Spot welding STC curve recommendation method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of data processing, in particular to a spot welding STC curve recommendation method and device, electronic equipment and a storage medium. The spot welding STC curve recommendation method comprises the steps that the real-time spot welding condition of current spot welding operation is obtained; inputting the real-time spot welding condition into a pre-constructed STC curve recommendation model to obtain an optimal STC curve; wherein the optimal STC curve describes an ideal output current value of the spot welding equipment in the spot welding operation process; and spot welding operation is conducted according to the optimal STC curve.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a spot welding STC curve recommendation method, a spot welding STC curve recommendation device, an electronic device and a computer readable storage medium. BACKGROUND

[0002] Spot welding technology can use the resistance heat effect to locally melt the metal plate, and form a welding point under the action of pressure. Due to its high efficiency and low cost, spot welding technology is widely used in the fields of automobiles, household appliances and aerospace. The process quality and efficiency of the spot welding technology are highly dependent on the dynamic optimization of welding parameters. Usually, the STC (Sheet Thickness Combination) technology is used to perform test piece welding and curve collection in the laboratory to quickly build a sheet combination parameter library, and then to dynamically optimize the welding parameters.

[0003] The traditional STC curve method has certain limitations, such as low data utilization, poor adaptability, dependence on manual experience, and insufficient recommendation accuracy. In addition, the method of using machine learning or data-driven optimization of welding parameters in related technologies still has defects: the general model is not optimized for the heat-force coupling characteristics of spot welding, which may result in insufficient prediction accuracy; the algorithm is disconnected from the domain knowledge, which may cause recommendation deviation; and the static modeling mode cannot update the model through real-time data iteration on the production line. SUMMARY

[0004] The present application aims to at least solve one of the technical problems existing in the prior art. To this end, the present application provides a spot welding STC curve recommendation method, which fully utilizes the basic language model combined with historical data to effectively improve the accuracy of the model output curve, and thus improves the welding quality and the production efficiency of the welding product.

[0005] In a first aspect, the embodiments of the present application provide a spot welding STC curve recommendation method, comprising: obtaining real-time spot welding conditions of a current spot welding operation; inputting the real-time spot welding conditions into a pre-constructed STC curve recommendation model to obtain an optimal STC curve; wherein the optimal STC curve describes the ideal output current value of the spot welding equipment during the spot welding operation; and performing the spot welding operation according to the optimal STC curve.

[0006] In some embodiments, the method further comprises: obtaining relevant data in the spot welding field, classifying all the relevant data according to attributes to obtain a plurality of spot welding data reference tables; wherein the relevant data comprises professional knowledge in the spot welding field and business data of historical spot welding operations; for each spot welding data reference table, each piece of data in the spot welding data reference table is semantized to obtain corresponding semantic spot welding data; wherein the semantic spot welding data represents the business data of the historical spot welding operations and the context information between each piece of business data; a training data set comprising the semantic spot welding data is constructed, a base language model is trained according to the training data set, and a pre-constructed STC curve recommendation model is obtained.

[0007] In some embodiments, the method further comprises: obtaining non-inferential welding data, and adding the non-inferential welding data to the training data set.

[0008] In some embodiments, the real-time spot welding conditions include at least one of the total thickness of the welding components of the current spot welding operation, the hierarchical structure between the welding components, the material parameters of each welding material, and the specifications of the spot welding equipment.

[0009] In some embodiments, inputting the real-time spot welding conditions into the pre-constructed STC curve recommendation model to obtain the optimal STC curve comprises: determining a mapping relationship between the real-time spot welding conditions and the ideal output current value of the spot welding equipment; sorting the mapping relationship in the time dimension to obtain a target mapping sequence; wherein the time dimension is aligned with the operation time of the spot welding operation; and determining the optimal STC curve according to the target mapping sequence.

[0010] In some embodiments, the method further comprises: integrating the real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation, and taking the integrated real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation as incremental data; and using the incremental data to perform incremental training on the pre-constructed STC curve recommendation model to obtain an updated STC curve recommendation model.

[0011] In some embodiments, the method further comprises: displaying the optimal STC curve on a visual graphical user interface.

