Parameter control method and system in friction stir welding process

Abstract

The invention provides a parameter control method and system in the friction stir welding process, and relates to the technical field of welding equipment of metal materials. The method comprises the steps that before friction stir welding is started, working condition information of a to-be-welded workpiece is input into a welding process simulation model for virtual test welding; screening out a plurality of groups of standard process parameter combinations meeting preset welding quality requirements; in the friction stir welding process, the change trend of all the technological parameters is determined according to the real-time technological parameter combination within the first preset time period; and if it is determined that the deviation risk exists in the current welding state within the second preset time period in the future according to the change trend, the standard technological parameter combination with the maximum similarity with the real-time technological parameter combination in the standard technological parameter library is determined as the target technological parameter combination to conduct parameter control adjustment on the welding equipment. By implementing the method, the welding defects caused by sudden change of the process window can be overcome, and the stability of the welding process and the consistency of the welding seam quality are improved.

Description

Technical Field

[0001] This application relates to the technical field of welding equipment for metal materials, and particularly to a method and system for controlling parameters in the friction stir welding process. Background Art

[0002] As an efficient and green solid-phase joining process, friction stir welding has been widely used in the fields of aerospace, rail transit, automobile manufacturing, etc. due to its superior weld quality and low energy consumption. With the continuous improvement of the requirements for the mechanical properties and reliability of welded structural parts in engineering, the refinement and intelligentization of the friction stir welding process control have gradually become an important development direction in the industry. Especially in the scenarios of complex components and mass production, how to improve the stability of the welding process and the consistency of welds has become a technical problem that urgently needs to be solved.

[0003] In the related art, to achieve precise control of the friction stir welding process, key process parameters such as welding temperature, tool rotation speed, axial pressure, welding speed, etc. are generally collected in real time by sensors, and the collected data is fed back to the control system in real time. The control system adjusts the parameters of the welding equipment according to the preset process window and control strategy to maintain the process parameters within the target range. For example, when it is detected that the welding temperature or the downward pressure changes, the control system can automatically adjust the tool rotation speed or the feed speed to strive to ensure the stable operation of the welding process and the controllability of the weld quality. Such methods can better cope with the fluctuations of parameters in the conventional production process and meet the process requirements of most application scenarios.

[0004] However, the dynamic parameter adjustment method based on real-time detection in the related art mainly relies on the process parameter information collected at the current moment for immediate adjustment, and it is difficult to cope with the influence of the coupled changes of multiple factors on the process window in the complex working conditions where multiple factors such as material properties, structural parameters, welding environment, and equipment status change simultaneously. When factors such as material batch changes, structural form mutations, rapid fluctuations in environmental temperature, or abnormal equipment status occur superimposed, the process parameters may undergo non-linear mutations, resulting in the existing real-time adjustment strategies being difficult to adapt to these rapid changes in a timely and accurate manner, and then adverse phenomena such as process window mutations, increased welding defects, and unstable welding quality occur during the welding process. Summary of the Invention

[0005] This application provides a method and system for controlling parameters in the friction stir welding process to address the problem that the process window may mutate due to changes in multiple factors such as material properties, structural parameters, welding environment, and equipment status during the friction stir welding process, thereby causing welding defects and unstable welding quality.

[0006] In a first aspect, the present application provides a method for controlling parameters in a friction stir welding process, which is applied to a friction stir welding control system. The method includes: Before the friction stir welding starts, input the working condition information of the workpiece to be welded into the welding process simulation model for virtual welding test, and screen out multiple sets of standard process parameter combinations that meet the preset welding quality requirements to obtain a standard process parameter library. The working condition information includes the material properties, structural parameters, welding environment information, and welding equipment status parameters of the workpiece to be welded; During the friction stir welding process, collect the process parameter information in real time to obtain a real-time process parameter combination; Determine the change trend of each process parameter based on the real-time process parameter combination within a first preset time period; If it is determined based on the change trend that there is a deviation risk in the current welding state within a future second preset time period, then determine the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination. The target process parameter combination is used to control and adjust the parameters of the welding equipment.

[0007] Through the above embodiments, the system conducts simulation virtual welding tests on different working condition information before welding, establishes a standard process parameter library, and collects process parameters in real time during welding. By combining the parameter change trend to predict future working conditions, it can timely identify the deviation risk of the welding state and dynamically select the optimal parameter combination for control and adjustment. This method breaks through the limitation of traditional reliance on real-time parameter adjustment and difficulty in dealing with complex multi-factor coupling changes, can cope with welding defects caused by sudden changes in the process window, and improves the stability of the welding process and the consistency of the weld quality.

[0008] In some embodiments, the step of inputting the working condition information of the workpiece to be welded into the welding process simulation model for virtual welding test and screening out multiple sets of standard process parameter combinations that meet the preset welding quality requirements specifically includes: Determine the fluctuation range of each working condition parameter based on the working condition information and the historical production data of workpieces of the same type as the working condition information; Divide multiple working condition parameter combinations within the fluctuation range to obtain multiple sets of derived working condition information; Input multiple sets of the derived working condition information into the welding process simulation model respectively, and screen out the standard process parameter combinations that meet the preset welding quality requirements under each set of derived working condition information.

