A method, apparatus, and medium for determining welding parameters

Abstract

The application discloses a welding parameter determination method and device and a medium thereof, relates to the technical field of welding, and is used for providing guidance for an actual welding process. In view of the problems of high requirement for technicians, low efficiency and inability to guarantee quality of the current welding parameter determination mode, the welding parameter determination method is provided. The welding parameter determination method is used for training a prediction model by taking historical data of welding material information, welding parameters and corresponding quality parameters as a data set in advance, so that corresponding welding parameters can be inversely deduced by inputting welding material information and requirements for output quality parameters. The welding quality expected by the welding parameters obtained is guaranteed. Compared with the method of repeatedly trying and debugging according to the experience of technicians, the welding parameter determination method is more efficient, does not cause the output of a large number of unqualified products due to trial production, avoids unnecessary losses, and better meets the production needs of actual welding processing technology.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of welding, in particular to a welding parameter determination method and device and a medium thereof. BACKGROUND

[0002] Welding technology is a key technology in various production and manufacturing processes, and the quality of finished products is largely determined by the welding quality in the production process. The quality of welding is closely related to the welding parameters set by the welding equipment, and when the material quality, material thickness, electrode size, etc. of the welding material change, the welding parameters need to be adjusted to obtain better welding quality.

[0003] Currently, before the welding process, technicians adjust the welding parameters based on their personal experience, but this parameter adjustment method has a low success rate and needs to be repeatedly tried to adjust to a relatively suitable parameter, which is a huge workload and the effect cannot be guaranteed. After adjusting the parameters, trial production needs to be continuously carried out for inspection, which will cause a large number of unqualified products, and if the welding material is frequently replaced, it will seriously affect the production.

[0004] Therefore, the technical personnel in the field urgently need a welding parameter determination method to solve the problem that the current welding parameter adjustment relying on manual experience requires high technical personnel, has low efficiency, and cannot guarantee the welding quality. SUMMARY

[0005] The purpose of the present application is to provide a welding parameter determination method, device and medium to solve the problem that the current welding parameter adjustment relying on manual experience requires high technical personnel, has low efficiency, and cannot guarantee the welding quality.

[0006] To solve the above technical problems, the present application provides a welding parameter determination method, comprising:

[0007] obtaining welding material information;

[0008] calling a prediction model; wherein the prediction model is a supervised multi-output regression model trained based on a deep learning framework by taking historical data of welding material information, welding parameters and quality parameters as a data set;

[0009] inputting the welding material information into the prediction model to obtain welding parameters meeting the preset quality standard.

[0010] Preferably, the welding material information includes thickness information and material quality information;

[0011] Correspondingly, after obtaining the welding material information, it further comprises:

[0012] regularizing the material quality information.

[0013] Preferably, the quality parameters include welding process stability, quality score and spatter rate.

[0014] Correspondingly, the preset quality standards include a first threshold corresponding to the welding process stability, a second threshold corresponding to the quality score and a third threshold corresponding to the spatter rate.

[0015] Preferably, inputting the welding material information into the prediction model to obtain the welding parameters satisfying the preset quality standards comprises:

[0016] Inputting the welding material information as the characteristic parameters into the prediction model to obtain the welding parameters whose welding process stability is higher than the first threshold, whose quality score is higher than the second threshold and whose spatter rate is lower than the third threshold output by the prediction model.

[0017] Preferably, the method further comprises:

[0018] Periodically acquiring welding material data, welding parameter data and corresponding quality parameter data in an actual welding process, updating the data set to optimize the prediction model.

[0019] Preferably, after obtaining the welding parameters satisfying the preset quality standards, the method further comprises:

[0020] Acquiring workstation information and sending the welding parameters to a corresponding workstation display screen according to the workstation information.

[0021] Preferably, the method further comprises:

[0022] If the welding parameters satisfying the preset quality standards are not obtained, returning preset default welding parameters and performing an alarm.