[0012] According to the spot welding STC curve recommendation method provided in the present application, the base language model is trained with spot welding field knowledge, so that the model can comprehensively understand the spot welding process, and the model can be updated and optimized by learning and analyzing new materials, further effectively improving the prediction accuracy of the spot welding STC curve, obtaining a more optimal STC spot welding curve that meets the actual demand, and performing spot welding operation according to the optimal STC curve, which can effectively meet the diversified welding demand, while ensuring the welding quality and improving the spot welding production efficiency.

[0013] In a second aspect, the embodiments of the present application provide a spot welding STC curve recommendation device, comprising:

[0014] An acquisition module configured to acquire real-time spot welding conditions of a current spot welding operation;

[0015] A prediction module configured to input the real-time spot welding conditions into a pre-constructed STC curve recommendation model to obtain an optimal STC curve; wherein the optimal STC curve describes an ideal output current value of a spot welding device in a spot welding operation process;

[0016] An output module configured to perform the spot welding operation according to the optimal STC curve.

[0017] According to the spot welding STC curve recommendation device provided by the present application, the base language model is trained with spot welding field knowledge, so that the model can comprehensively understand the spot welding process, and at the same time, the model can be updated and optimized through learning and analysis of new materials, thereby further effectively improving the prediction accuracy of the spot welding STC curve, obtaining a more optimal STC spot welding curve that meets actual needs, and performing the spot welding operation according to the optimal STC curve, which can effectively meet diversified welding needs, ensure welding quality, and improve spot welding production efficiency.

[0018] In a third aspect, the embodiments of the present application provide an electronic device, comprising a processor and a memory, wherein the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the steps of the spot welding STC curve recommendation method of the first aspect.

[0019] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, wherein the readable storage medium stores programs or instructions, and the programs or instructions are executed by the processor to implement the steps of the spot welding STC curve recommendation method of the first aspect.

[0020] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0021] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the appended drawings.

[0022] Figure 1 A flowchart of the spot welding STC curve recommendation method provided by the embodiments of the present application;

[0023] Figure 2 A flowchart of pre-processing all related data provided by the embodiments of the present application;

[0024] Figure 3A visual graphical user interface input interface schematic diagram provided by an embodiment of the present application;

[0025] Figure 4 A visual graphical user interface output interface schematic diagram provided by an embodiment of the present application;

[0026] Figure 5 A flowchart of a spot welding STC curve recommendation method provided by an embodiment of the present application;

[0027] Figure 6 A schematic diagram of a spot welding STC curve recommendation device provided by an embodiment of the present application;

[0028] Figure 7 A structure schematic diagram of an electronic device in an embodiment of the present application.

[0029] The reference signs: 610-acquisition module; 620-prediction module; 630-output module; 710-processor; 720-memory; 730-input / output interface; 740-communication interface; 750-bus. DETAILED DESCRIPTION

[0030] Embodiments of the present application will be described in more detail by referring to the attached drawings. Although certain embodiments of the present application are shown in the drawings, it is understood that the present application can be implemented in various forms and should not be interpreted as being limited to the embodiments set forth herein, but rather these embodiments are provided to make the present application more thorough and complete. It is understood that the drawings and embodiments of the present application are for exemplary purposes only and are not intended to limit the scope of protection of the present application.

[0031] It should be understood that each of the steps recited in the method embodiments of the present application can be executed in different orders and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the execution of the steps shown. The scope of the present application is not limited in this respect.

[0032] As recited in the background section, the process quality of spot welding technology is highly dependent on the dynamic optimization of welding parameters (such as current, time, electrode pressure, etc.), and usually the STC technology is used to realize the rapid construction of the plate combination parameter library through the test piece welding and the collection of curves, so as to reduce the on-site debugging cost.

[0033] STC (Sheet Thickness Combination) is a method for quickly debugging spot welding process parameters. Its core goal is to simulate the welding conditions of different sheet combinations (such as different materials and thicknesses) in the laboratory, avoiding reliance on actual vehicles or complex workpieces when debugging directly in the production site. By using test pieces made of the same material as those used in actual production (such as steel and aluminum plates) in the laboratory, welding is performed according to different thickness combinations. Furthermore, sensors monitor the dynamic resistance changes during the welding process in real time, forming a resistance-time curve. Based on this resistance curve analysis, optimized welding parameters (such as current) are obtained, and a process solution is developed that is suitable for this sheet combination.

