A method and apparatus for optimizing process parameters for ultrasonic metal welding

Abstract

The application provides a process parameter optimization method and device for ultrasonic metal welding, which comprises the following steps: receiving multiple groups of process parameters to be optimized in multiple process stages of ultrasonic metal welding; inputting the multiple groups of process parameters to be optimized into a process parameter prediction layer to obtain multiple predicted process parameters corresponding to the multiple process stages; when the multiple predicted process parameters exceed a set parameter range, adjusting the multiple groups of process parameters to be optimized based on a parameter adjustment module to obtain multiple groups of optimized process parameters; and when the multiple predicted process parameters are within the set parameter range, inputting the multiple predicted process parameters into a welding quality prediction layer to obtain a predicted welding quality. The process parameter prediction layer can predict the process parameters of each process stage, and the parameter adjustment module can adjust the multiple groups of process parameters to be optimized based on the prediction results of the process parameters, so that the consistency of the optimized process parameters can be improved, and the accuracy of the predicted welding quality can be improved.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic metal welding technology, and specifically to a method and equipment for optimizing process parameters in ultrasonic metal welding. Background Technology

[0002] Ultrasonic Metal Welding (UMW) is a solid-state welding technology that utilizes high-frequency ultrasonic energy. Because UMW has low material dependence and minimizes the formation of intermetallic compounds and energy loss at the contact zone, it has been widely used in the production of automotive parts such as lithium-ion battery tabs, wire terminals, and electronic components. A major challenge for UMW technology is accurately assessing weld quality, which is closely related to the input or set process parameters. To ensure weld quality meets requirements, appropriate process parameters must be input or set.

[0003] The existing method for ensuring welding quality involves determining the process parameters corresponding to the desired welding quality during a debugging phase before actual welding. Specifically, different process parameters are manually input to obtain the corresponding welding quality. If the welding quality does not meet the requirements, the process parameters are manually adjusted until optimized process parameters corresponding to the desired welding quality are found. This method has the following technical problems: 1. Ultrasonic metal welding is a phased, sequential process with numerous process parameters, and these parameters are coupled between different stages, leading to a complex debugging process and low efficiency in process parameter optimization. 2. There are process parameters between the process parameters and the final quality prediction in ultrasonic metal welding. Existing technologies only consider the final quality during process parameter debugging, neglecting the process parameters. During production, due to inconsistencies in production materials, environment, and equipment, the same process parameters cannot guarantee the same product quality. This results in poor adaptability of the optimized process parameters, failing to meet welding quality requirements under arbitrary production conditions.

[0004] Therefore, there is an urgent need to provide a method and equipment for optimizing process parameters in ultrasonic metal welding, enabling automated optimization and improving the efficiency of process parameter optimization. Furthermore, the optimization process should consider process parameters to ensure compatibility between process parameters and factors such as production materials and the environment, thereby improving the quality of products produced using the optimized process parameters. Summary of the Invention

[0005] In view of this, it is necessary to provide a method and equipment for optimizing process parameters of ultrasonic metal welding, so as to solve the technical problems in the prior art, such as low optimization efficiency due to manual optimization of process parameters and failure to consider the process parameters of ultrasonic metal welding, resulting in low quality of products produced by optimizing process parameters.

[0006] In a first aspect, in order to solve the above-mentioned technical problems, the present invention provides a method for optimizing process parameters of ultrasonic metal welding. The method optimizes the process parameters of ultrasonic metal welding based on a process parameter optimization model. The process parameter optimization model includes a hierarchical analysis model and a parameter tuning module. The hierarchical analysis model includes a process parameter prediction layer and a welding quality prediction layer connected in sequence. The input and output of the parameter tuning module are respectively connected to the output and input of the process parameter prediction layer.

[0007] The method includes:

[0008] Receive multiple sets of process parameters to be optimized in multiple stages of ultrasonic metal welding process;

[0009] The multiple sets of process parameters to be optimized are input into the process parameter prediction layer to predict process parameters, thereby obtaining multiple predicted process parameters that correspond one-to-one with the multiple process stages.

[0010] When the multiple prediction process parameters exceed the set parameter range, the multiple sets of process parameters to be optimized are optimized based on the parameter tuning module to obtain multiple sets of optimized process parameters; the multiple prediction control parameters corresponding to the multiple sets of optimized process parameters are within the set parameter range.

[0011] When the multiple prediction process parameters are within the set parameter range, the multiple prediction process parameters are input into the welding quality prediction layer to perform quality prediction and obtain the predicted welding quality.

[0012] In one possible implementation, the plurality of process stages include a pressure-down stage, an ultrasonic vibration stage, a pressure-holding stage, and a post-weld treatment stage. Then, the plurality of process parameters to be optimized include pressure-down parameters to be optimized, ultrasonic vibration parameters to be optimized, pressure-holding parameters to be optimized, and post-weld treatment parameters to be optimized. The plurality of predicted process parameters include pressure-down process predicted parameters, ultrasonic vibration process predicted parameters, pressure-holding process predicted parameters, and post-weld treatment process predicted parameters.

[0013] In one possible implementation, the process parameter prediction layer includes a pressure-down process parameter prediction unit, an ultrasonic vibration process parameter prediction unit, a pressure-holding process parameter prediction unit, and a post-weld processing process parameter prediction unit.

[0014] The compression process parameter prediction unit is used to predict the process parameters of the compression stage based on the compression parameters to be optimized, and obtain the compression process prediction parameters.

[0015] The ultrasonic vibration process parameter prediction unit is used to predict the process parameters of the ultrasonic vibration stage based on the ultrasonic vibration parameters to be optimized and the compression process prediction parameters, so as to obtain the ultrasonic vibration process prediction parameters.