[0023] To solve the above technical problems, the application further provides a welding parameter determination device, comprising:

[0024] An information acquisition module for acquiring welding material information;

[0025] A model calling module for calling a prediction model; wherein the prediction model is a supervised multi-output regression model trained based on a deep learning framework and with historical data of welding material information, welding parameters and quality parameters as a data set;

[0026] A parameter prediction module for inputting the welding material information into the prediction model to obtain welding parameters satisfying preset quality standards.

[0027] Preferably, the welding parameter determination device further comprises:

[0028] A data processing module for performing regularization processing on the material information.

[0029] The model optimization module is configured to periodically acquire welding material data, welding parameter data and corresponding quality parameter data in an actual welding process, update a data set, and optimize the prediction model.

[0030] The parameter display module is configured to acquire workstation information and send welding parameters to a corresponding workstation display screen according to the workstation information.

[0031] The abnormality alarm module is configured to return preset default welding parameters and perform an alarm if welding parameters meeting preset quality standards are not acquired.

[0032] To solve the above technical problems, the application further provides a welding parameter determination device, which comprises:

[0033] The memory is configured to store a computer program.

[0034] The processor is configured to implement the steps of the welding parameter determination method when executing the computer program.

[0035] To solve the above technical problems, the application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the welding parameter determination method.

[0036] The welding parameter determination method provided by the application uses historical data of welding material information, welding parameters and corresponding quality parameters as a data set in advance, trains a supervised multi-output regression model based on a deep learning framework, and thus can predict the quality parameters corresponding to the input welding material information and welding parameters in a subsequent welding process through a prediction model. Therefore, welding parameters can be acquired by taking a preset quality standard as an output target of the prediction model and taking current welding material information as a characteristic parameter. When the performance of the trained prediction model is sufficient, the acquired welding parameters can meet the quality acceptance requirements of an actual welding process. Compared with the current method of repeatedly trying and debugging based on the experience of technical personnel, the method is more efficient and the welding quality is more guaranteed. In addition, the method does not cause the production of a large number of unqualified products due to trial production, avoids unnecessary losses, and better meets the production requirements of an actual welding process.

[0037] The welding parameter determination device and the computer readable storage medium provided by the application correspond to the above method and have the same effects. BRIEF DESCRIPTION OF DRAWINGS

[0038] To more clearly illustrate the embodiments of the application, the drawings required in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative effort.

[0039] Figure 1 A flow chart of a welding parameter determination method provided by the present application is shown in the figure;

[0040] Figure 2 A flow chart of another welding parameter determination method provided by the present application is shown in the figure;

[0041] Figure 3 A structure diagram of a welding parameter determination device provided by the present application is shown in the figure;

[0042] Figure 4 A structure diagram of another welding parameter determination device provided by the present application is shown in the figure. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0044] The core of the present application is to provide a welding parameter determination method, device and medium.

[0045] In order to enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0046] At present, the determination of welding parameters is mainly based on the experience of workers. After trial production, the welding parameters are repeatedly adjusted according to the advantages and disadvantages of the welding quality of the produced products, until the welding quality of the produced products meets the needs, and then the welding parameter determination process ends. This method not only requires high technical personnel, but also wastes a lot of manpower, material resources and time in the process of welding parameter adjustment and trial production. Low efficiency will also bring unnecessary losses, especially for such as a vehicle body-in-white, which has thousands of welding points. The welding materials and welding parameters required for different welding points are not necessarily the same, so it is more difficult to determine all the welding parameters of the entire body-in-white.

[0047] To solve the above problems, the present application provides a welding parameter determination method, as shown in Figure 1 The method comprises the following steps:

[0048] S11: Obtain welding material information.

[0049] S12: Call a prediction model.

[0050] The prediction model is a supervised multi-output regression model trained based on a deep learning framework, with historical data of welding material information, welding parameters and quality parameters as a data set.

[0051] S13: inputting the welding material information into the prediction model to obtain welding parameters meeting preset quality standards.

[0052] In actual welding processes, different products have different welding processes, and different welding materials and welding parameters are required.

