser welding parameters, i.e., the cross-population, is generated.The process also includes step S550, which involves simulating laser welding based on multiple laser welding parameters from the crossover population and the historical population, and calculating the welding fitness using the laser welding function. Laser welding parameters with higher welding fitness are retained to obtain an updated historical population. Based on the laser welding parameters from the crossover population and the historical population, the welding process is simulated to obtain welding result data. This data is then evaluated using the laser welding function to calculate a welding fitness value. Laser welding parameters from the crossover population and the historical population are then screened, retaining those with higher welding fitness and discarding those with lower fitness, thus forming a new historical population. The process also includes step S560, which involves iterative optimization based on the updated historical population until convergence, outputting the laser welding parameters with the highest welding fitness within the final historical population to obtain the optimal laser welding parameters. In each iteration, laser welding is simulated based on the updated laser welding parameters in the historical population, and the welding fitness is calculated in conjunction with the laser welding function. Based on the size of the welding fitness, the welding parameters with higher welding fitness are retained. This process is repeated until the convergence condition is met. This condition is usually set based on the rate of change of welding fitness or the number of iterations. When the rate of change of welding fitness is lower than a certain threshold, or when the number of iterations reaches a preset maximum value, the laser welding parameters with the highest welding fitness in the historical population are output, which are the optimal laser welding parameters.

[0031] In one possible implementation, step S530 further includes step S531, which involves acquiring a set of sample galvanizing characteristic data, a set of sample composition characteristic data, a set of sample laser welding parameters, a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample simulated weld seam accumulation scale information based on laser welding data records of similar parts. This set of sample data obtained from laser welding data records of similar parts can be used as welding simulation construction data. By using computer simulation software to accurately simulate and predict the laser welding process, and by continuously adjusting and optimizing the simulation parameters, the welding results under different parameter combinations can be predicted, thereby finding the optimal laser welding parameters. The method also includes step S532, which involves using the welding simulation construction data to construct a laser welding simulator, and simulating laser welding using multiple laser welding parameters combined with part characteristic data within the updated population to obtain multiple sets of simulated spatter scale information, multiple sets of simulated porosity scale information, and multiple sets of simulated weld seam accumulation scale information. Using the collected welding simulation data, a laser welding simulator is constructed to simulate the laser welding process and predict the welding results. Multiple laser welding parameters from the updated population are combined with part feature data and input into the laser welding simulator. The laser welding simulator simulates the actual laser welding process and outputs multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld accumulation scale information, reflecting the welding defects and weld morphology that may occur under different laser welding parameters.