[0016] The pressure holding process parameter prediction unit is used to predict the process parameters of the pressure holding stage based on the pressure holding parameters to be optimized and the ultrasonic vibration process prediction parameters, and obtain the pressure holding process prediction parameters.

[0017] The post-weld processing parameter prediction unit is used to predict the process parameters of the post-weld processing stage based on the post-weld processing parameters to be optimized and the pressure holding process prediction parameters, so as to obtain the post-weld processing predicted parameters.

[0018] In one possible implementation, the welding quality prediction layer includes a pressure prediction unit, an ultrasonic vibration prediction unit, a pressure holding quality prediction unit, and a post-weld treatment quality prediction unit.

[0019] The compression quality prediction unit is used to predict the quality of the compression stage based on the compression process prediction parameters, and obtain the implicit information of the compression quality.

[0020] The ultrasonic vibration quality prediction unit is used to predict the quality of the ultrasonic vibration stage based on the ultrasonic vibration process prediction parameters and the implicit information of the pressure quality, so as to obtain the implicit information of ultrasonic vibration quality.

[0021] The pressure holding quality prediction unit is used to predict the quality of the pressure holding stage based on the pressure holding process prediction parameters and the implicit information of ultrasonic vibration quality, and to obtain the implicit information of pressure holding quality.

[0022] The post-weld treatment quality prediction unit is used to predict the quality of the post-weld treatment stage based on the prediction parameters of the post-weld treatment process and the implicit information of the pressure holding quality, so as to obtain the predicted welding quality.

[0023] In one possible implementation, the parameter tuning module includes a pressure parameter tuning unit, an ultrasonic vibration parameter tuning unit, a pressure holding parameter tuning unit, and a post-weld treatment parameter tuning unit.

[0024] The pressure parameter optimization unit is used to optimize the pressure parameter to be optimized when the pressure process prediction parameter exceeds the set pressure parameter range, so as to obtain the target pressure parameter; the pressure process prediction parameter corresponding to the target pressure parameter is within the set pressure parameter range.

[0025] The ultrasonic vibration parameter optimization unit is used to optimize the ultrasonic vibration parameters to be optimized when the predicted parameters of the ultrasonic vibration process exceed the set range of ultrasonic vibration parameters, so as to obtain the target ultrasonic vibration parameters; the target ultrasonic vibration parameters are within the set range of ultrasonic vibration parameters.

[0026] The pressure holding parameter optimization unit is used to optimize the pressure holding parameters to be optimized when the predicted parameters of the pressure holding process exceed the set pressure holding parameter range, so as to obtain the target pressure holding parameters; the target pressure holding parameters are within the set pressure holding parameter range.

[0027] The post-weld processing parameter optimization unit is used to optimize the post-weld processing parameters to be optimized when the predicted parameters of the post-weld processing process exceed the set post-weld processing parameter range, so as to obtain the target post-weld processing parameters; the target post-weld processing parameters are within the set post-weld processing parameter range.

[0028] In one possible implementation, the hierarchical analysis model further includes a process parameter input layer, which includes a pressure holding parameter input unit, an ultrasonic vibration parameter input unit, a pressure holding parameter input unit, and a post-weld treatment parameter input unit.

[0029] The pressure holding parameter input unit is used to input the pressure-down parameter to be optimized into the pressure-down process parameter prediction unit;

[0030] The ultrasonic vibration parameter input unit is used to input the ultrasonic vibration parameters to be optimized into the ultrasonic vibration process parameter prediction unit.

[0031] The pressure holding parameter input unit is used to input the pressure holding parameter to be optimized into the pressure holding process parameter prediction unit;

[0032] The post-weld processing parameter input unit is used to input the post-weld processing parameters to be optimized into the post-weld processing parameter prediction unit.

[0033] In one possible implementation, the hierarchical analysis model further includes a first multi-head self-attention layer disposed between the process parameter input layer and the process parameter prediction layer;

[0034] The first multi-head self-attention layer is used to perform self-attention learning on the multiple sets of process parameters to be optimized, and obtain attention process parameter features;

[0035] The process parameter prediction layer is used to predict process parameters based on the attention process parameter features, thereby obtaining the plurality of process prediction parameters.

[0036] In one possible implementation, the hierarchical analysis model further includes a second multi-head self-attention layer disposed between the process parameter prediction layer and the weld quality prediction layer;

[0037] The second multi-head self-attention layer is used to perform self-attention learning on the multiple process prediction parameters to obtain attention process parameter features;

[0038] The welding quality prediction layer is used to predict the quality of the attention process parameters and obtain the predicted welding quality.

[0039] In one possible implementation, the pressure-down parameters to be optimized include the pressure-down force, the ultrasonic vibration parameters to be optimized include the first amplitude of the amplitude transformer, the first ultrasonic vibration power, and the welding time, the pressure-holding parameters to be optimized include the pressure-holding time, and the post-weld processing parameters to be optimized include the second amplitude of the amplitude transformer, the second ultrasonic vibration frequency, and the post-processing time.

[0040] The predicted parameters for the pressing process include the peak value of the weld head depth and the average value of the weldment temperature during the pressing stage. The predicted parameters for the ultrasonic vibration process include the peak value of the weld head depth, the average value of the weld head depth, the standard deviation of the weld head depth, the peak value of the weldment temperature, the average value of the weldment temperature, the standard deviation of the weldment temperature, the peak value of the weld head amplitude, the average value of the weld head amplitude, and the standard deviation of the weld head amplitude during the ultrasonic vibration stage. The predicted parameters for the holding pressure process include the peak value of the weld head depth and the average value of the weldment temperature during the holding pressure stage. The predicted parameters for the post-weld treatment process include the average value of the weldment temperature, the peak value of the weld head amplitude, the average value of the weld head amplitude, and the standard deviation of the weld head amplitude during the post-weld treatment stage.

[0041] The predicted weld quality includes predicted joint tensile strength and predicted contact resistance.