[0053] Taking the production of a white body as an example, the welding material information includes but is not limited to the number of layers of welding plates, thickness, material quality, and whether to coat glue and other information. Considering the influence of different welding material information on welding quality and the convenience of subsequent prediction model training and calculation, the preferred welding material information is thickness information and material quality information. The welding parameters include but are not limited to welding current, electrode pressure, welding time, and grinding frequency. Similarly, the quality parameters for evaluating welding quality include but are not limited to welding process stability, quality score, spatter rate, semi-destroyed qualified rate, fully-destroyed qualified rate, and eddy current measurement qualified rate.

[0054] It should be noted that the above welding material information and welding parameters can be obtained from devices such as group control equipment in the production workshop that control and supervise the production process. For example, the group control equipment collects actual welding data including welding material information and welding parameters.

[0055] It should also be noted that for the acquisition of quality parameters, since welding quality is determined relatively according to actual conditions, before the prediction model is established, manual inspection is performed through sampling, breaking, etc. to input original quality parameter data. The quality parameters can be judged from whether the welding process is stable, the welding quality quantitative score, and the probability of spatter occurrence in the welding process, thereby obtaining the original quality parameter data.

[0056] When a sufficient amount of the above data is obtained, the prediction model can be trained through machine learning. It is easy to understand that the welding material information + welding parameters correspond to the welding quality parameters, i.e. the data set is a data set with input and output. The corresponding machine learning method is supervised learning, with welding material information and welding parameters as input and quality parameters as output.

[0057] Since the welding material information, welding parameters, and quality parameters as output each contain multiple sub-information, the prediction model is a multiple-input and multiple-output model, which can be implemented by various machine learning methods, such as multiple multiple-input and single-output models, MultiOutputRegressor, and the preferred automatic learning framework (autosklearn, autokeras, etc.) in the above embodiment. The skilled person can select a suitable method to establish the prediction model according to the actual manufacturing process and welding process, and the mean absolute error (MAE) can be used as an evaluation index to judge the prediction effect. For example, when the application scenario is the vehicle body-in-white manufacturing process, the multiple-layer neural network model established by the automatic learning framework autokeras can be used as the prediction model.

[0058] In addition, since various welding materials are involved in the welding process, for example, more than 100 different materials are involved in the vehicle body-in-white manufacturing process, and using the material name as the material information is not standardized and cannot determine the subsequent welding parameters, the embodiment provides a preferred implementation: the material information is subjected to a regularization process.

[0059] Specifically, the regularization process can be achieved by characterizing material-specific parameters, such as yield strength, tensile strength, hardness, elongation at break, and other material-related parameters of different materials, which are subjected to different regularization processes according to the needs of different welding processes.

[0060] It should be noted that even the same characteristic parameter of the same material is not completely consistent, but varies within a certain range, which also brings some trouble to the subsequent model training and prediction, so in the regularization process of the material information, a more representative characteristic value can be used as the specific value of a characteristic parameter of a material to simplify the subsequent process and improve the prediction accuracy.

[0061] The welding parameter determination method provided in the present application can use the pre-trained prediction model to determine the welding parameters that can meet the actual engineering requirements of the predicted welding quality by inputting the welding material information as the characteristic parameters, and the welding parameters are recommended as the preferred welding parameters to each welding station, which provides guidance for the actual welding process, helps to improve the welding quality, and does not rely on manual repeated debugging by technical personnel, has low requirements for technical personnel, is more efficient, and does not need repeated trial production, avoids unnecessary losses, and better meets the requirements of actual welding production.

[0062] From the above embodiments, it can be seen that the welding quality parameters of different welding processes are not necessarily the same, but based on the commonality and general requirements, the present embodiment provides a preferred embodiment, and the above-mentioned quality parameters include: welding process stability, quality score and spatter rate.

[0063] The welding process stability is an evaluation of the stability of the entire welding process, usually expressed in percentage, and the higher the value, the better the stability of the welding; the quality score is determined according to the actual welding needs, and the quality score of the same welding finished product may be different in different application scenarios, and is usually represented by a score, and the higher the score, the better the welding quality. In one possible implementation, the full score of the quality score is 120; the spatter rate is the probability of metal spatter occurring during the welding process, which is also expressed in percentage, and is usually lower the better.