[0034] However, the generation of traditional STC curves relies on empirical formulas, which require engineers' experience and experimental data to generate recommended STC curves. These formulas have poor adaptability to new materials and processes, and are difficult to cope with complex welding conditions, which may affect welding quality.

[0035] When implementing this application, the applicant discovered that there are still some defects when using machine learning or data-driven methods to optimize the STC curve. For example, the general model is not specifically optimized for the spot welding field, which makes the model's prediction results less accurate; the professional knowledge in the spot welding field (such as welding physics, material properties, etc.) is not fully integrated into the model training process, which makes the prediction results deviate from actual needs; it is difficult to iteratively update the model through real-time data from the production line, etc. In this context, the applicant uses a basic language model to train the model through a large amount of welding data, further combines various physical laws to build a knowledge system and embed the model's prediction mechanism. At the same time, it can also integrate production line data to form a more flexible and more accurate STC curve recommendation system.

[0036] The following describes a spot welding STC curve recommendation method according to an embodiment of the present application with reference to the accompanying drawings.

[0037] refer to Figure 1 , which is a flow chart of the spot welding STC curve recommendation method provided in an embodiment of the present application.

[0038] Step S101, obtaining the real-time spot welding conditions of the current spot welding operation.

[0039] Specifically, the real-time spot welding conditions may be obtained through sensors or detection equipment, or may be directly input by a user.

[0040] As an optional embodiment, the real-time spot welding conditions include at least one of the total thickness of the welded parts in the current spot welding operation, the layered structure between the welded parts, the material parameters of each welding material, and the specifications of the spot welding equipment.

[0041] Specifically, the total thickness of the welded component can refer to the overall thickness of the stacked multiple layers of metal plates after welding (for example: the total thickness of two layers of 1.5mm steel plates is 3mm); the hierarchical structure between the welded components can refer to the number of layers, arrangement order and combination mode of the welded components (for example: a sandwich structure of steel plate-aluminum plate-steel plate); the material parameters of each welding material can include the physical properties of the metal such as resistivity, thermal conductivity, melting point, etc. (for example, the resistivity of low-carbon steel is about 1.5*10 -7 The specification of the spot welding equipment can be the maximum output current of the equipment (such as 20kA), the electrode pressure range (such as 2-8kN), the control accuracy and other hardware capability indicators, which are used to represent the process boundary that can be achieved by the spot welding equipment.

[0042] In step S102, the real-time spot welding conditions are input into the pre-constructed STC curve recommendation model to obtain an optimal STC curve; wherein the optimal STC curve describes the ideal output current value of the spot welding equipment in the spot welding operation process.

[0043] Specifically, the pre-constructed STC curve recommendation model is a basic language model that is pre-trained and fine-tuned in the field. The model is for spot welding technology and is obtained by adjusting the parameters of the initial field basic language model using a pre-stored data set to improve the recommendation accuracy of the model. When the real-time spot welding conditions of the current spot welding operation are obtained, the real-time spot welding conditions are input into the pre-constructed STC curve recommendation model, which will calculate and reason inside the model according to the input spot welding conditions to generate the optimal STC curve. In the embodiments of the present application, the optimal STC curve can be a curve describing the relationship between the ideal output current value of the spot welding equipment and time change in the spot welding operation process.

[0044] In step S103, the spot welding operation is performed according to the optimal STC curve.

[0045] Specifically, the optimal STC curve is output, and the user can guide the actual spot welding process according to the generated optimal STC curve to achieve the best spot welding effect.

[0046] As an optional embodiment, the method further comprises: obtaining related data in the spot welding field, classifying and arranging all related data according to attributes to obtain a plurality of spot welding data reference tables; wherein the related data includes professional knowledge in the spot welding field and business data of historical spot welding operations; for each spot welding data reference table, each data in the spot welding data reference table is semantized to obtain corresponding semantic spot welding data; wherein the semantic spot welding data represents the business data of the historical spot welding operations and the context information between each business data; a training data set including the semantic spot welding data is constructed, a basic language model is trained according to the training data set, and a pre-constructed STC curve recommendation model is obtained

[0047] Specifically, in the training process of the base language model, first, relevant data in the spot welding field needs to be obtained.