[0042] Secondly, the present invention also provides a process parameter optimization device for ultrasonic metal welding, including a process parameter acquisition device and a process parameter optimization module;

[0043] The process parameter acquisition device is used to acquire multiple sets of process parameters to be optimized in multiple process stages of ultrasonic metal welding process.

[0044] The process parameter optimization module includes a memory and a processor, wherein,

[0045] The memory is used to store programs;

[0046] The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the ultrasonic metal welding process parameter optimization method described in any of the above possible implementations.

[0047] The beneficial effects of this invention are as follows: The ultrasonic metal welding process parameter optimization method provided by this invention achieves automatic optimization of ultrasonic metal welding process parameters through a process parameter optimization model, eliminating the need for manual parameter adjustment and improving the optimization efficiency of process parameters. Furthermore, this invention sets up a hierarchical analysis model including a process parameter prediction layer, which predicts process parameters for each process stage. A parameter tuning module is also included to tune multiple sets of process parameters to be optimized when multiple predicted process parameters exceed the set parameter range, obtaining optimized process parameters. This optimizes the matching degree between process parameters and production materials, environment, and equipment; that is, the optimized process parameters are optimized in real time as production materials, environment, and equipment change, reducing the adverse effects of material and environmental factors on production quality. This further improves the rationality of the determined optimized process parameters, ensuring the accuracy of ultrasonic metal welding quality prediction.

[0048] Furthermore, the hierarchical analysis model in this invention uses the concept of hierarchical analysis to characterize the nonlinear relationship between process parameters, process parameters, and welding quality. It decouples process parameters and process parameters based on process stages. For example, the process parameters and process parameters of the current process stage will not affect the previous process stage, thereby reducing the model solution speed and further improving the optimization efficiency of process parameters. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A schematic flowchart of an embodiment of the ultrasonic metal welding process parameter optimization method provided by the present invention;

[0051] Figure 2 A schematic diagram of an embodiment of the process parameter optimization model provided by the present invention;

[0052] Figure 3 This is a schematic diagram of an embodiment of the ultrasonic metal welding process parameter optimization device provided by the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0054] It should be understood that the illustrative drawings are not drawn to scale. The flowcharts used in this invention illustrate operations implemented according to some embodiments of the invention. It should be understood that the operations in the flowcharts may be implemented out of order, and steps without logical contextual relationships may be reversed or performed simultaneously. Furthermore, those skilled in the art, guided by the content of this invention, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor systems and / or microcontroller systems.

[0055] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0056] This invention provides a method and equipment for optimizing process parameters in ultrasonic metal welding, which will be described below.

[0057] Figure 1 This is a schematic flowchart of an embodiment of the ultrasonic metal welding process parameter optimization method provided by the present invention. Figure 2 A schematic diagram of an embodiment of the process parameter optimization model provided by the present invention is shown below. Figure 1 and Figure 2 As shown, the process parameter optimization model includes a hierarchical analysis model and a parameter tuning module. Figure 2 The module is composed of double-dotted-line units, and the hierarchical analysis model includes a process parameter prediction layer connected in sequence. Figure 2 The module consists of dotted and dashed units and a welding quality prediction layer. The input and output of the parameter tuning module are connected to the output and input of the process parameter prediction layer, respectively.

[0058] The optimization methods for ultrasonic metal welding process parameters include:

[0059] S101, Receive multiple sets of process parameters to be optimized in multiple process stages of ultrasonic metal welding process.

[0060] The number of multiple sets of process parameters to be optimized corresponds one-to-one with the number of multiple process stages. For example, if the ultrasonic metal welding process includes four stages, then the multiple sets of process parameters to be optimized include four sets, which correspond one-to-one with the process stages.

[0061] S102. Input multiple sets of process parameters to be optimized into the process parameter prediction layer to perform process parameter prediction, and obtain multiple predicted process parameters that correspond one-to-one with multiple process stages.

[0062] It should be noted that, due to the temporal nature of ultrasonic metal welding, the process parameter prediction layer should be a temporal prediction model, specifically a Long Short-Term Memory (LSTM) network model.

[0063] S103. When multiple prediction process parameters exceed the set parameter range, the parameter tuning module is used to tune multiple sets of process parameters to be optimized to obtain multiple sets of optimized process parameters; the multiple prediction control parameters corresponding to the multiple sets of optimized process parameters are within the set parameter range.

[0064] The parameter tuning module includes built-in optimization algorithms, such as particle swarm optimization, ant colony optimization, and gray wolf optimization, which aim to optimize the process parameters within a set parameter range and obtain the optimized process parameters.

[0065] In a specific embodiment of the present invention, the process of obtaining multiple sets of optimized process parameters in step S103 will go through at least one iteration. Each iteration results in multiple sets of iterated process parameters. Based on the iterated process parameters, the corresponding predictive control parameters are determined. It is determined whether the predictive control parameters are within the set parameter range. If so, the current iterated process parameters are the optimized process parameters.

[0066] S104. When multiple prediction process parameters are within the set parameter range, input the multiple prediction process parameters into the welding quality prediction layer to perform quality prediction and obtain the predicted welding quality.

[0067] It should be understood that the range of parameters can be set or adjusted according to the actual application scenario, and no specific limitation is made here.

[0068] The welding quality prediction layer should also be a time-series prediction model; specifically, the welding quality prediction layer should also be an LSTM model.

[0069] The memory units in the LSTM model employ an input gate, forget gate, and output gate structure to remember long-term information between time sequences and discard useless information, enabling it to effectively learn temporal relationships between high-dimensional data. The specific structure of the LSTM model will not be elaborated upon here.

[0070] It should be noted that the process parameter optimization model needs to be trained, verified, and tested based on actual data before it can be used.