[0064] Corresponding to the preferred scheme of the above-mentioned quality parameters, the preset quality standard can be preferably three preset thresholds, respectively corresponding to the welding process stability, the quality score and the spatter rate. Since the welding process stability and the quality score are higher the better, so greater than the corresponding threshold is to meet the preset quality standard, and the spatter rate is smaller the better, so less than the corresponding threshold is to meet the preset quality requirement.

[0065] More specifically, the present embodiment also provides an instantiated application scenario example. For the welding manufacturing process of the vehicle body-in-white, the first threshold corresponding to the welding process stability is 95%, that is, stability exceeding 95% is considered to meet the standard; the second threshold corresponding to the quality score is 103, that is, the quality score exceeding 103 is considered to meet the standard; the third threshold corresponding to the spatter rate is 5%, that is, the spatter rate below 5% is considered to meet the standard. Further, when the above three quality parameters all meet the standard, it is considered that the corresponding welding parameters meet the preset quality standard, and can be used as recommended welding parameters to guide the actual welding process.

[0066] On the basis of the present embodiment, the welding parameters meeting the preset quality standard are obtained by the prediction model, that is, the current welding material information is input into the prediction model as a feature parameter, the process stability greater than the first threshold, the quality score greater than the second threshold, and the spatter rate lower than the third threshold are taken as target parameters, the welding parameters are calculated, and the obtained welding parameters are the preferred welding parameters for guiding the actual welding production.

[0067] It should be further pointed out that, as can be seen from the above, the prediction accuracy of the prediction model is the most direct influencing factor of the welding quality of the welding parameters determined by the welding parameter determination method provided by the present application, so for improving the accuracy of the prediction model, the present embodiment also provides a preferred scheme:

[0068] Periodically acquire welding material data, welding parameter data and corresponding quality parameter data in actual welding process, update the data set to optimize the prediction model.

[0069] In addition to the above preferred solution of increasing the number of data set samples by continuously collecting new actual welding data to improve the prediction accuracy of the prediction model, the multi-output regression prediction model based on the deep learning framework also has self-learning ability, and can continuously learn and optimize the mechanism model through the prediction of its own welding parameters, thereby further improving the prediction accuracy.

[0070] The preferred solution provided by the embodiment gives a relatively universal combination of quality parameters, and presets a corresponding threshold value for each quality parameter. Whether the preset quality standard is met is determined by comparing the size of the corresponding threshold value, so as to determine the welding parameters with expected good quality according to the prediction model, thereby providing guidance for the actual welding process. In addition, to ensure the guidance effect of the above method and improve the welding quality, the embodiment also updates the data set by continuously collecting new actual welding data, and continuously optimizes the mechanism model in combination with the self-learning ability of the regression model, so as to improve the accuracy of the prediction, and thus the welding quality of the preferred welding parameters of the prediction model is also more guaranteed.

[0071] It is easy to know that in the actual production environment, the production workshop usually carries out welding manufacturing in units of stations, and the welding processes of different stations may be different, so for a station, it is only interested in the welding parameters required for the welding process being carried out in the current station. Therefore, in order to make the above method have a targeted guidance effect, the embodiment provides a preferred implementation solution, after obtaining the welding parameters that meet the preset quality standard, as shown in Figure 2 The above method further comprises:

[0072] S14: Obtain station information, and send the welding parameters to the corresponding station display screen according to the station information.

[0073] In the production workshop, a station is usually configured with a plurality of group control devices, and these group control devices are used to control various welding devices for realizing welding. Therefore, from these group control devices, not only welding actual data can be acquired, but also corresponding station information can be obtained to distinguish to which station the group control device belongs, and similarly, the welding data corresponding to which station can also be distinguished. In addition, a station display screen is usually provided at the station in the production workshop to provide some necessary information for workers to view, so the station corresponding to the welding parameters can be determined through the station information, and then the welding parameters are sent to the station display screen at the corresponding station to guide the technical personnel at the corresponding station to carry out welding operation.