[0048] The relevant data can be collected manually, for example, for enterprise users, spot welding related data can be manually collected from internal documents, production records, equipment logs, quality detection reports, etc.; for spot welding process data, it can be real-time collected running data of spot welding equipment (including current, voltage, pressure, welding time, etc.) through sensors, data acquisition systems, etc. automation means; relevant data can also be obtained from public data sets, academic literature, industry reports, standard documents, etc. spot welding related technical parameters, material properties, industry standards, etc.

[0049] The relevant data in the spot welding field includes professional knowledge in the spot welding field and business data of historical spot welding operations.

[0050] The professional knowledge in the spot welding field can be spot welding process parameters, spot welding equipment information, material properties, quality control standards, and industry standards, etc., wherein the spot welding process parameters include current, voltage, pressure, welding time, pulse parameters, etc.; the spot welding equipment information includes equipment model, manufacturer, maintenance record, etc.; the material properties include welding material type, thickness, conductivity, thermal conductivity, etc.; the quality control standards include welding strength, weld size, defect detection standards, etc.; the industry standards include international and industry standards such as ISO, AWS, etc.; the academic literature includes the latest research results, experimental data, etc. of spot welding technology.

[0051] The business data of historical spot welding operations can be internal enterprise data, welding environment data, welding process optimization data, welding defect database, and welding material supplier data, etc., wherein the internal enterprise data includes production records, equipment operation data, quality detection reports, etc.; the welding environment data includes environmental conditions such as temperature and humidity in the welding workshop; the welding process optimization data includes optimization test data of process parameters, etc.; the welding defect database includes common welding defect types and their solutions, etc.; the welding material supplier data includes material performance data and welding recommendations provided by the supplier, etc.

[0052] It should be noted that the relevant data needs to be updated and maintained regularly. Regularly obtaining the latest spot welding data can ensure that the data always reflects the latest production environment and process changes, etc., to avoid model deviation caused by information lag, so that the recommendation model can accurately match the current actual working conditions; Regularly check and clean up the data, through regular data maintenance and cleaning, the accuracy of the data can be maintained, to prevent errors or biases in the recommended results caused by errors or missing, and at the same time, it can also avoid too much redundant information to bring computational burden to the model.

[0053] Reference Figure 2 The flowchart for preprocessing all relevant data provided for the embodiments of the present application.

[0054] Further, the relevant data is preprocessed according to the attributes of all relevant data, so that the original data can be converted into a format suitable for analysis and modeling.

[0055] Step S201, cleaning, outlier processing and repeated value processing of the relevant data.

[0056] Specifically, the cleaning of the data includes missing value processing, which can check whether there is missing data or incomplete data due to sensor failure in the data. The processing of missing values needs to delete the data with too many missing values first, and then use the mean interpolation method to fill in the missing values. For the null value of the category type, the mode (i.e. the highest frequency option of this field) also needs to be filled; there may be outliers in the relevant data, such as current or voltage values that exceed the reasonable range, which need to be filtered based on the statistical 3σ principle (values outside the range of mean ± 3 times standard deviation are considered abnormal), and the outliers of current, voltage and other parameters are selected; further, combined with process knowledge (such as the maximum current limit of the device), the nature of the abnormal value is judged to determine whether the abnormal value needs to be removed or adjusted to a reasonable threshold; in the update iteration of the model, due to the approximation or consistency of the welding conditions, repeated data may be input into the model, in order to avoid the burden on the model calculation, the repeated data values need to be deleted.

[0057] Step S202, standardizing the relevant data.

[0058] Specifically, in the relevant data, due to the different spot welding conditions, there may be various units of measurement in the experimental data, which need to be converted to a unified scale, for example, the unit of material thickness is unified to millimeters (mm), the unit of current is unified to amperes (A), etc., to avoid analysis errors caused by unit confusion; at the same time, there may be data with the same semantics but different expressions in the relevant data, which need to be converted to a standardized and unified name, for example, "welding gun pressure" and "electrode pressure" are unified to "electrode pressure (kN)".

[0059] Step S203, storing the relevant data.

[0060] Further, the preprocessed relevant data is classified and arranged to obtain a spot welding data reference table, and the spot welding data reference table is stored in the database of the basic language model, so as to facilitate subsequent analysis and modeling. The spot welding data reference table can include:

[0061] Spot Welding Process Parameter Table: This table is used to store real-time collected process parameter data during spot welding. It is the core table for spot welding data collection and records key parameters such as material name, thickness, plating, glue application, number of layers, total thickness, electrode cap diameter, welding gun, STC curve number, current, voltage, pressure, and welding time.