[0071] Specifically, to prevent large fluctuations in validation scores due to insufficient validation set data points, k-fold cross-validation is used to train the model. During training, the root mean square error (RMSE) is used as the model's loss function. During parameter optimization, gradient descent is used to adjust the model parameters until the loop termination condition is met, at which point the final model parameters are saved, resulting in a usable optimized process parameter model.

[0072] It should be understood that the ultrasonic metal welding process parameter optimization method in the embodiments of the present invention can be implemented in any device based on the ultrasonic metal welding process parameter optimization method, such as ultrasonic metal welding control equipment. Specifically, the ultrasonic metal welding process parameter optimization method is stored in the aforementioned device as a pre-programmed program. When the device is started, the program is invoked, and the ultrasonic metal welding process parameter optimization method is implemented.

[0073] Compared with existing technologies, the ultrasonic metal welding process parameter optimization method provided in this invention achieves automatic optimization of ultrasonic metal welding process parameters through a process parameter optimization model, eliminating the need for manual parameter tuning and improving the optimization efficiency. Furthermore, this invention sets up a hierarchical analysis model including a process parameter prediction layer, which predicts process parameters for each process stage. A parameter tuning module is also included to tune multiple sets of process parameters when multiple predicted process parameters exceed a set range, obtaining optimized process parameters. This optimizes the matching degree between process parameters and production materials, environment, and equipment; that is, the optimized process parameters are optimized in real time as production materials, environment, and equipment change, reducing the adverse effects of material and environmental factors on production quality. This further improves the rationality of the determined optimized process parameters, ensuring the accuracy of ultrasonic metal welding quality prediction.

[0074] Furthermore, the hierarchical analysis model in this embodiment of the invention uses the concept of hierarchical analysis (AHP) to characterize the nonlinear relationship between process parameters, process parameters and welding quality, and decouples process parameters and process parameters based on process stages. For example, the process parameters and process parameters of the current process stage will not affect the previous process stage, thereby reducing the model solution speed and further improving the optimization efficiency of process parameters.

[0075] In some embodiments of the present invention, multiple process stages include a pressing stage, an ultrasonic vibration stage, a pressure holding stage, and a post-weld treatment stage.

[0076] (1) Pressing stage: In this stage, the welding head of the welding equipment applies a certain pressure to the metal workpiece to be welded placed below under the action of the cylinder or other driving device, so that the upper and lower metal workpieces are in close contact, and the pressure is maintained until the end of the holding stage. During this period, the key process parameter is the pressing pressure P.

[0077] (2) Ultrasonic vibration stage: The ultrasonic generator generates a high-frequency electrical signal, which is converted into mechanical vibration energy by the transducer. After the amplitude is amplified by the amplitude transformer, it is transmitted to the welding head, causing the welding head to generate high-frequency ultrasonic vibration. The parameters that affect the welding quality in this process are the first amplitude C1 of the amplitude transformer, the first ultrasonic vibration power W1, and the welding time T1.

[0078] (3) Pressure holding stage: When entering the pressure holding stage, the ultrasonic vibration stops, but the welding head still maintains the pressure on the workpiece, so that the welded joint gradually cools and solidifies under the pressure to form a stable solid connection. P remains unchanged in this stage. Another important parameter is the pressure holding time T2.

[0079] (4) Post-processing stage: To prevent the workpiece from sticking to the welding head, a small vibration is often applied to the welding head after the pressure holding is completed. This vibration can destroy any microscopic connection that may exist between the workpiece and the welding head, thereby separating them. This stage is similar to the ultrasonic vibration stage, and the main parameters are the second amplitude C2, the second ultrasonic vibration power W2, and the post-processing time T3.

[0080] The process parameters to be optimized in each of the above stages can be expressed as:

[0081]

[0082] In the formula X 1=P, X 2=(C1,W1,T2) T , X 3=T2, X 4 = (C2, W2, T3) T .

[0083] The multiple sets of process parameters to be optimized include the pressure-down parameters to be optimized. X 1. Ultrasonic vibration parameters to be optimized X 2. Pressure holding parameters to be optimized X 3. Post-weld treatment parameters to be optimized X 4.

[0084] Similarly, multiple prediction process parameters include prediction parameters for the pressing process, ultrasonic vibration process, pressure holding process, and post-weld treatment process.

[0085] In a specific embodiment of the present invention, the prediction parameters for the pressure-down process include the peak value of the weld head depth D1 and the average value of the weldment temperature E1 during the pressure-down stage; the prediction parameters for the ultrasonic vibration process include the peak value of the weld head depth D2, the average value of the weld head depth D3, the standard deviation of the weld head depth D4, the peak value of the weldment temperature E2, the average value of the weldment temperature E3, the standard deviation of the weldment temperature E4, the peak value of the weld head amplitude S1, the average value of the weld head amplitude S2, and the standard deviation of the weld head amplitude S3 during the ultrasonic vibration stage; the prediction parameters for the pressure-holding process include the peak value of the weld head depth D5 and the average value of the weldment temperature E5 during the pressure-holding stage; and the prediction parameters for the post-weld treatment process include the average value of the weldment temperature E6, the peak value of the weld head amplitude S4, the average value of the weld head amplitude S5, and the standard deviation of the weld head amplitude S6 during the post-weld treatment stage.

[0086] Specifically, the prediction process parameters are represented by Z, and are as follows:

[0087]

[0088] In the formula, Z1=(D1,E1) T , Z2=(D2,D3,D4,E2,E3,E4,S1,S2,S3) T Z3 = (D5, E5) T Z3 = (E6, S4, S5, S6) T .

[0089] In a specific embodiment of the present invention, predicting weld quality includes predicting the tensile strength of the joint and predicting the contact resistance. The predicted weld quality is represented by Y, specifically as follows:

[0090]

[0091] In the formula, F To predict the tensile strength of the joint; R To predict contact resistance.