[0074] It should be noted that in the welding manufacturing process such as the vehicle body-in-white, the welding operation is carried out in a basic unit of a welding spot, and the welding spot is uniquely represented by a welding spot identity document (ID), so the correspondence of various welding parameters and other data can be simply realized through the welding spot ID.

[0075] A preferred scheme provided by the embodiment is to use the station display screen provided by the existing workshop at each station to respectively send and display welding data corresponding to different stations, so as to provide targeted guidance for the welding operation of each station, which helps to improve the welding efficiency and welding quality.

[0076] Since the working conditions in the actual manufacturing process are constantly changing, the welding process may change, resulting in the addition of some welding spots. At this time, the sample historical data of the added welding spots is insufficient, the welding parameters predicted by the prediction model are inaccurate or cannot be predicted to meet the preset quality standard. In order to ensure the guidance of the method, if Figure 2 As shown in the embodiment, a preferred embodiment is provided:

[0077] S15: If the welding parameters meeting the preset quality standard are not obtained, return the preset default welding parameters and perform an alarm.

[0078] At this time, if the preferred scheme of step S14 is established, the welding parameters sent to the corresponding station display screen for display in step S14 are the default welding parameters determined in step S15.

[0079] However, in fact, step S14 and step S15 can also be implemented independently, and the above is only one possible embodiment. When step S14 and step S15 are implemented independently and do not affect each other, there is no sequence relationship between them.

[0080] When the welding parameters meeting the quality standard can be obtained, it is highly probable that the sample data in the prediction model data set is not sufficient enough, and the prediction ability for this new working condition needs to be improved. At this time, the default welding parameters pre-recorded can be returned. The default welding parameters can be realized by a group of basic welding parameters summarized in the actual welding process, or can be realized by other ways. However, since the welding parameters obtained in the above manner have a guarantee in welding quality, an alarm needs to be performed to prompt the technical personnel to perform differential treatment on the welding spot. In addition, the alarm information is also helpful to prompt the relevant personnel to collect the actual welding data of the welding spot, enrich the data set of the prediction model, and optimize the prediction model mechanism and improve the effective prediction range.

[0081] In the above embodiments, a welding parameter determination method is described in detail, and the application also provides a corresponding welding parameter determination device. It should be noted that the embodiments of the device part are described from two angles, one is based on the functional module, and the other is based on the hardware.

[0082] Based on the functional module, as shown in Figure 3 The embodiment provides a welding parameter determination device, which comprises:

[0083] The information acquisition module 21 is configured to acquire welding material information.

[0084] The model calling module 22 is configured to call a prediction model, wherein the prediction model is a supervised multi-output regression model trained based on a deep learning framework and by taking historical data of the welding material information, welding parameters and quality parameters as a data set.

[0085] The parameter prediction module 23 is configured to input the welding material information into the prediction model to obtain welding parameters meeting the preset quality standard.

[0086] Preferably, the welding parameter determination device further comprises:

[0087] The data processing module is configured to perform regularization processing on the material information.

[0088] The model optimization module is configured to periodically acquire welding material data, welding parameter data and corresponding quality parameter data in an actual welding process, update the data set, and optimize the prediction model.

[0089] The parameter display module is configured to acquire workstation information and send the welding parameters to a corresponding workstation display screen according to the workstation information.

[0090] The abnormality alarm module is configured to return preset default welding parameters and perform alarm if the welding parameters meeting the preset quality standard are not acquired.

[0091] Since the embodiments of the device part correspond to the embodiments of the method part, the embodiments of the device part are described in the description of the embodiments of the method part, and will not be described here.

[0092] The welding parameter determination device provided in the embodiment can obtain welding material information in a welding process through an information acquisition module; then, through a model calling module and a parameter prediction module, a pre-trained prediction model is used to back-calculate corresponding welding parameters according to output quality parameters meeting preset standards, so as to determine welding parameters meeting actual engineering requirements of expected welding quality, and recommend the welding parameters as preferred welding parameters to each welding station, thereby providing guidance for an actual welding process and helping to improve welding quality. Meanwhile, the device does not need to rely on manual repeated debugging by technical personnel, is more efficient, greatly reduces the requirements for technical personnel, does not need to produce a large number of unqualified products due to repeated trial production, avoids unnecessary losses, and better meets the requirements of actual welding production.