[0009] Through the above embodiments, the system analyzes the working condition information and historical production data, divides the fluctuation range of the working condition parameters, and screens multiple groups of standard process parameter combinations that meet the quality requirements in the simulation model. This step expands the coverage range of the process parameters and enhances the representativeness of the parameter library. The constructed standard process parameter library can better adapt to complex working conditions such as different batches of materials and structural changes in actual production, and plays a supporting role in the parameter adjustment and quality guarantee of the subsequent welding process.

[0010] In some embodiments, after the step of inputting the multiple groups of the derived working condition information into the welding process simulation model respectively and screening out the standard process parameter combinations that meet the preset welding quality requirements under each derived working condition information, it specifically includes: Obtain the first quality score corresponding to each of the standard process parameter combinations, where the first quality score is calculated by weighted calculation based on the distance between each process parameter and the corresponding index in the preset welding quality requirements; Adjust the first quality score according to the importance weight corresponding to each derived working condition information to obtain the second quality score corresponding to each of the standard process parameter combinations; Select the standard process parameter combinations that meet the corresponding preset quantity ratio and have the largest second quality score among the multiple standard process parameter combinations corresponding to each derived working condition information to construct the standard process parameter library.

[0011] Through the above embodiments, the system performs multi-dimensional quality scoring on the standard process parameter combinations and performs secondary optimization in combination with the importance weight of the working condition information, and finally optimally selects the parameter combinations with high reliability and high adaptability to enter the parameter library. This method not only improves the accuracy of parameter screening, but also can dynamically adjust the structure of the parameter library according to the differences in actual working conditions, which helps to improve the quality consistency and stability of the welding process.

[0012] In some embodiments, the step of determining the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination specifically includes: Based on the real-time process parameter combination, calculate the time required for parameter adjustment between each standard process parameter combination in the standard process parameter library and the real-time process parameter combination respectively; Screen out the standard process parameter combinations in the standard process parameter library where the time required for parameter adjustment is less than or equal to the preset time threshold to obtain the first candidate parameter combination set; Screen out the standard process parameter combinations with a second quality score greater than the preset score threshold from the first candidate parameter combination set to obtain the second candidate parameter combination set; Determine the standard process parameter combination with the shortest time required for parameter adjustment in the second set of candidate parameter combinations as the target process parameter combination.

[0013] Through the above embodiments, the system calculates multiple screening conditions such as the time required for parameter adjustment, the set time threshold, and the quality score threshold, ensuring that during the actual welding process, the selected target parameter combination is not only highly similar to the real-time process but also can complete the adjustment within a limited time, guaranteeing the timeliness and efficiency of parameter switching. This multi-level and multi-constraint screening mechanism improves the response efficiency and parameter adaptation accuracy during the welding process and can reduce the risk of welding defects caused by slow parameter switching.

[0014] In some embodiments, after the step of determining the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination, it further includes: Adjust the welding process simulation model according to the current working condition information; Input the target process parameter combination into the adjusted welding process simulation model to obtain a new first quality score; Determine whether the new first quality score is less than the preset score threshold; If it is detected that the new first quality score is less than the preset score threshold, input the standard process parameter combination with the second highest similarity into the adjusted welding process simulation model to re-obtain a new first quality score for re-judgment.

[0015] Through the above embodiments, the system adjusts the simulation model in real time according to the current working condition information and verifies the quality score of the target parameter combination. If the score does not meet the standard, it automatically switches to the sub-optimal parameters for re-evaluation, realizing the closed-loop optimization of parameter adjustment. This dynamic feedback mechanism improves the self-adaptability and intelligence level of parameter selection, ensures that each parameter adjustment is verified by simulation, and maximally guarantees the reliability of the welding process and the final weld quality.

[0016] In some embodiments, after the step of determining the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination, it further includes: Obtain the proportion value of the target process parameters with deviation risks in the real-time process parameter combination; If it is detected that the proportion value exceeds the preset proportion threshold, use the target process parameter combination to perform an overall adjustment of all process parameters; If it is detected that the proportion value does not exceed the preset proportion threshold, determine the adjustment value of the process parameters with deviation risks based on historical production data and perform a local adjustment according to the adjustment value.

[0017] Through the above embodiments, the system proposes a control strategy that combines overall adjustment and local adjustment for different degrees of parameter deviation risks. When the deviation risk is large, it can quickly adjust the overall parameters to avoid large-area quality problems; while when the risk is small, it makes local fine-tuning of specific parameters based on historical data to achieve more refined process control. This hierarchical adjustment strategy enhances the flexibility and pertinence of parameter control, not only improving the adaptability of welding production, but also effectively reducing human intervention and operation difficulty.

[0018] In some embodiments, the step of determining the adjustment value of the process parameter with deviation risk based on historical production data and performing local adjustment according to the adjustment value specifically includes: Retrieving historical production data with a similarity less than a preset threshold to the working condition information to obtain target historical production data; Extracting the historical process parameter values and adjustment responses corresponding to the process parameters with deviation risk from the target historical production data to obtain a historical parameter dataset, where the adjustment response includes a historical adjustment value and a corresponding historical first quality score; Based on the working condition information and the historical parameter dataset, determining the adjustment value corresponding to the process parameter with deviation risk through an optimization algorithm; Performing local adjustment on the process parameter with deviation risk according to the adjustment value.