[0062] Material Property Table: This table is used to store the properties of welding materials, including material type, thickness, electrical conductivity, and thermal conductivity.

[0063] Industry Standard Table: This table is used to store industry standards related to spot welding, including international standards (such as ISO) and industry standards (such as AWS).

[0064] Spot Welding Knowledge Table: This table is used to store relevant knowledge and explanations about spot welding processes. It is a core reference in the field of spot welding and includes key knowledge such as basic concepts, welding principles, process characteristics, equipment requirements, and solutions to common problems.

[0065] Spot Welding Equipment Information Table: This table is used to store basic information about spot welding equipment, including equipment model, manufacturer, and maintenance records.

[0066] Quality Control Standard Table: This table is used to store control standards for welding quality, such as welding strength, weld size, and defect detection standards.

[0067] Academic Literature Table: This table is used to store the latest research findings and experimental data related to spot welding technology, which can be used as a reference and guide for improving welding processes.

[0068] Internal Enterprise Data Table: This table is used to store internal production records, equipment operation data, and quality detection reports, which can be used to manage internal production and control product quality.

[0069] Welding Process and Material Association Table: This table is used to store the association between welding process parameters and material properties, which can determine the applicable welding parameters for specific materials.

[0070] Welding Process and Standard Association Table: This table is used to store the association between welding process parameters and industry standards to ensure that the welding process meets relevant standards.

[0071] Quality Detection Report Table: This table is used to store quality detection reports for each welding operation, including detection results and defect information, which can be used to track the quality of spot welding and further improve the reference.

[0072] Equipment Maintenance Record Table: This table is used to store detailed maintenance records of equipment, including maintenance date, maintenance content, and maintenance personnel.

[0073] Welding Environment Data Table: Used to store environmental conditions such as temperature, humidity, etc. in the welding workshop, providing data support for analyzing the impact of environmental factors on welding quality.

[0074] Welding Process Optimization Data Table: Used to store optimization test data of process parameters, including parameter values before and after optimization and optimization effect description, to guide continuous improvement of welding process.

[0075] Welding Defect Database: Used to store common welding defect types, descriptions, discovery times, and solutions to facilitate quick positioning and solving of welding quality problems.

[0076] Welding Material Supplier Data Table: Used to store material performance data and welding recommendations provided by suppliers, providing reference for the selection and use of welding materials.

[0077] Step S204, semanticize the relevant data.

[0078] Specifically, in order to facilitate the basic language model to understand the data more deeply, it is necessary to semanticize each piece of data in the spot welding data reference table. Data semanticization can convert each piece of data information in the spot welding data reference table into an expression form that contains clear business meaning and associated background information, making it go beyond simple numerical values or codes, and can intuitively reflect the operation logic and context relationship in the actual scene, to enhance the readability and comprehensiveness of the data, and thus improve the reliability of the output results of the basic language model.

[0079] For the above different spot welding data reference tables, the meaning represented by the semanticized data can be as follows:

[0080] The semantic spot welding data corresponding to each piece of data in the spot welding process parameter table can represent a specific combination of welding process parameters, used to guide welding operations. For example: when using a B20 / 8 electrode cap and a Gun11 welding gun to weld a 2-layer plate with a total thickness of 2.65 mm, where the first layer is a material (HC340 / 590DPD) with a thickness of 0.65 mm, a Z 50\50 plating layer, and no adhesive, and the second layer is a material (HC260LAD) with a thickness of 2 mm, no plating layer, and no adhesive, the recommended parameter settings are: current 9000A, pressure 3000N, and welding time 500ms.

[0081] The semantic spot welding data corresponding to each piece of data in the spot welding equipment information table can represent the basic information and maintenance history of a spot welding device. For example: the device model X2000, manufactured by ABC Company, purchased on January 1, 2020, the last maintenance record is September 15, 2023, and the device status is running.

[0082] The semantic spot welding data corresponding to each data in the material properties table can describe the physical and chemical properties of a welding material. For example: Material B is an aluminum alloy with a thickness of 2mm, electrical conductivity of 35MS / m, and thermal conductivity of 200W / m·K, suitable for high-strength welding.