[0032] In one possible implementation, step S540 further includes step S541, which involves obtaining a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample simulated weld seam accumulation scale information, as well as a set of sample sealing performance influence amplitude and a set of sample smoothness influence amplitude, based on finished product processing data records of similar parts to the target part. The set of sample spatter scale information, the set of sample porosity scale information, and the set of sample simulated weld seam accumulation scale information are obtained based on finished product processing data records of similar parts, reflecting various defects and weld seam morphologies that may occur during the actual welding process; the set of sample sealing performance influence amplitude and the set of sample smoothness influence amplitude are obtained based on the sealing and smoothness test results of the parts in the finished product processing data records, and sealing and smoothness are directly related to the performance and appearance of the parts. The method also includes step S542, which uses the sample spatter scale information set, sample porosity scale information set, sample simulated weld accumulation scale information set, sample sealing performance influence amplitude set, and sample smoothness influence amplitude set to construct a part finished product influence analyzer. Based on the multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld accumulation scale information, the analyzer predicts and obtains multiple sealing performance influence amplitudes and multiple smoothness influence amplitudes, and calculates multiple influence amplitudes by weighting. The component finished product impact analyzer is a tool built based on known processing data of similar component finished products, capable of predicting and analyzing the impact of different welding parameters on the performance of the component finished product. It uses existing data to predict and evaluate new simulated welding results. Specifically, data from a set including sample spatter scale information, sample porosity scale information, sample simulated weld seam buildup scale information, sample sealing performance impact amplitude, and sample smoothness impact amplitude are input into the component finished product impact analyzer for analysis and prediction. It outputs multiple predicted sealing performance impact amplitudes and multiple predicted smoothness impact amplitudes to measure the specific degree of influence of different combinations of simulated welding parameters on the performance of the component finished product. Finally, the obtained impact amplitudes are weighted and calculated to obtain a comprehensive impact amplitude value, which is the multiple impact amplitudes. The analysis also includes step S543, which generates multiple cross probabilities based on the multiple impact amplitudes, wherein the magnitude of the impact amplitude and the magnitude of the cross probability are positively correlated. In the process of optimizing laser welding parameters, the crossover probability of different parameter combinations in the genetic algorithm is determined by analyzing the impact of simulated welding results on the performance of finished parts. Specifically, the crossover probability determines the probability of two individuals crossing over during the crossover operation. The crossover operation refers to combining two different welding parameter combinations to generate a new parameter combination. When the impact is large, it means that the welding parameter combination may lead to poor part performance, and the crossover probability is also greater, making it more likely to cross over with other excellent parameter combinations, thereby generating a new parameter combination that is more likely to bring performance improvement.The process also includes step S544, where, according to the multiple crossover probabilities, several laser welding parameters from the updated population and several laser welding parameters from the historical population are randomly selected for cross-updating to obtain a cross-population. The number of selected laser welding parameters is less than the total number of individuals in the updated population. Based on the obtained multiple crossover probabilities, several laser welding parameters are randomly selected from the updated population, typically less than the total number of individuals in the updated population. Simultaneously, a corresponding number of laser welding parameters are selected from the historical population for cross-updating. By exchanging the welding power or welding speed values ​​of corresponding individuals in the updated and historical populations, a new parameter combination, i.e., a cross-population, can be generated.

[0033] In the above text, refer to Figure 1 A laser welding parameter optimization method based on part features according to an embodiment of the present invention is described in detail. Next, reference will be made to... Figure 2 A laser welding parameter optimization apparatus based on part features according to an embodiment of the present invention is described.

[0034] The laser welding parameter optimization device based on part features according to embodiments of the present invention addresses the technical problems of existing laser welding parameter optimization methods, which rely on experience-based laser welding parameter settings and lack precise adjustment and optimization based on part features, leading to welding defects and affecting product quality and performance. The device achieves the technical effect of precisely controlling welding parameters and improving welding quality and product performance. The laser welding parameter optimization device based on part features includes: a part feature data acquisition module 10, a constraint setting module 20, a laser welding function construction module 30, a historical laser welding parameter selection module 40, and a laser welding parameter optimization module 50.

[0035] The part feature data acquisition module 10 is used to acquire part feature data of the target part to be welded, wherein the part feature data includes galvanizing feature data and composition feature data. The constraint setting module 20 is used to set constraints on laser welding parameters based on laser welding data records of similar parts to the target part. The constraints include welding power constraints and welding speed constraints. The laser welding function construction module 30 is used to construct a laser welding function that optimizes laser welding parameters in order to reduce the spatter scale, porosity scale and weld accumulation scale during the laser process. Historical laser welding parameter selection module 40 is used to select multiple historical laser welding parameters whose distribution dispersion meets the dispersion requirements based on the laser welding data records of the same type of parts, as a historical population. The laser welding parameter optimization module 50 is used to optimize the laser welding parameters based on the constraints and the laser welding function, according to the historical population and part feature data, to obtain the optimal laser welding parameters as the optimization result. The optimization process includes cross-updating, which is selected based on the influence of welding power and welding speed on spatter, porosity and weld seam within the laser welding parameters.

[0036] The specific configuration of the part feature data acquisition module 10 will be described in detail below. The part feature data acquisition module 10 may further include: acquiring the zinc coating thickness of the target part to obtain zinc coating feature data; acquiring the composition information of the target part to obtain composition feature data, and combining the zinc coating feature data as part feature data.

[0037] The specific configuration of the constraint setting module 20 will be described in detail below. The constraint setting module 20 further includes: constructing a welding power range and a welding speed range based on laser welding data records of similar parts; setting the constraint conditions by ensuring that the welding power in the laser welding parameters falls within the welding power range as a welding power constraint and the welding speed falls within the welding speed range as a welding speed constraint.