[0092] Since parameters of one process can affect parameters of another process in actual production, to improve the accuracy of overall process parameter prediction, in some embodiments of the present invention, such as... Figure 2As shown, the process parameter prediction layer includes a pressure-down process parameter prediction unit, an ultrasonic vibration process parameter prediction unit, a pressure-holding process parameter prediction unit, and a post-weld treatment process parameter prediction unit.

[0093] The compression process parameter prediction unit is used to predict the process parameters of the compression stage based on the compression parameters to be optimized, and obtain the compression process prediction parameters.

[0094] The ultrasonic vibration process parameter prediction unit is used to predict the process parameters of the ultrasonic vibration stage based on the ultrasonic vibration parameters to be optimized and the compression process prediction parameters, so as to obtain the ultrasonic vibration process prediction parameters.

[0095] The pressure holding process parameter prediction unit is used to predict the process parameters of the pressure holding stage based on the pressure holding parameters to be optimized and the ultrasonic vibration process prediction parameters, so as to obtain the pressure holding process prediction parameters.

[0096] The post-weld processing parameter prediction unit is used to predict the process parameters of the post-weld processing stage based on the post-weld processing parameters to be optimized and the pressure holding process prediction parameters, and obtain the post-weld processing predicted parameters.

[0097] In the embodiments of the present invention, each subsequent parameter prediction unit considers the influence of the predicted process parameters from the previous parameter prediction unit when predicting process parameters, thereby further improving the accuracy of the predicted process parameters at each process stage. Furthermore, the orderly prediction can also improve the prediction efficiency of the process parameters.

[0098] In specific embodiments of the present invention, such as Figure 2 As shown, the welding quality prediction layer includes a pressure prediction unit, an ultrasonic vibration prediction unit, a pressure holding quality prediction unit, and a post-weld treatment quality prediction unit.

[0099] The compression quality prediction unit is used to predict the quality of the compression stage based on the compression process prediction parameters, and to obtain the implicit information of the compression quality.

[0100] The ultrasonic vibration quality prediction unit is used to predict the quality of the ultrasonic vibration stage based on the ultrasonic vibration process prediction parameters and the implicit information of the compression quality, thereby obtaining the implicit information of ultrasonic vibration quality.

[0101] The pressure holding quality prediction unit is used to predict the quality of the pressure holding stage based on the pressure holding process prediction parameters and the implicit information of ultrasonic vibration quality, thereby obtaining the implicit information of pressure holding quality.

[0102] The post-weld treatment quality prediction unit is used to predict the quality of the post-weld treatment stage based on the prediction parameters of the post-weld treatment process and the implicit information of the pressure holding quality, so as to obtain the predicted welding quality.

[0103] In this embodiment of the invention, each subsequent quality prediction unit considers the influence predicted by the previous quality prediction unit when predicting process parameters, thereby further improving the accuracy of the implicit quality information at each process stage and thus improving the accuracy of the final predicted welding quality.

[0104] In specific embodiments of the present invention, such as Figure 2 As shown, the parameter tuning module includes a pressure parameter tuning unit, an ultrasonic vibration parameter tuning unit, a pressure holding parameter tuning unit, and a post-weld treatment parameter tuning unit.

[0105] The downpressure parameter optimization unit is used to optimize the downpressure parameters to be optimized when the predicted downpressure process parameters exceed the set downpressure parameter range, so as to obtain the target downpressure parameters; the predicted downpressure process parameters corresponding to the target downpressure parameters are within the set downpressure parameter range.

[0106] The ultrasonic vibration parameter optimization unit is used to optimize the ultrasonic vibration parameters to be optimized when the predicted parameters of the ultrasonic vibration process exceed the set range of ultrasonic vibration parameters, so as to obtain the target ultrasonic vibration parameters; the target ultrasonic vibration parameters are within the set range of ultrasonic vibration parameters.

[0107] The pressure holding parameter optimization unit is used to optimize the pressure holding parameters to obtain the target pressure holding parameters when the predicted parameters of the pressure holding process exceed the set pressure holding parameter range; the target pressure holding parameters are within the set pressure holding parameter range.

[0108] The post-weld processing parameter optimization unit is used to optimize the post-weld processing parameters to obtain the target post-weld processing parameters when the predicted parameters of the post-weld processing process exceed the set post-weld processing parameter range; the target post-weld processing parameters are within the set post-weld processing parameter range.

[0109] Since inaccurate predictions of parameters in subsequent processes can result from the predicted parameters being outside the set range, in a preferred embodiment of the present invention, the ultrasonic vibration parameter tuning unit is tuned only if the pressure-lowering parameter tuning unit has already obtained the target pressure-lowering parameter. That is, the parameters for the subsequent process stage are tuned only after ensuring the accuracy of the previous process stage. Similarly, the pressure-holding parameter tuning unit is tuned only if the target ultrasonic vibration signal has been obtained, and the post-weld processing parameter tuning unit is tuned only if the target pressure-holding parameter has been obtained.

[0110] Because the input process parameters to be optimized are diverse, in order to further improve the optimization efficiency of process parameters, in some embodiments of the present invention, such as... Figure 1 As shown, the hierarchical analysis model also includes a process parameter input layer, which includes a pressure holding parameter input unit, an ultrasonic vibration parameter input unit, a pressure holding parameter input unit, and a post-weld treatment parameter input unit.

[0111] The pressure holding parameter input unit is used to input the pressure-down parameters to be optimized into the pressure-down process parameter prediction unit;

[0112] The ultrasonic vibration parameter input unit is used to input the ultrasonic vibration parameters to be optimized into the ultrasonic vibration process parameter prediction unit.

[0113] The pressure holding parameter input unit is used to input the pressure holding parameters to be optimized into the pressure holding process parameter prediction unit;

[0114] The post-weld processing parameter input unit is used to input the post-weld processing parameters to be optimized into the post-weld processing parameter prediction unit.