[0093] Figure 4 A structural diagram of a welding parameter determination device provided for another embodiment of the application is shown in FIG. 2, which includes a memory 30 for storing a computer program. Figure 4

[0094] A processor 31 is configured to implement steps of a welding parameter determination method according to any one of the above embodiments when executing the computer program.

[0095] The welding parameter determination device provided in the embodiment can include, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, or the like.

[0096] The processor 31 can include one or more processing cores, such as a 4-core processor, an 8-core processor, or the like. The processor 31 can be implemented in at least one of a hardware form of a digital signal processor (DSP), a field-programmable gate array (FPGA), a programmable logic array (PLA), or the like. The processor 31 can also include a main processor and a coprocessor. The main processor is a processor for processing data in an awake state, also known as a central processing unit (CPU). The coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 31 can be integrated with a graphics processor (GPU) that is responsible for rendering and drawing content to be displayed on a display screen. In some embodiments, the processor 31 can also include an artificial intelligence (AI) processor that is configured to process computing operations related to machine learning. ​

[0097] The memory 30 can include one or more computer-readable storage media. The computer-readable storage media can be non-transitory. The memory 30 can also include high-speed random access memory and nonvolatile, computer-readable storage media such as one or more magnetic disk storage devices, flash memory devices. In this embodiment, the memory 30 is at least used to store the following computer program 301, wherein the computer program is loaded and executed by the processor 31, and can realize the related steps of the welding parameter determination method disclosed in any of the foregoing embodiments. In addition, the resources stored by the memory 30 can also include an operating system 302 and data 303, etc., and the storage mode can be temporary storage or permanent storage. The operating system 302 can include Windows, Unix, Linux, etc. The data 303 can include but is not limited to a welding parameter determination method, etc.

[0098] In some embodiments, the welding parameter determination device can also include a display screen 32, an input / output interface 33, a communication interface 34, a power supply 35, and a communication bus 36.

[0099] Those skilled in the art can understand that the structure shown in the above embodiments does not constitute a limitation on the welding parameter determination device, and can include more or fewer components than those shown in the drawings. Figure 4 Those skilled in the art can understand that the structure shown in the above embodiments does not constitute a limitation on the welding parameter determination device, and can include more or fewer components than those shown in the drawings.

[0100] The welding parameter determination device provided by the embodiment of the present application includes a memory and a processor. When the processor executes the program stored in the memory, the following method can be realized: a welding parameter determination method.

[0101] The welding parameter determination device provided by the embodiment of the present application can input the obtained material information into the prediction model through the pre-trained prediction model by executing the computer program stored in the memory by the processor, so as to determine the welding parameters that can meet the preset standard according to the output, and recommend the welding parameters as the preferred welding parameters to each welding station, thereby providing guidance for the actual welding process, and being beneficial to improve the welding quality of the actual welding production. At the same time, the device does not need to rely on manual repeated debugging of technical personnel, and is more efficient while greatly reducing the requirements for technical personnel. Also, a large number of unqualified products are not needed to be produced due to repeated trial production, unnecessary losses are avoided, and the requirements of the actual welding production are better met.

[0102] Finally, the present application also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, and the computer program is executed by the processor to realize the steps recorded in the above method embodiment.

[0103] It can be understood that if the method in the above embodiment is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and performs all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0104] The computer readable storage medium provided in the embodiment can input the obtained material information into the prediction model through the pre-trained prediction model when the computer program stored therein is executed, so that the welding parameters whose expected quality meets the preset standard can be determined according to the output, and the welding parameters are recommended to each welding station as preferred welding parameters, thereby providing guidance for the actual welding process, and the welding quality of the actual welding production is improved. At the same time, the device does not need to rely on manual repeated debugging of technical personnel, and the efficiency is higher, and the requirement for technical personnel is also greatly reduced. Also, a large number of unqualified products are not needed to be produced due to repeated trial production, unnecessary losses are avoided, and the requirements of actual welding production are better met.