[0083] The semantic spot welding data corresponding to each data in the quality control standard table can define the standard requirements of a welding quality. For example: for Material C, the welding strength should not be less than 500MPa, the weld diameter should be between 5mm and 7mm, and the defect detection method is ultrasonic detection.

[0084] The semantic spot welding data corresponding to each data in the industry standard table can represent an international or industry standard related to spot welding. For example: ISO9001 standard is applicable to welding quality management system, which requires that the welding process must be recorded and regularly audited.

[0085] The semantic spot welding data corresponding to each data in the academic literature table can represent a research literature related to spot welding technology. For example: the literature "Spot welding process optimization research" was published by Zhang San et al. in 2022, which studied the variation law of welding strength under high current.

[0086] The semantic spot welding data corresponding to each data in the enterprise internal data table can represent the detailed data of a production activity. For example: on October 1, 2023, 1000 welds were produced using equipment X2000 and process parameter combination P001, and the quality detection report showed that the qualified rate was 98%.

[0087] The semantic spot welding data corresponding to each data in the welding process and material association table can describe the applicable relationship between a welding process parameter combination and a specific material. For example: process parameter combination P001 is suitable for material A, which can achieve high-strength welding.

[0088] The semantic spot welding data corresponding to each data in the equipment maintenance record table can describe the detailed information of a device maintenance activity. For example: equipment X2000 was regularly maintained on September 15, 2023, and the electrode tip was replaced, and the maintenance personnel was Li Si.

[0089] The semantic spot welding data corresponding to each data in the quality detection report table can represent the results of a welding quality detection. For example: the detection report on October 1, 2023 showed that the weld qualified rate of production batch B001 was 98%, and the main defect was the small size of the weld.

[0090] The semantic spot welding data corresponding to each data in the welding process and standard association table can describe the compliance relationship between a welding process parameter combination and an industry standard. For example: process parameter combination P001 conforms to ISO9001 standard, suitable for high-strength welding.

[0091] The semantic spot welding data corresponding to each piece of data in the welding environment data table can reflect the environmental conditions of the welding workshop at a specific time point, such as temperature and humidity, providing data support for analyzing the impact of environmental factors on welding quality. For example, the record on October 1, 2023 shows that the workshop temperature was 22°C and the humidity was 45%, and at that time the welding quality of production batch B001 was good.

[0092] The semantic spot welding data corresponding to each piece of data in the welding process optimization data table can store the optimization test data of process parameters, including the parameter values before and after optimization and the description of optimization effect, for guiding the continuous improvement of welding process. For example, the optimization test for process parameter combination P001 showed that after increasing the current from 2000A to 2100A, the weld strength increased by 12%.

[0093] The semantic spot welding data corresponding to each piece of data in the welding defect database can record the common welding defect types, descriptions, discovery times and solutions for each record, facilitating quick positioning and solving of welding quality problems. For example, the welding defect found on October 1, 2023 was porosity, which occurred in production batch B001, and this problem was solved by adjusting the welding time.

[0094] The semantic spot welding data corresponding to each piece of data in the welding material supplier data table can store the material performance data and welding suggestions provided by the supplier, providing reference basis for the selection and use of welding materials. For example, the high-strength steel material provided by supplier XYZ suggests a welding current of 2000A, which is suitable for production batch B001.

[0095] Further, a training data set including semantic spot welding data is constructed, and a base language model is trained according to the training data set to obtain a pre-constructed STC curve recommendation model.

[0096] Specifically, the training data set of the semantic spot welding data after being sorted is input into the base language model, and the base language model can be trained according to the input training data set to obtain the constructed STC curve recommendation model. The base language model can be a large language model such as GPT4, GPT-4o, Claude, and DeepSeek-V3. In the embodiments of the present application, the DeepSeek-V3 model is preferably used for training.