[0038] The specific configuration of the laser welding function construction module 30 will be described in detail below. The laser welding function construction module 30 may further include the following formula for the laser welding function: ; Where LAW represents weldability, , and These are spatter weight, porosity weight, and weld weight, respectively. , and These are information on spatter size, porosity size, and weld buildup size obtained by welding according to laser welding parameters. , and These are preset splash size information, preset porosity size information, and preset weld deposit size information, respectively.

[0039] The specific configuration of the historical laser welding parameter selection module 40 will be described in detail below. The historical laser welding parameter selection module 40 may further include: obtaining a sample laser welding parameter set, as well as a sample spatter scale information set, a sample porosity scale information set, and a sample weld seam accumulation scale information set after welding, based on laser welding data records of similar parts. The sample laser welding parameter set includes a sample welding power set and a sample welding speed set. Based on the sample welding power set and sample welding speed set, welding power dispersion parameters and welding speed dispersion parameters are calculated. Based on the sample spatter scale information set, sample porosity scale information set, and sample weld seam accumulation scale information set, a sample welding fitness set is calculated based on the laser welding function, and welding fitness dispersion parameters are calculated. Based on the welding power dispersion parameters, welding speed dispersion parameters, and welding fitness dispersion parameters, a historical discreteness parameter is calculated, and it is determined whether it exceeds a discreteness parameter threshold. If it does, the sample laser welding parameter set is used as the historical population; otherwise, a new sample laser welding parameter set is obtained.

[0040] The specific configuration of the laser welding parameter optimization module 50 will be described in detail below. The laser welding parameter optimization module 50 may further include: randomly generating multiple new laser welding parameters based on the constraints and the number of laser welding parameters in the historical population, constructing a basic population, and randomly constructing an index relationship with the historical population; adjusting and updating multiple laser welding parameters in the basic population according to the index relationship, using the historical population as the adjustment direction, to obtain an updated population; performing simulated laser welding on the multiple laser welding parameters in the updated population based on the part feature data, obtaining multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information; and adjusting and updating the multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information based on the multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information. The model information is used to predict the impact range on the sealing and smoothness of the target part, obtaining multiple impact ranges. Based on the magnitude of the impact range, several laser welding parameters in the updated population are randomly selected and cross-updated with several laser welding parameters in the historical population to obtain a cross-population. Simulated laser welding is performed based on multiple laser welding parameters in the cross-population and the historical population, and the welding fitness is calculated by combining the laser welding function. Laser welding parameters with higher welding fitness are retained to obtain an updated historical population. Based on the updated historical population, iterative optimization is performed until convergence, and the laser welding parameters with the highest welding fitness in the final historical population are output to obtain the optimal laser welding parameters.

[0041] The specific configuration of the laser welding parameter optimization module 50 will be described in detail below. The laser welding parameter optimization module 50 further includes: acquiring a set of sample galvanizing characteristic data, a set of sample composition characteristic data, a set of sample laser welding parameters, a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample simulated weld seam accumulation scale information based on laser welding data records of similar parts, as welding simulation construction data; using the welding simulation construction data, constructing a laser welding simulator, and performing simulated laser welding on multiple laser welding parameters within the updated population combined with part characteristic data to obtain multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information.

[0042] The specific configuration of the laser welding parameter optimization module 50 will be described in detail below. The laser welding parameter optimization module 50 further includes: acquiring a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample simulated weld seam accumulation scale information, as well as a set of sample sealing performance influence amplitude and a set of sample smoothness influence amplitude, based on finished product processing data records of similar parts of the target part; constructing a finished product influence analyzer using the set of sample spatter scale information, sample porosity scale information, sample simulated weld seam accumulation scale information, sample sealing performance influence amplitude, and sample smoothness influence amplitude; analyzing and predicting multiple sealing performance influence amplitudes and multiple smoothness influence amplitudes based on the multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information, and weighted calculating multiple influence amplitudes; generating multiple cross-probabilities based on the multiple influence amplitudes, wherein the magnitude of the influence amplitude is positively correlated with the magnitude of the cross-probability; and randomly selecting several laser welding parameters in the update population and several laser welding parameters in the historical population for cross-updating according to the multiple cross-probabilities to obtain a cross-population, wherein the number of selected laser welding parameters is less than the total number of individuals in the update population.