[0105] The welding parameter determination method, device and medium provided by the present application are described in detail above. The embodiments in the specification are described in a progressive manner, and each embodiment mainly describes the differences from other embodiments. The same or similar parts of each embodiment can be referred to. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be referred to the method part. It should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, the present application can be improved and modified in several ways, and these improvements and modifications also fall within the protection scope of the claims of the present application.

[0106] It also needs to be explained that in the present specification, the relational terms such as first and second and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

Claims

1. A method for determining welding parameters, characterized in that: include: Obtaining welding material information; wherein the welding material information includes material thickness information and material information; Regularization processing is performed on the material information; wherein the material information is a characteristic value of a characteristic parameter corresponding to the material; different regularization processing is selected according to the requirements of the material corresponding to different welding processes; Calling a prediction model; wherein the prediction model is a supervised multi-output regression model pre-trained based on a deep learning framework using historical data on welding material information, welding parameters, and quality parameters as a data set, and the prediction model is also a multi-layer neural network model established using an automatic learning framework; the quality parameters include welding process stability, quality score, and spatter rate; the preset quality standards include: a first threshold corresponding to welding process stability, a second threshold corresponding to the quality score, and a third threshold corresponding to the spatter rate; Inputting the welding material information into the prediction model to obtain welding parameters that meet preset quality standards; The step of inputting the welding material information into the prediction model to obtain welding parameters that meet preset quality standards includes: The welding material information is input into the prediction model as a characteristic parameter, and the welding parameters output by the prediction model are obtained, wherein the welding process stability is higher than the first threshold, the quality score is higher than the second threshold, and the spatter rate is lower than the third threshold. In the welding manufacturing process of the vehicle body-in-white, the first threshold is 95%, the second threshold is 103, and the third threshold is 5%.

2. The welding parameter determination method according to claim 1, characterized in that: Also includes: Periodically obtain welding material data, welding parameter data, and corresponding quality parameter data during the actual welding process, and update the data set to optimize the prediction model.

3. The method for determining welding parameters according to claim 1, wherein: After obtaining the welding parameters that meet the preset quality standards, the method further includes: The workstation information is obtained, and the welding parameters are sent to the corresponding workstation display screen according to the workstation information.

4. The method for determining welding parameters according to claim 1, wherein: Also includes: If the welding parameters that meet the preset quality standards are not obtained, the preset default welding parameters are returned and an alarm is issued.

5. A welding parameter determination device, characterized in that: include: An information acquisition module is used to acquire welding material information; wherein the welding material information includes material thickness information and material information; A data processing module is used to perform regularization processing on material information; wherein the material information is a characteristic value corresponding to a characteristic parameter of the material; different regularization processing is selected according to the needs of the material corresponding to different welding processes; A model calling module is used to call a prediction model; wherein the prediction model is a supervised multi-output regression model pre-trained based on a deep learning framework using historical data on welding material information, welding parameters, and quality parameters as a data set, and the prediction model is also a multi-layer neural network model established using an automatic learning framework; the quality parameters include welding process stability, quality score, and spatter rate; the preset quality standards include: a first threshold corresponding to welding process stability, a second threshold corresponding to the quality score, and a third threshold corresponding to the spatter rate; A parameter prediction module, configured to input the welding material information into a prediction model to obtain welding parameters that meet preset quality standards; The step of inputting the welding material information into the prediction model to obtain welding parameters that meet preset quality standards includes: The welding material information is input into the prediction model as a characteristic parameter, and the welding parameters output by the prediction model are obtained, wherein the welding process stability is higher than the first threshold, the quality score is higher than the second threshold, and the spatter rate is lower than the third threshold. In the welding manufacturing process of the vehicle body-in-white, the first threshold is 95%, the second threshold is 103, and the third threshold is 5%.

6. A welding parameter determination device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the welding parameter determination method according to any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the welding parameter determination method according to any one of claims 1 to 4 are implemented.

Images