[0097] In order for the basic language model to more comprehensively and accurately understand the forecast demand, the raw data in the training dataset needs to be converted into standardized training corpus that adapts to the machine learning model input specifications. The data in the training dataset should include at least one of the preprocessed relevant data and semantic spot welding data. Typically, the training corpus includes prompt information and completion information, corresponding to the preprocessed relevant data and semantic spot welding data respectively. Taking the spot welding process as an example, the generated training corpus can be:

[0098] {

[0099] "Prompt":"Welding point id":"40604573",

[0100] Total thickness: 2.65,

[0101] "Level": "2",

[0102] "Material 1": "HC340 / 590DPD",

[0103] "Thickness 1": "0.65",

[0104] "Coating 1": "Z 50\50",

[0105] "Glue 1": "",

[0106] "Material 2": "HC260LAD",

[0107] "Thickness 2": "2.0",

[0108] "Coating 2": "",

[0109] "Glue 2": "",

[0110] "Welding gun": "Gun11",

[0111] "Electrode cap diameter": "B20 / 8",

[0112] Current: 9000,

[0113] "Pressure": "3000",

[0114] "Welding time": "500",

[0115] "Completion":”

[0116] In the welding of a 2-layer plate with a total thickness of 2.65 mm using a B20 / 8 electrode cap and a Gun11 welding gun, wherein the first layer is a material (HC340 / 590DPD) with a thickness of 0.65 mm, a Z50\50 plating layer, and no adhesive, and the second layer is a material (HC260LAD) with a thickness of 2 mm, no plating layer, and no adhesive, the recommended parameter settings are: current 9000 A, pressure 3000 N, and welding time 500 ms.

[0117] As an optional embodiment, the real-time spot welding conditions are input into the pre-constructed STC curve recommendation model to obtain the optimal STC curve, which further includes: determining the mapping relationship between the real-time spot welding conditions and the ideal output current value of the spot welding equipment; sorting the mapping relationship in the time dimension to obtain a target mapping sequence; wherein the time dimension is aligned with the operation time of the spot welding operation; and determining the optimal STC curve according to the target mapping sequence.

[0118] Specifically, in the embodiments of the present application, the DeepSeek-V3 model can determine the complex mapping relationship between the real-time spot welding conditions and the ideal output current value of the spot welding equipment according to the real-time spot welding conditions, and further arrange the real-time spot welding conditions at each time (such as every millisecond or every welding point period) and the corresponding ideal current value in chronological order to form a target mapping sequence.

[0119] The model can first analyze the professional knowledge in the field of spot welding, such as the welding physical mechanism: in the process of spot welding, the size of the current and the welding time will affect the nugget size, strength and other key indicators of the welding spot, preliminarily understand the interaction between different related parameter data, and form the mapping relationship between the real-time welding conditions and the ideal output current value of the welding equipment, for example, for thicker plates, larger current and longer welding time may be needed to ensure the strength of the welding spot, while also considering avoiding welding defects caused by overheating.

[0120] The model will also combine the real-time welding conditions and the corresponding STC curve in the historical data, and analyze the relationship between the historical data using a machine learning algorithm (such as a neural network), wherein the historical data contains successful welding parameter combinations under different welding conditions, for example, the model analyzes the data of the ideal output current value and the welding time corresponding to different plate thicknesses and plating types to obtain the corresponding mapping relationship, time t1:{material thickness = 2mm, temperature = 25℃}→current = 10kA; time t2:{material thickness = 2mm, temperature = 28℃}→current = 10.2kA, and so on to form a time sequence [I(t1), I(t2),..., I(t n )] corresponding to the real-time spot welding conditions, and further align the time sequence with the current to form the optimal STC curve corresponding to the real-time spot welding conditions.

[0121] As an optional embodiment, the method further comprises: obtaining non-reasoning welding data, and adding the non-reasoning welding data to the training data set.

[0122] Specifically, during the long-term operation and continuous updating of the model, there may be regular error data, and the model may integrate the error data into the model during the updating process, thereby causing the performance of the model to decrease (degrade), and therefore, non-reasoning data needs to be fused into the model to improve the accuracy and robustness of the model.

[0123] Non-reasoning data is data that cannot be directly calculated by model reasoning, such as abnormal data or welding data generated under special conditions (such as device failure, extreme changes in environmental conditions, material batch differences, etc.), which can provide error data reference for the model and is beneficial to the positive optimization of the model.

[0124] The model can fuse the non-reasoning data with the data for determining the mapping relationship, wherein the processing technology for fusing the data can be to use an anomaly detection algorithm to identify abnormal data and analyze the causes thereof, or to analyze abnormal welding results caused by device failure, learn the influence of device failure on welding parameters, and thus enable the model to consider the device state when making recommendations according to real-time spot welding conditions in the future.