[0043] The laser welding parameter optimization device based on part features provided in this embodiment of the invention can execute the laser welding parameter optimization method based on part features provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0044] Although this application makes various references to certain modules in the apparatus according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not intended to limit the scope of protection of this invention.

[0045] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A laser welding parameter optimization method based on part features, characterized in that, The method includes: Collect part feature data of the target part to be welded, including galvanization feature data and composition feature data; Based on the laser welding data records of similar parts to the target part, constraints are set for the laser welding parameters, including welding power constraints and welding speed constraints. To reduce the spatter, porosity, and weld buildup during the laser process, a laser welding function is constructed to optimize the laser welding parameters. Based on the laser welding data records of the same type of parts, several historical laser welding parameters whose distribution dispersion meets the dispersion requirements are selected as historical populations. Based on the constraints and laser welding function, the laser welding parameters are optimized according to the historical population and part feature data to obtain the optimal laser welding parameters as the optimization result. The optimization process includes cross-updating, which is selected based on the influence of welding power and welding speed on spatter, porosity and weld seam within the laser welding parameters. The process of collecting part feature data of the target part to be welded includes: Collect the zinc coating thickness of the target part and obtain zinc coating characteristic data; The composition information of the target part is collected to obtain composition feature data, which is then combined with the galvanizing feature data to form the part feature data. Specifically, to reduce the spatter, porosity, and weld buildup during the laser process, a laser welding function is constructed to optimize the laser welding parameters, as shown in the following equation: ； Where LAW represents weldability, , and These are spatter weight, porosity weight, and weld weight, respectively. , and These are information on spatter size, porosity size, and weld buildup size obtained by welding according to laser welding parameters. , and These are preset splash size information, preset porosity size information, and preset weld deposit size information, respectively; Specifically, based on laser welding data records of similar parts, multiple historical laser welding parameters whose distribution dispersion meets the dispersion requirement are selected as historical populations, including: Based on the laser welding data records of similar parts, a set of sample laser welding parameters, as well as a set of sample spatter scale information, sample porosity scale information, and sample weld accumulation scale information were obtained. The sample laser welding parameter set includes a set of sample welding power and a set of sample welding speed. Based on the sample welding power set and sample welding speed set, the welding power dispersion parameter and welding speed dispersion parameter are calculated. Based on the sample spatter scale information set, sample porosity scale information set, and sample weld accumulation scale information set, the sample welding fitness set is calculated based on the laser welding function, and the welding fitness dispersion parameter is calculated. Based on the welding power dispersion parameter, welding speed dispersion parameter, and welding fitness dispersion parameter, the historical discrete parameter is calculated and it is determined whether it is greater than the discrete parameter threshold. If it is, the sample laser welding parameter set is taken as the historical population; otherwise, the sample laser welding parameter set is re-acquired. Specifically, based on the constraints and laser welding function, and according to the historical population and part characteristic data, the laser welding parameters are optimized to obtain the optimal laser welding parameters, which, as the optimization result, include: Based on the constraints and the number of laser welding parameters in the historical population, multiple new laser welding parameters are randomly generated to construct a basic population, and an index relationship with the historical population is randomly constructed. Using the historical population as the adjustment direction, and according to the index relationship, multiple laser welding parameters within the basic population are adjusted and updated to obtain an updated population; Based on the part feature data, simulated laser welding is performed on multiple laser welding parameters in the updated population to obtain multiple simulated spatter scale information, multiple simulated porosity scale information and multiple simulated weld accumulation scale information. Based on multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information, the impact on the sealing and smoothness of the target part is predicted, and multiple impact ranges are obtained. Based on the magnitude of the impact range, several laser welding parameters in the update population are randomly selected and cross-updated with several laser welding parameters in the historical population to obtain a cross-population. Simulated laser welding was performed based on multiple laser welding parameters within the cross-population and historical population. The welding fitness was calculated by combining the laser welding function, and the laser welding parameters with higher welding fitness were retained to obtain the updated historical population. Based on the updated historical population, iterative optimization is performed until convergence, and the laser welding parameters with the highest welding fitness within the final historical population are output, thus obtaining the optimal laser welding parameters.