[0115] In this embodiment of the invention, a parameter input unit is set up for each group of process parameters to be optimized, so as to realize a one-to-one correspondence between the input process parameters and the process parameter prediction unit. This avoids the problem of the process parameter prediction unit selecting or inputting incorrect parameters, thereby improving the efficiency of process parameter optimization and ensuring the accuracy of process parameter optimization.

[0116] To further improve the accuracy of process prediction parameters and provide optimization direction for process parameter optimization, in some embodiments of the present invention, such as... Figure 1 As shown, the hierarchical analysis model also includes a first multi-head self-attention (MHSA) layer set between the process parameter input layer and the process parameter prediction layer;

[0117] The first multi-head self-attention layer is used to perform self-attention learning on multiple sets of process parameters to be optimized, and obtain the attention process parameter features;

[0118] The process parameter prediction layer is used to predict process parameters based on attention process parameter features, and obtain multiple process prediction parameters.

[0119] This invention, through the establishment of a first multi-head self-attention layer, can evaluate the contribution rate of different historical inputs to the output based on multiple sets of input parameters to be optimized, thereby adding more weight to key feature information and improving the accuracy of process prediction parameters. Simultaneously, since the first multi-head attention layer assigns different weights to different inputs, these weights represent the degree of influence of each input process parameter on the process prediction parameters. Therefore, it can provide direction for subsequent optimization of process parameters, thereby improving the optimization efficiency and accuracy of process parameters.

[0120] The process of the multi-head attention layer is as follows: the query vector, key vector and value vector are obtained based on the input by mapping the parameter matrices of the three science departments. Then, the similarity between features is measured by the dot product of the query vector and the key vector. Then, the attention scores are weighted by the softmax function. Finally, the outputs of all heads are concatenated and transformed by the learnable output weight matrix to obtain the final feature representation.

[0121] Furthermore, to improve the accuracy of welding quality prediction, in some embodiments of the present invention, such as... Figure 2 As shown, the hierarchical analysis model also includes a second multi-head self-attention layer set between the process parameter prediction layer and the welding quality prediction layer;

[0122] The second multi-head self-attention layer is used to perform self-attention learning on multiple process prediction parameters to obtain the attention process parameter features;

[0123] The welding quality prediction layer is used to predict the quality of welding processes by analyzing the parameters of the welding process.

[0124] By setting a second multi-head self-attention layer, this invention can assign different weights to different process prediction parameters based on the input parameters, thereby increasing the weight of key process prediction parameters and improving the accuracy of the final predicted welding quality.

[0125] To verify the effectiveness of the process parameter optimization model proposed in the embodiments of the present invention, in some embodiments of the present invention, 10 sets of test sets were used to predict welding quality. The error between the prediction results and the experimental results is shown in Table 1:

[0126] Table 1. Predicted and experimental results of tensile strength and contact resistance.

[0127]

[0128] As shown in Table 1, the average relative percentage errors of tensile strength and contact resistance obtained by the process parameter optimization model proposed in this embodiment are 3.21% and 3.7%, respectively. This indicates that the model's prediction of the test set samples is very close to the actual measured values, and the model has high predictive and generalization capabilities. Furthermore, although the prediction error for contact resistance is also very small, it is still higher than the error for tensile strength. This result is mainly due to the measurement error of contact resistance.

[0129] Furthermore, to verify whether introducing the multi-head attention mechanism into the process parameter optimization model effectively improves the prediction ability of welding quality, and to verify the efficiency of the proposed model, the same test set samples were used to compare the quality predictions of three trained models: MHSA-LSTM-AHP (including the first and second multi-head self-attention layers), MHSA-AHP (excluding the first and second multi-head self-attention layers), and GA-BP. The comparison metrics used were mean square error (MSE), RMSE, and mean absolute percentage error (MAPE). The comparison results are shown in Table 2.

[0130] Table 2 Performance comparison of different models

[0131]

[0132] Table 2 shows that, in predicting tensile strength, the MAE, RMSE, and MAPE of the MHSA-LSTM-AHP model are 15.1, 16.16, and 0.032, respectively. These represent significant reductions compared to the LSTM-AHP and GA-BP algorithms. Similarly, the error in resistance prediction is also significantly reduced, demonstrating that the MHSA-LSTM-AHP model has better predictive performance and generalization ability. Compared to the performance metrics of MHSA-LSTM, the MAE, RMSE, and MAPE of MHSA-LSTM-AHP are reduced by 17%, 13.4%, and 16.3%, respectively; and even more significantly by 24%, 21.5%, and 24.7% in resistance prediction. This indicates that the self-attention mechanism introduced in LSTM-AHP significantly improves the model's predictive ability. GA-BP, LSTM-AHP, and MHSA-LSTM-AHP all showed increased errors in resistance prediction compared to tensile strength prediction, but GA-BP showed a greater increase in error. This indicates that LSTM-AHP and MHSA-LSTM-AHP, which can learn time-dependent factors, can extract more information and exhibit better robustness, while the traditional GA-BP performs somewhat poorly.

[0133] In summary, compared with the traditional GA-BP and LSTM-AHP models, MHSA-LSTM-AHP has better predictive ability and better robustness. It outperforms the other two models in all indicators, and its error in predicting resistance is much smaller than that of the GA-BP and LSTM-AHP models.

[0134] like Figure 3As shown, the present invention also provides a process parameter optimization device 300 for ultrasonic metal welding. This ultrasonic metal welding process parameter optimization device 300 includes a process parameter acquisition unit 310 and a process parameter optimization module 320;

[0135] The process parameter acquisition device 310 is used to acquire multiple sets of process parameters to be optimized in multiple process stages of ultrasonic metal welding process.