[0019] Through the above embodiments, the system realizes precise fine-tuning of the parameters with deviation risk by efficiently retrieving, extracting historical parameters and adjustment responses similar to the current working condition, and calculating the optimal adjustment value in combination with an optimization algorithm. This method exploits the value of historical data, makes parameter adjustment more scientific and traceable, improves the intelligent level of the welding process and the consistency of welding quality, and helps to ensure welding stability under complex working conditions.

[0020] In a second aspect, the present application provides a friction stir welding control system, where the friction stir welding control system includes: one or more processors and a memory; The memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions so that the friction stir welding control system can implement a parameter control method for a friction stir welding process provided by the above embodiments, which will not be elaborated here.

[0021] In a third aspect, the present application provides a computer-readable storage medium including instructions that, when running on a friction stir welding control system, enable the friction stir welding control system to implement a method for controlling parameters in a friction stir welding process provided in the above embodiments, which will not be elaborated here.

[0022] In a fourth aspect, the present application provides a computer program product that, when running on a friction stir welding control system, enables the friction stir welding control system to implement a method for controlling parameters in a friction stir welding process provided in the above embodiments, which will not be elaborated here.

[0023] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Through working condition information simulation, parameter trend prediction, and dynamic parameter matching, active identification and timely response to the risk of deviation from the welding state are achieved. Compared with the traditional passive adjustment relying only on real-time feedback, this solution can improve the stability of the welding process and the consistency of weld quality in a complex and variable production environment, and cope with the defect risk brought by sudden changes in the process window.

[0024] 2. Through a hierarchical quality scoring and weight adjustment mechanism, an optimal target process parameter combination with both representativeness and high adaptability is selected, improving the coverage and response ability of the standard process parameter library to abnormal working conditions and complex environments.

[0025] 3. Combining parameter adjustment time, quality scoring, multi-level candidate screening, and an optimization algorithm driven by historical data, rapid and accurate matching and hierarchical adjustment of target parameters are achieved. Especially under the hierarchical judgment of parameter deviation risk, overall or partial adjustment can be selected according to the risk level, and the adjustment value can be optimized through historical data feedback closed-loop to ensure the scientific and effective adjustment plan. This multi-level and dynamic closed-loop control mechanism improves the response speed and adjustment accuracy of the welding process to sudden abnormalities and complex working conditions, ensuring the consistency and reliability of welding quality. Description of the Drawings

[0026] Figure 1 is a flowchart of a method for controlling parameters in a friction stir welding process according to an embodiment of the present application; Figure 2 is another flowchart of a method for controlling parameters in a friction stir welding process according to an embodiment of the present application; Figure 3 is a flowchart of a process for adjusting process parameters by a friction stir welding control system according to an embodiment of the present application; Figure 4 is a schematic structural diagram of a physical device of a friction stir welding control system according to an embodiment of the present application. Specific Embodiments

[0027] The terms used in the following embodiments of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. As used in the specification and appended claims of this application, the singular forms "a", "an", "the", "above-mentioned", "said", and "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations including one or more of the listed items.

[0028] Hereinafter, the terms "first" and "second" are only for descriptive purposes and cannot be construed as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of this application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0029] For ease of understanding, the method provided in this embodiment is described in terms of a process below. Please refer to Figure 1 , which is a schematic flow diagram of a method for controlling parameters in a friction stir welding process in an embodiment of this application.

[0030] S101. Before the friction stir welding starts, input the working condition information of the workpiece to be welded into the welding process simulation model for virtual welding tests, and screen out multiple groups of standard process parameter combinations that meet the preset welding quality requirements to obtain a standard process parameter library.

[0031] Among them, the working condition information refers to a set of various parameters directly related to the welding process, including material properties (such as material, hardness, thermal conductivity), structural parameters (such as plate thickness, groove form, butt gap), welding environment information (such as ambient temperature, humidity, gas protection conditions), and welding equipment status parameters (such as the wear degree of the stirring head, main shaft motor power); the welding process simulation model is used to represent a computer model established based on welding physical mechanisms (such as heat generation by friction, plastic metal flow), simulate the welding process by inputting the working condition information, and predict results such as weld formation, temperature field distribution, stress and strain, etc.; the standard process parameter combination represents a combination of parameters such as the rotation speed of the stirring head, welding speed, and axial pressure that meet the preset welding quality requirements (such as no cracks, incomplete penetration, pores and other defects, and the mechanical properties meet the standards).

[0032] Specifically, in the welding preparation stage, the operator first collects the complete working condition information of the workpiece to be welded, including the chemical composition of the material, geometric dimensions, temperature and humidity data of the welding environment, and the current state of the welding equipment (such as the wear amount of the stirring head and the working state of the cooling system). Subsequently, this information is input into a pre-trained welding process simulation model (such as a model based on finite element analysis or a machine learning model). The model simulates the welding process under different combinations of process parameters through algorithms and outputs the weld quality prediction results (such as nugget size, hardness distribution, defect probability). The system filters out all parameter combinations that meet the requirements according to the preset quality standards (such as industry specifications or enterprise internal standards) and stores them in the standard process parameter library for subsequent welding process calls.