2. The laser welding parameter optimization method based on part features according to claim 1, characterized in that, Based on laser welding data records of similar parts to the target part, constraints are set for laser welding parameters, including: Based on laser welding data records of similar parts, welding power range and welding speed range are constructed; The welding power in the laser welding parameters falls within the welding power range as a welding power constraint, and the welding speed falls within the welding speed range as a welding speed constraint. These constraints are then set.

3. The laser welding parameter optimization method based on part features according to claim 1, characterized in that, Based on the component feature data, simulated laser welding is performed on multiple laser welding parameters within the updated population to obtain multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information, including: Based on laser welding data records of similar parts, a set of sample zinc plating characteristic data, a set of sample composition characteristic data, a set of sample laser welding parameters, a set of sample spatter scale information, a set of sample porosity scale information, and a set of sample simulated weld accumulation scale information are obtained as welding simulation construction data; Using the welding simulation data, a laser welding simulator is constructed. Multiple laser welding parameters in the updated population are combined with part feature data to simulate laser welding, thereby obtaining multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld accumulation scale information.

4. The laser welding parameter optimization method based on part features according to claim 1, characterized in that, Based on multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld seam accumulation scale information, the impact on the sealing and smoothness of the target part is predicted, resulting in multiple impact magnitudes. Based on the magnitude of the impact magnitude, several laser welding parameters within the current population are randomly selected for cross-updating with several laser welding parameters from the historical population, including: Based on the finished product processing data records of similar parts of the target part, obtain the sample spatter scale information set, sample porosity scale information set, sample simulated weld accumulation scale information set, as well as the sample sealing performance influence amplitude set and sample smoothness influence amplitude set; Using the sample spatter scale information set, sample porosity scale information set, sample simulated weld accumulation scale information set, sample sealing performance influence amplitude set, and sample smoothness influence amplitude set, a part finished product influence analyzer is constructed. Based on the multiple simulated spatter scale information, multiple simulated porosity scale information, and multiple simulated weld accumulation scale information, multiple sealing performance influence amplitudes and multiple smoothness influence amplitudes are analyzed and predicted, and multiple influence amplitudes are obtained by weighted calculation. Based on the multiple influence magnitudes, multiple crossover probabilities are generated, wherein the magnitude of the influence magnitude and the magnitude of the crossover probability are positively correlated; Based on the aforementioned multiple crossover probabilities, several laser welding parameters within the updated population are randomly selected and their welding power or welding speed is cross-updated with several laser welding parameters within the historical population to obtain a crossover population. The number of selected laser welding parameters is less than the total number of individuals in the updated population.

5. A laser welding parameter optimization device based on part features, characterized in that, The apparatus is used to implement the laser welding parameter optimization method based on part features as described in any one of claims 1-4, and the apparatus comprises: The part feature data acquisition module is used to acquire part feature data of the target part to be welded, wherein the part feature data includes galvanizing feature data and composition feature data. The constraint setting module is used to set constraints on laser welding parameters based on laser welding data records of similar parts to the target part. The constraints include welding power constraints and welding speed constraints. A laser welding function construction module is used to construct a laser welding function that optimizes laser welding parameters in order to reduce the spatter, porosity and weld buildup during the laser process. The historical laser welding parameter selection module is used to select multiple historical laser welding parameters whose distribution dispersion meets the discreteness requirements based on the laser welding data records of the same type of parts, as a historical population. A laser welding parameter optimization module is used to optimize laser welding parameters based on the constraints and laser welding function, according to the historical population and part feature data, to obtain the optimal laser welding parameters as the optimization result. The optimization process includes cross-updating, which is selected based on the influence of welding power and welding speed on spatter, porosity and weld seam within the laser welding parameters.

Images