[0136] The process parameter optimization module 320 includes a processor 321, a memory 322, and a display 323. Figure 3 Only some components of the ultrasonic metal welding process parameter optimization equipment 300 are shown. However, it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.

[0137] In some embodiments, processor 321 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in memory 322 or process data, such as the ultrasonic metal welding process parameter optimization method of the present invention.

[0138] In some embodiments, memory 322 may be an internal storage unit of the ultrasonic metal welding process parameter optimization device 300, such as a hard disk or memory of the ultrasonic metal welding process parameter optimization device 300. In other embodiments, memory 322 may also be an external storage device of the ultrasonic metal welding process parameter optimization device 300, such as a pluggable hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the ultrasonic metal welding process parameter optimization device 300.

[0139] Furthermore, the memory 322 may include both internal storage units of the ultrasonic metal welding process parameter optimization device 300 and external storage devices. The memory 322 is used to store application software and various types of data for installing the ultrasonic metal welding process parameter optimization device 300.

[0140] In some embodiments, display 323 may be an LED display, a liquid crystal display, a touch-screen liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 323 is used to display information from the ultrasonic metal welding process parameter optimization equipment 300 and to display a visual user interface. Components 321-323 of the ultrasonic metal welding process parameter optimization equipment 300 communicate with each other via a system bus.

[0141] In some embodiments of the present invention, when the processor 321 executes the ultrasonic metal welding process parameter optimization program in the memory 322, the following steps can be implemented:

[0142] Receive multiple sets of process parameters to be optimized in multiple stages of ultrasonic metal welding process;

[0143] Multiple sets of process parameters to be optimized are input into the process parameter prediction layer to predict process parameters, thereby obtaining multiple predicted process parameters that correspond one-to-one with multiple process stages.

[0144] When multiple prediction process parameters exceed the set parameter range, the parameter tuning module is used to tune multiple sets of process parameters to be optimized to obtain multiple sets of optimized process parameters; the multiple prediction control parameters corresponding to the multiple sets of optimized process parameters are within the set parameter range.

[0145] When multiple prediction process parameters are within the set parameter range, the multiple prediction process parameters are input into the welding quality prediction layer to perform quality prediction and obtain the predicted welding quality.

[0146] It should be understood that when the processor 321 executes the ultrasonic metal welding process parameter optimization program in the memory 322, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.

[0147] The ultrasonic metal welding process parameter optimization device 300 in this embodiment of the invention can be installed in passenger cars and commercial vehicles.

[0148] Accordingly, embodiments of the present invention also provide a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions in the ultrasonic metal welding process parameter optimization methods provided in the above-described method embodiments.

[0149] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0150] The above provides a detailed description of the method and equipment for optimizing process parameters of ultrasonic metal welding provided by the present invention. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method of optimizing process parameters for ultrasonic metal welding, characterized in that, The process parameters of ultrasonic metal welding are optimized based on a process parameter optimization model, the process parameter optimization model comprising an analytic hierarchy process model and a parameter tuning module, the analytic hierarchy process model comprising a process parameter prediction layer and a welding quality prediction layer connected in sequence, and the input and output of the parameter tuning module being connected to the output and input of the process parameter prediction layer, respectively. The method comprises: receiving a plurality of groups of process parameters to be optimized in a plurality of process stages of ultrasonic metal welding; inputting the plurality of groups of process parameters to be optimized into the process parameter prediction layer for process parameter prediction, to obtain a plurality of predicted process parameters corresponding to the plurality of process stages one by one; when the plurality of predicted process parameters are outside a set parameter range, tuning the plurality of groups of process parameters to be optimized based on the parameter tuning module to obtain a plurality of groups of optimized process parameters, the plurality of predicted control parameters corresponding to the plurality of groups of optimized process parameters being within the set parameter range; when the plurality of predicted process parameters are within the set parameter range, inputting the plurality of predicted process parameters into the welding quality prediction layer for quality prediction, to obtain a predicted welding quality; the analytic hierarchy process model further comprises a process parameter input layer for inputting the plurality of groups of process parameters to be optimized into the process parameter prediction layer, and a first multi-head self-attention layer arranged between the process parameter input layer and the process parameter prediction layer; the first multi-head self-attention layer is used for self-attention learning on the plurality of groups of process parameters to be optimized, to obtain attention process parameter features; the process parameter prediction layer is used for process parameter prediction on the attention process parameter features, to obtain the plurality of predicted process parameters.

2. The method of process parameter optimization for ultrasonic metal welding of claim 1, wherein, When the plurality of process stages comprise a pressing stage, an ultrasonic vibration stage, a holding stage and a post-welding treatment stage, the plurality of groups of process parameters to be optimized comprise process parameters to be optimized in the pressing stage, process parameters to be optimized in the ultrasonic vibration stage, process parameters to be optimized in the holding stage and process parameters to be optimized in the post-welding treatment stage, and the plurality of predicted process parameters comprise predicted process parameters in the pressing stage, predicted process parameters in the ultrasonic vibration stage, predicted process parameters in the holding stage and predicted process parameters in the post-welding treatment stage.

3. The method of claim 2, wherein the process parameters are optimized for ultrasonic metal welding. The process parameter prediction layer comprises a pressing process parameter prediction unit, an ultrasonic vibration process parameter prediction unit, a holding process parameter prediction unit and a post-welding treatment process parameter prediction unit; the pressing process parameter prediction unit is used for predicting process parameters in the pressing stage based on the process parameters to be optimized in the pressing stage, to obtain the predicted process parameters in the pressing stage; the ultrasonic vibration process parameter prediction unit is used for predicting process parameters in the ultrasonic vibration stage based on the process parameters to be optimized in the ultrasonic vibration stage and the predicted process parameters in the pressing stage, to obtain the predicted process parameters in the ultrasonic vibration stage; the holding process parameter prediction unit is used for predicting process parameters in the holding stage based on the process parameters to be optimized in the holding stage and the predicted process parameters in the ultrasonic vibration stage, to obtain the predicted process parameters in the holding stage; the post-welding treatment process parameter prediction unit is used for predicting process parameters in the post-welding treatment stage based on the process parameters to be optimized in the post-welding treatment stage and the predicted process parameters in the holding stage, to obtain the predicted process parameters in the post-welding treatment stage.