[0033] Optionally, the system can use commercial welding simulation software (such as Simufact Welding, ANSYS Workbench) to build a simulation model, perform mesh generation, boundary condition setting, and solution calculation after inputting the working condition information; or use a machine learning model (such as a neural network) to train the historical working condition information and corresponding qualified process parameters to establish a working condition-parameter mapping relationship and automatically generate parameter combinations that meet the requirements; or through the orthogonal experimental design method, generate multiple groups of experimental combinations within the fluctuation range of the working condition parameters, and input them into the simulation model one by one to screen out qualified parameters, which is not limited here.

[0034] S102. During the friction stir welding process, the process parameter information is collected in real time to obtain the real-time process parameter combination.

[0035] Among them, the process parameter information refers to the process parameters that directly affect the welding quality, including the stirring head rotation speed, welding speed, axial pressure, welding temperature, torque, etc.

[0036] Specifically, the system collects the original data of each process parameter in real time through an integrated sensor network (such as a rotation speed sensor, a pressure sensor, a temperature sensor). After the data is conditioned by a signal conditioning module (such as filtering, amplification) and analog-to-digital conversion, it is transmitted to the central processing unit (CPU) or digital signal processor (DSP) of the control system. The processor combines the parameter values at the same moment into a real-time process parameter combination and stores it in the system memory or database for subsequent step calls and analysis.

[0037] S103. Determine the change trend of each process parameter based on the real-time process parameter combination within the first preset time period.

[0038] Specifically, the system extracts data points within a first preset time period (such as 300 data points in the past 30 seconds) from the historical data of real-time process parameter combinations, and conducts trend analysis on each process parameter (such as rotational speed, temperature) respectively. For example, for the rotational speed parameter, the time-rotational speed curve is fitted by the least squares method, and the slope is calculated to determine whether the rotational speed is increasing, decreasing, or remaining stable; for the temperature parameter, the moving average method is used to smooth the data, and it is observed whether the fluctuation range gradually expands. The analysis results are output in the form of a trend vector (such as rotational speed change rate +5 rpm / s, temperature fluctuation amplitude ±10 °C) for subsequent deviation risk judgment.

[0039] S104. If it is determined based on the change trend that there is a deviation risk in the current welding state within a second preset time period in the future, then the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library is determined as the target process parameter combination.

[0040] Among them, the deviation risk indicates the risk that the welding state may exceed the preset process window (such as the parameters exceeding the qualified range), resulting in welding defects (such as lack of fusion, excessive flash). It is predicted through trend analysis whether the parameters will reach the risk threshold within the second preset time period in the future (such as 5 seconds, 10 seconds).

[0041] Specifically, the system predicts the values of each process parameter within the second preset time period in the future according to the change trend (such as through linear extrapolation or model prediction), and compares them with the boundaries of the preset process window (such as the allowable range of rotational speed 800 - 1200 rpm, the allowable range of temperature 400 - 500 °C). If the predicted value exceeds the boundary, it is determined that there is a deviation risk. At this time, the system calls the similarity calculation algorithm, compares the real-time process parameter combination with all combinations in the standard process parameter library, calculates the distance (such as Euclidean distance) or similarity coefficient (such as cosine similarity) between each standard combination and the real-time combination, screens out the combination with the smallest distance or the largest similarity as the target process parameter combination, and then uses the target process parameter combination to adjust the parameters of the welding equipment.

[0042] Optionally, the system can also calculate the time required for each standard process parameter combination in the standard process parameter library to switch to the real-time parameters based on the real-time process parameter combination, through the dynamic model or historical data of the welding equipment. For example, calculate the rotation speed adjustment time according to the motor response curve, the pressure adjustment delay, etc., to obtain the parameter adjustment time required for each combination; then compare these times with the preset time threshold, screen out the combinations with the adjustment time less than or equal to the threshold to form the first candidate parameter combination set, and exclude the combinations that cannot complete the adjustment in time; then extract the second quality score of each combination from the first candidate set, compare it with the preset score threshold, and retain the combinations with the score greater than the threshold to obtain the second candidate parameter combination set, ensuring that the selected combinations have both timeliness and quality assurance; finally, sort the combinations in the second candidate set in ascending order according to the parameter adjustment time required, select the combination with the shortest time as the target process parameter combination, and if the times are the same, further compare the quality scores and preferentially select the combination with the higher score, so as to achieve a fast and high-quality parameter adjustment response.

[0043] In the above embodiment, the system conducts simulation virtual welding on different working condition information before welding, establishes a standard process parameter library, and in real time collects process parameters during welding. By combining the parameter change trend to predict future working conditions, it can timely identify the deviation risk of the welding state and dynamically select the optimal parameter combination for control and adjustment. This method breaks through the limitation of traditional reliance on real-time parameter adjustment and difficulty in coping with complex multi-factor coupling changes, can handle welding defects caused by sudden changes in the process window, and improves the stability of the welding process and the consistency of the weld quality.

[0044] The following further describes the method provided in this embodiment in a more specific process. Please refer to Figure 2 , which is another process schematic diagram of a parameter control method for a friction stir welding process in an embodiment of the present application.

[0045] S201. Determine the fluctuation range of each working condition parameter according to the working condition information and the historical production data of the workpieces of the same type as the working condition information, and divide multiple working condition parameter combinations to obtain multiple groups of derived working condition information.

[0046] Among them, the same-type workpieces refer to the welded workpieces with similar material properties and structural parameters (such as plate thickness, groove form) to the workpieces to be welded. For example, different specifications of profiles made of the same material aluminum alloy; the fluctuation range of working condition parameters refers to the change interval that each working condition parameter (such as material hardness, ambient temperature) may appear in actual production. For example, the fluctuation range of the aluminum alloy plate thickness is 2-4 mm, and the fluctuation range of the ambient temperature is 15-35 °C; the derived working condition information is used to represent multiple groups of simulated working conditions generated by parameter combination based on the original working condition information. For example, the original working condition is a plate thickness of 3 mm and a temperature of 25 °C, and the derived working conditions can include combinations such as a plate thickness of 2.5 mm / a temperature of 20 °C, a plate thickness of 3.5 mm / a temperature of 30 °C, etc.

[0047] This step is executed before the virtual welding test in step S101 and is applicable to scenarios where the material batches of workpieces are unstable, the production environment changes frequently, or the equipment status fluctuates. The purpose is to expand the coverage of working conditions by analyzing historical data and improve the robustness of the standard process parameter library.

[0048] Specifically, the system first retrieves the workpiece data of the same type as the current working condition information (such as welding records of the same material and similar structure) from the historical database, and extracts the actual value ranges of each working condition parameter (such as the minimum plate thickness of 2.1 mm and the maximum plate thickness of 3.9 mm). Combining the design tolerance of the current workpiece (such as the nominal plate thickness of 3 mm ± 0.5 mm) and process experience, the final fluctuation range of each parameter is determined. Subsequently, a parameter space partitioning algorithm (such as uniform sampling, Latin hypercube sampling) is used to generate multiple combinations of working condition parameters within the fluctuation range. Each combination includes specific values of material properties, structural parameters, environmental information, and equipment status, forming a set of derived working condition information.

[0049] S202. Input the multiple groups of derived working condition information into the welding process simulation model respectively, and screen out the standard process parameter combinations that meet the preset welding quality requirements under each derived working condition information.

[0050] Among them, the preset welding quality requirements refer to the preset welding quality evaluation criteria, including weld formation indexes (such as the nugget width ≥ 8 mm, flash height ≤ 0.5 mm), mechanical property indexes (such as tensile strength ≥ 200 MPa, elongation rate ≥ 15%), and defect grades (such as porosity ≤ 1%).

[0051] The system inputs each set of derived working condition information into the welding process simulation model in sequence. The model simulates the welding process according to the working condition parameters and outputs the prediction results of the weld quality. For each set of simulation results, the system checks against the preset quality requirements, such as checking whether the nugget size meets the standard and whether the crack risk is predicted. If the simulation results of a certain set of process parameter combinations meet all quality indicators, it is marked as qualified and included in the candidate set of standard process parameter combinations; if not, the parameters are adjusted and simulated again or the combination is eliminated.

[0052] In the above embodiment, the system divides the fluctuation range of the working condition parameters by introducing the analysis of the working condition information and historical production data, and screens multiple sets of standard process parameter combinations that meet the quality requirements in the simulation model. This step expands the coverage range of the process parameters and enhances the representativeness of the parameter library. The constructed standard process parameter library can better adapt to complex working conditions such as different batches of materials and structural changes in actual production, and plays a supporting role in the parameter adjustment and quality assurance of the subsequent welding process.

[0053] S203. Obtain the first quality score corresponding to each standard process parameter combination.

[0054] Specifically, the system establishes a quality evaluation model for each qualified standard process parameter combination, calculates the absolute distance between each process parameter (such as the stirring head speed, welding speed) and the corresponding target value in the preset quality indicators, and normalizes the distance to the interval [0, 1] (for example, if the allowable deviation is ±200 rpm and the actual deviation is 100 rpm, the normalized value is 0.5). Through a preset parameter weight matrix (such as the temperature weight is higher than the pressure weight), the normalized distances of each parameter are weighted and summed to obtain the first quality score. For example, if the scores of the speed, temperature, and pressure of a certain combination are 0.8, 0.9, and 0.7 respectively, and the weights are 0.3, 0.4, and 0.3 respectively, then the first quality score is 0.8×0.3 + 0.9×0.4 + 0.7×0.3 = 0.81.

[0055] S204. Adjust the first quality score according to the importance weight corresponding to each derived working condition information to obtain the second quality score corresponding to each standard process parameter combination.

[0056] The system assigns importance weights to each set of derivative working condition information (for example, the weight of common working conditions is 1.0, and the weight of rare working conditions is 0.6). The weight value reflects the priority of this working condition in actual production. Then, multiply the first quality score of each standard process parameter combination by the weight of the corresponding derivative working condition to obtain the second quality score. For example, a certain combination corresponds to a rare working condition (weight 0.6), and its first quality score is 0.8, then the second quality score is 0.8×0.6 = 0.48; another combination corresponds to a common working condition (weight 1.0), and the first quality score is 0.75, then the second quality score is 0.75×1.0 = 0.75. The latter has a higher score because the working condition is more common.

[0057] It should be noted that the importance weight of derivative working condition information represents the occurrence probability or influence degree of different derivative working conditions in actual production. For example, the weight of a high humidity environment (rare working condition) is relatively low, and the weight of a conventional room temperature environment (common working condition) is relatively high. The weight can be determined through statistical analysis of historical production data or process risk assessment, which is not limited here.