4. The method of claim 2, wherein the method further comprises: The welding quality prediction layer comprises a pressing quality prediction unit, an ultrasonic vibration quality prediction unit, a holding quality prediction unit and a post-welding treatment quality prediction unit; The pressing quality prediction unit is configured to predict the quality of the pressing stage based on the pressing process prediction parameter, and obtain pressing quality implicit information; The ultrasonic vibration quality prediction unit is configured to predict the quality of the ultrasonic vibration stage based on the ultrasonic vibration process prediction parameter and the pressing quality implicit information, and obtain ultrasonic vibration quality implicit information; The holding quality prediction unit is configured to predict the quality of the holding stage based on the holding process prediction parameter and the ultrasonic vibration quality implicit information, and obtain holding quality implicit information; The post-welding treatment quality prediction unit is configured to predict the quality of the post-welding treatment stage based on the post-welding treatment process prediction parameter and the holding quality implicit information, and obtain the predicted welding quality.

5. The method of claim 3, wherein the method further comprises: The parameter optimization module comprises a pressing parameter optimization unit, an ultrasonic vibration parameter optimization unit, a holding parameter optimization unit and a post-welding treatment parameter optimization unit; The pressing parameter optimization unit is configured to optimize the to-be-optimized pressing parameter when the pressing process prediction parameter exceeds the set pressing parameter range, and obtain a target pressing parameter; the pressing process prediction parameter corresponding to the target pressing parameter is within the set pressing parameter range; The ultrasonic vibration parameter optimization unit is configured to optimize the to-be-optimized ultrasonic vibration parameter when the ultrasonic vibration process prediction parameter exceeds the set ultrasonic vibration parameter range, and obtain a target ultrasonic vibration parameter; the target ultrasonic vibration parameter is within the set ultrasonic vibration parameter range; The holding parameter optimization unit is configured to optimize the to-be-optimized holding parameter when the holding process prediction parameter exceeds the set holding parameter range, and obtain a target holding parameter; the target holding parameter is within the set holding parameter range; The post-welding treatment parameter optimization unit is configured to optimize the to-be-optimized post-welding treatment parameter when the post-welding treatment process prediction parameter exceeds the set post-welding treatment parameter range, and obtain a target post-welding treatment parameter; the target post-welding treatment parameter is within the set post-welding treatment parameter range.

6. The method of claim 3, wherein the method further comprises: The process parameter input layer comprises a holding parameter input unit, an ultrasonic vibration parameter input unit, a holding parameter input unit and a post-welding treatment parameter input unit; The holding parameter input unit is configured to input the to-be-optimized pressing parameter to the pressing process parameter prediction unit; The ultrasonic vibration parameter input unit is configured to input the to-be-optimized ultrasonic vibration parameter to the ultrasonic vibration process parameter prediction unit; The holding parameter input unit is configured to input the to-be-optimized holding parameter to the holding process parameter prediction unit; The post-welding treatment parameter input unit is configured to input the to-be-optimized post-welding treatment parameter to the post-welding treatment process parameter prediction unit.

7. The method of optimizing process parameters for ultrasonic metal welding according to any one of claims 1-5, wherein, The analytic hierarchy process model further comprises a second multi-head self-attention layer arranged between the process parameter prediction layer and the welding quality prediction layer; The second multi-head self-attention layer is configured to perform self-attention learning on the plurality of process prediction parameters to obtain attention process parameter features. The welding quality prediction layer is configured to perform quality prediction on the attention process parameters to obtain the predicted welding quality.

8. The method of claim 2, wherein, The to-be-optimized down-pressing parameters include a down-pressing pressure, the to-be-optimized ultrasonic vibration parameters include a first amplitude of an amplitude transformer, a first ultrasonic vibration power and a welding time, the to-be-optimized pressure maintaining parameters include a pressure maintaining time, and the to-be-optimized post-welding treatment parameters include a second amplitude of the amplitude transformer, a second ultrasonic vibration frequency and a post-welding treatment time. The down-pressing process prediction parameters include a peak value of a welding head depth in a down-pressing stage and an average value of a welding part temperature, the ultrasonic vibration process prediction parameters include a peak value of a welding head depth, an average value of a welding head depth, a standard deviation of a welding head depth, a peak value of a welding part temperature, an average value of a welding part temperature, a standard deviation of a welding part temperature, a peak value of a welding head amplitude, an average value of a welding head amplitude and a standard deviation of a welding head amplitude in an ultrasonic vibration stage, the pressure maintaining process prediction parameters include a peak value of a welding head depth and an average value of a welding part temperature in a pressure maintaining stage, and the post-welding treatment process prediction parameters include an average value of a welding part temperature, a peak value of a welding head amplitude, an average value of a welding head amplitude and a standard deviation of a welding head amplitude in a post-welding treatment stage. The predicted welding quality includes a predicted joint tensile strength and a predicted contact resistance.

9. An apparatus for optimizing process parameters of ultrasonic metal welding, characterized in that, The process parameter collector is configured to collect a plurality of groups of to-be-optimized process parameters in a plurality of process stages in an ultrasonic metal welding process. The process parameter optimization module includes a memory and a processor, wherein The memory is configured to store a program. The processor is coupled to the memory and is configured to execute the program stored in the memory to implement the steps in the process parameter optimization method of the ultrasonic metal welding in any one of claims 1 to 8. ​

Images