[0058] S205. Select the standard process parameter combinations that meet the corresponding preset quantity ratio and have the largest second quality score among the multiple standard process parameter combinations corresponding to each derivative working condition information to construct a standard process parameter library.

[0059] The system sorts the standard process parameter combinations under each derivative working condition information in descending order according to the second quality score, and intercepts the top N combinations (N = total number of combinations × ratio) according to the preset quantity ratio (such as 20%). For example, there are 50 qualified combinations under a certain working condition, and the preset ratio is 20%, then the 10 groups with the highest scores are selected for storage in the library. At the same time, the system checks the parameter differences of the combinations stored in the library for each working condition to ensure that the combinations under the same working condition cover different parameter adjustment directions (such as the coexistence of high speed - low speed and low speed - high speed combinations) to meet the diverse adjustment requirements in actual welding.

[0060] In the above embodiment, the system performs multi-dimensional quality scoring on the standard process parameter combinations and combines the importance weights of the working condition information for secondary optimization, and finally selects parameter combinations with high reliability and high adaptability to enter the parameter library. This method not only improves the accuracy of parameter screening, but also can dynamically adjust the structure of the parameter library according to the differences in actual working conditions, which helps to improve the quality consistency and stability of the welding process.

[0061] S206. After determining the target process parameter combination, adjust the welding process simulation model according to the current working condition information, and input the target process parameter combination into the adjusted welding process simulation model to obtain a new first quality score.

[0062] The system collects the real-time working condition information during the current welding process, compares it with the original working condition information input before welding, and identifies the changed parameters (such as ambient temperature, equipment wear). Then, according to the changed working condition parameters, the input parameters or boundary conditions of the welding process simulation model are adjusted. After the adjustment is completed, the target process parameter combination is input into the adjusted model, and the weld quality prediction results (such as temperature field distribution, stress value) are recalculated, and a new first quality score is obtained through distance weighting calculation based on the preset quality index.

[0063] S207. Determine whether the new first quality score is less than the preset score threshold.

[0064] The system numerically compares the newly calculated first quality score with the preset score threshold. For example, if the preset threshold is 0.75 and the new score is 0.82, it is determined to be qualified; if the new score is 0.70, it is determined to be unqualified. The setting of the score threshold is based on process standards and historical data, and is usually set slightly lower than the average score of the high-quality parameter combinations to reserve a safety margin, which is not limited here.

[0065] S208. Input the standard process parameter combination with the second highest similarity into the adjusted welding process simulation model to obtain a new first quality score again.

[0066] Specifically, when the system determines that the new first quality score is less than the preset threshold, it immediately retrieves the standard combination with the second highest similarity to the real-time process parameter combination from the standard process parameter library (for example, if the score of the original combination with the highest similarity does not meet the standard, the second closest combination is selected). Then, this second highest combination is input into the adjusted welding process simulation model (the model parameters have been updated according to the current working conditions), and its first quality score is recalculated. If the new score meets the standard, this combination is used as the new target parameter combination; if it still does not meet the standard, continue to retrieve the combination with the third highest similarity for verification until a qualified combination is found or an alarm of no valid solution is output.

[0067] S209. Use the target process parameter combination to control and adjust the welding equipment.

[0068] The system converts the parameter values in the target process parameter combination (such as rotational speed 1000 rpm, welding speed 60 mm / min, axial pressure 2.5 kN) into control signals recognizable by the equipment (such as analog voltage signals, digital pulse signals), and transmits them to the actuators of the welding equipment (such as servo motors, pressure controllers) through the industrial bus. After receiving the instruction, the equipment gradually adjusts the actual operating parameters through a closed-loop control system (such as PID adjustment) until the target value is reached. At the same time, the system continuously monitors the real-time data during the parameter adjustment process to ensure the smoothness of the adjustment process and avoid welding defects caused by sudden parameter changes.

[0069] In the above embodiments, the system adjusts the simulation model in real time according to the current working condition information, and conducts a quality score verification on the target parameter combination. If the score does not meet the standard, it automatically switches to the sub-optimal parameters for re-evaluation, realizing the closed-loop optimization of parameter adjustment. This dynamic feedback mechanism improves the adaptability and intelligence level of parameter selection, ensures that each parameter adjustment is verified by simulation, and maximally guarantees the reliability of the welding process and the quality of the final weld.

[0070] The following is a further and more specific process description of the method provided in this embodiment. Please refer to Figure 3 , which is a schematic flowchart of the process parameter adjustment of the friction stir welding control system in the embodiment of the present application.

[0071] S301. Obtain the proportion value of the target process parameters with deviation risks in the real-time process parameter combination.

[0072] Specifically, the system first extracts all process parameters (such as n parameters) from the real-time process parameter combination, and then, according to the change trends of each parameter determined in step S103, judges whether each parameter has a deviation risk one by one (that is, whether the predicted value exceeds the process window boundary). Count the number m of parameters with deviation risks, and calculate the proportion value as m / n×100%. For example, if the real-time process parameter combination includes 5 parameters: rotational speed, temperature, pressure, welding speed, and torque, and among them, the predicted values of rotational speed, temperature, and pressure exceed the process window, then the proportion value is 60%.

[0073] S302. Whether the proportion value exceeds the preset proportion threshold.

[0074] Specifically, the system numerically compares the proportion value calculated in S301 with the preset proportion threshold. For example, if the preset threshold is 50%, when the proportion value ≥ 50%, it is determined that the deviation risk is relatively large, and the overall adjustment process of step S303 is triggered; when the proportion value < 50%, it is determined that the deviation risk is relatively small, and the local adjustment process from step S304 to step S306 is triggered. Among them, the setting of the threshold needs to be combined with the process stability requirements and historical data. For example, in the aerospace scenario with extremely high requirements for welding quality, the threshold can be set to 40% to trigger more stringent adjustment measures in advance. It is not limited here.

[0075] S303. Use the target process parameter combination to conduct an overall adjustment on all process parameters.

[0076] This step is the same as step S209 and will not be elaborated here.

[0077] S304. Retrieve the historical production data with a similarity greater than the preset threshold to the working condition information to obtain the target historical production data.

[0078] Specifically, the system first extracts the key features in the current working condition information (such as material type, plate thickness, ambient temperature, equipment wear amount), then retrieves the working condition information of all historical records from the historical production database, and calculates the similarity between each record and the current working condition. The historical production data with similarity greater than the preset threshold (such as similarity > 80%) is screened according to the preset threshold as the target historical production data.

[0079] S305. Extract the historical process parameter values and adjustment responses corresponding to the process parameters at risk of deviation from the target historical production data to obtain a historical parameter dataset.

[0080] Specifically, the system traverses each record in the target historical production data according to the list of deviation parameters (such as rotation speed, temperature) determined in step S301, and extracts the historical values of these parameters in each record (such as the rotation speed of a certain record is 900 rpm and the temperature is 450 °C). At the same time, the adjustment operation information for these parameters in this record is extracted, such as the adjustment value (such as the rotation speed is adjusted from 900 rpm to 1000 rpm, and the adjustment value is + 100 rpm) and the quality score after adjustment (such as the first quality score is 0.85). These information are organized into a structured dataset, for example, stored in tabular form, with columns of parameter name, historical value, adjustment value, and quality score, and rows of each historical record.

[0081] S306. Based on the working condition information and the historical parameter dataset, determine the adjustment value corresponding to the process parameter at risk of deviation through an optimization algorithm for local adjustment.

[0082] Specifically, the system inputs the current working condition information (such as material properties, structural parameters, etc.) and the historical parameter dataset into the optimization algorithm model. The model analyzes the adjustment rules of the deviation parameters under different working conditions through training or calculation, and outputs the optimal adjustment value for the current deviation parameters. For example, use a random forest model to train the historical data, input the current working condition characteristics and the current values of the deviation parameters, and predict the best adjustment amount (such as the predicted adjustment value is - 8 °C when the temperature deviates). Then, the system converts the adjustment value into an equipment control instruction, and only adjusts the parameters at risk of deviation, such as gradually adjusting the temperature from 460 °C to 452 °C (current value + adjustment value = 460 - 8 = 452 °C) through a PID controller, while keeping other parameters (such as rotation speed, pressure) unchanged.

[0083] In the above embodiments, the system proposes a control strategy that combines overall adjustment and local adjustment for different degrees of parameter deviation risks. When the deviation risk is large, it can quickly adjust the overall parameters to avoid large-area quality problems; while when the risk is small, it locally fine-tunes the specific parameters based on historical data to achieve more refined process control. This hierarchical adjustment strategy enhances the flexibility and pertinence of parameter control, not only improving the adaptability of welding production, but also effectively reducing the need for human intervention and operation difficulty.

[0084] The friction stir welding control system according to an embodiment of the present invention is applied to an electronic device. Figure 4 The schematic diagram of the architecture of the electronic device suitable for implementing the embodiment of the present invention is shown.

[0085] It should be noted that Figure 4 The shown electronic device is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present invention.

[0086] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions (computer programs), or the relevant hardware can be controlled by instructions (computer programs). The instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. The electronic device of this embodiment includes a storage medium and a processor. Among them, multiple instructions are stored in the storage medium, and these instructions can be loaded by the processor to execute any step of the method provided by the embodiments of the present invention.

[0087] Specifically, the storage medium and the processor are electrically connected directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other through one or more signal lines. The storage medium stores computer-executable instructions for implementing the data access control method, including at least one software function module that can be stored in the storage medium in the form of software or firmware. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium. The storage medium can be, but is not limited to, a random access memory (Random Access Memory, abbreviated as RAM), a read-only memory (Read Only Memory, abbreviated as ROM), a programmable read-only memory (Programmable Read-Only Memory, abbreviated as PROM), an erasable programmable read-only memory (Erasable Programmable Read-Only Memory, abbreviated as EPROM), an electrically erasable programmable read-only memory (Electric Erasable Programmable Read-Only Memory, abbreviated as EEPROM), etc. Among them, the storage medium is used to store programs, and the processor executes the programs after receiving the execution instructions.

[0088] Further, the software programs and modules in the above storage medium may further include an operating system, which may include various software components and / or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and may communicate with various hardware or software components to provide a running environment for other software components. The processor can be an integrated circuit chip with signal processing capabilities. The above-mentioned processor can be a general-purpose processor, including a central processing unit (Central Processing Unit, abbreviated as CPU), a network processor (Network Processor, abbreviated as NP), etc., which can implement or execute the various methods, steps, and logic flow block diagrams disclosed in this embodiment. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0089] Since the instructions stored in the storage medium can execute the steps in any of the methods provided in the embodiments of the present invention, the beneficial effects of any of the methods provided in the embodiments of the present invention can be achieved. For details, please refer to the previous embodiments and will not be repeated here.

[0090] As described above, it is only the preferred specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A parameter control method for friction stir welding process, applied to a friction stir welding control system, characterized in that The method includes: Before the friction stir welding starts, input the working condition information of the workpiece to be welded into the welding process simulation model for virtual welding test, and screen out multiple groups of standard process parameter combinations that meet the preset welding quality requirements to obtain a standard process parameter library. The working condition information includes the material properties, structural parameters, welding environment information, and welding equipment status parameters of the workpiece to be welded; During the friction stir welding process, collect the process parameter information in real time to obtain the real-time process parameter combination; Determine the change trend of each process parameter according to the real-time process parameter combination within the first preset time period; If it is determined according to the change trend that there is a deviation risk in the current welding state within the next second preset time period, then determine the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination, and the target process parameter combination is used to control and adjust the parameters of the welding equipment.

2. The method according to claim 1, wherein The step of inputting the working condition information of the workpiece to be welded into the welding process simulation model for virtual welding test and screening out multiple groups of standard process parameter combinations that meet the preset welding quality requirements specifically includes: Determine the fluctuation range of each working condition parameter according to the working condition information and the historical production data of the workpieces of the same type as the working condition information; Divide multiple working condition parameter combinations within the fluctuation range to obtain multiple groups of derived working condition information; Input multiple groups of the derived working condition information into the welding process simulation model respectively, and screen out the standard process parameter combinations that meet the preset welding quality requirements under each derived working condition information.

3. The method according to claim 2, wherein After the step of inputting multiple groups of the derived working condition information into the welding process simulation model respectively and screening out the standard process parameter combinations that meet the preset welding quality requirements under each derived working condition information, it specifically includes: Obtain the first quality score corresponding to each standard process parameter combination, and the first quality score is calculated by weighted calculation based on the distance between each process parameter and the corresponding index in the preset welding quality requirements; Adjust the first quality score according to the importance weight corresponding to each derived working condition information to obtain the second quality score corresponding to each standard process parameter combination; Select the standard process parameter combinations that meet the corresponding preset quantity ratio and have the largest second quality score among the multiple standard process parameter combinations corresponding to each derived working condition information to construct the standard process parameter library.

4. The method according to claim 3, wherein The step of determining the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination specifically includes: Calculate the parameter adjustment time required between each standard process parameter combination in the standard process parameter library and the real-time process parameter combination respectively based on the real-time process parameter combination; Screen out the standard process parameter combinations in the standard process parameter library whose parameter adjustment time required is less than or equal to the preset time threshold to obtain the first candidate parameter combination set; Screen out the standard process parameter combinations in the first candidate parameter combination set whose second quality score is greater than the preset score threshold to obtain the second candidate parameter combination set; Determine the standard process parameter combination with the shortest parameter adjustment time in the second candidate parameter combination set as the target process parameter combination.

5. The method according to claim 1, wherein After the step of determining the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination, the following steps are further included: Adjust the welding process simulation model according to the current working condition information; Input the target process parameter combination into the adjusted welding process simulation model to obtain a new first quality score; Judge whether the new first quality score is less than the preset score threshold; If it is detected that the new first quality score is less than the preset score threshold, input the standard process parameter combination with the second highest similarity into the adjusted welding process simulation model to re-obtain the new first quality score for re-judgment.

6. The method according to claim 1, wherein After the step of determining the standard process parameter combination with the highest similarity to the real-time process parameter combination in the standard process parameter library as the target process parameter combination, the following steps are further included: Obtain the proportion value of the target process parameters with deviation risks in the real-time process parameter combination; If it is detected that the proportion value exceeds the preset proportion threshold, use the target process parameter combination to perform overall adjustment on all process parameters; If it is detected that the proportion value does not exceed the preset proportion threshold, determine the adjustment value of the process parameters with deviation risks based on historical production data, and perform local adjustment according to the adjustment value.

7. The method according to claim 6, wherein The step of determining the adjustment value of the process parameters with deviation risks based on historical production data and performing local adjustment according to the adjustment value specifically includes: Retrieve historical production data with a similarity less than the preset threshold to the working condition information to obtain target historical production data; Extract the historical process parameter values and adjustment responses corresponding to the process parameters with deviation risks from the target historical production data to obtain a historical parameter data set, and the adjustment response includes the historical adjustment value and the corresponding historical first quality score; Based on the working condition information and the historical parameter data set, determine the adjustment value corresponding to the process parameters with deviation risks through an optimization algorithm; Perform local adjustment on the process parameters with deviation risks according to the adjustment value.

8. A friction stir welding control system, characterized in that, The friction stir welding control system includes: one or more processors and a memory; The memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the friction stir welding control system to execute the method according to any one of claims 1-7.

9. A computer-readable storage medium, comprising instructions, characterized in that, When the instructions run on the friction stir welding control system, enable the friction stir welding control system to execute the method according to any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on the friction stir welding control system, enable the friction stir welding control system to execute the method according to any one of claims 1-7.

Images