[0125] In addition, the model can also adjust the parameters of the model through some optimization algorithms such as Bayesian optimization, and can comprehensively consider the non-reasoning data and the data for determining the mapping relationship, so that the model can maintain good performance in the environment with non-reasoning data, for example, when the temperature of the environment where the device is located changes greatly, by fusing the non-reasoning data generated in this environment, the model can adjust the recommendation strategy of the welding parameters according to the new environmental conditions to ensure the effect of the welding process.

[0126] As an optional embodiment, when the STC curve recommendation model is constructed, the model also needs to be fine-tuned by LoRA to save the calculation amount of the model.

[0127] Specifically, when fine-tuning a large language model, to avoid directly updating all parameters and consuming huge video memory resources, the low-rank matrix can be used to approximate the parameter increment to reduce the training parameter amount, thereby saving the calculation video memory and calculation resources.

[0128] The fine-tuning parameters include batch size, initial learning rate, learning rate scheduling strategy, LoRA Rank, optimizer, and training rounds, etc. The batch size is the number of training samples per time, used to balance the GPU utilization and gradient stability. The initial learning rate is used to control the initial amplitude of parameter update, determine the step size of parameter adjustment at the beginning of training, and affect the convergence speed and stability. The learning rate scheduling strategy is used to dynamically adjust the learning step of the model, that is, to dynamically adjust the learning rate during training to improve the convergence effect. The LoRA Rank is used to determine the representation ability of the adapter. The higher the rank value of the low-rank matrix, the stronger the representation ability (close to full parameter fine-tuning), but the corresponding GPU memory occupancy will increase. The optimizer is used to ensure the smooth update of model parameters, such as AdamW (adaptive momentum optimization) and SGD (stochastic gradient descent), which can control the parameter update direction and step size. The training rounds are used to describe the number of iterations of the model on the data set, which can prevent the model from underfitting (too few rounds) or overfitting (too many rounds).

[0129] The fine-tuning techniques include parameter freezing strategy, mixed precision training, and dynamic load balancing. The parameter freezing strategy can be to fix 90% of the parameters of the base model and only fine-tune the key parts such as the MoE routing layer and the last 3 Transformer blocks. The MoE routing layer can control weight distribution, and the last 3 Transformer blocks usually carry task-related features. The parameter freezing strategy can reduce GPU memory occupancy while preserving the core knowledge of the model. The mixed precision training can use FP8 format to store gradient accumulators to reduce GPU consumption by 30%, while keeping FP16 precision for attention matrix calculation to avoid numerical overflow or underflow. By using low precision (FP8) for memory-sensitive parts and high precision (FP16) for numerical-sensitive parts, the numerical stability can be ensured. Dynamic load balancing dynamically adjusts the expert selection threshold based on the STC curve classification, for example: when inputting long text, the STC curve predicts that more experts are needed to participate, and the routing layer threshold is reduced. At the same time, domain knowledge is introduced as a reward signal (such as code generation task, grammatical correctness as auxiliary reward) to guide the optimization of the routing layer.

[0130] It should be noted that in actual application, the Rank and training strategy need to be adjusted according to the task complexity to balance the effect and resource consumption.

[0131] As an optional embodiment, the method further comprises: integrating the real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation, taking the integrated real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation as incremental data; and using the incremental data to perform incremental training on the pre-constructed STC curve recommendation model to obtain an updated STC curve recommendation model.

[0132] Specifically, the pre-constructed STC curve recommendation model also needs to be dynamically optimized and updated during application. The real-time spot welding conditions of the current input spot welding job and the optimal STC curve of the current spot welding job need to be integrated to form an incremental data pair of "state-action" (i.e., to form a training sample that can drive model updating). Further, the new data is input into the model, and the model is locally updated or the parameters are efficiently fine-tuned. This can avoid repeated training of full data, reduce the computational overhead of the model, and further, after the training is completed, the updated model needs to be evaluated to check its performance on new data. This can be done by cross-validation, ROC curve, and other evaluation indicators to comprehensively evaluate the accuracy and robustness of the model. The model can also determine whether further adjustment or optimization of the model is needed based on the evaluation results. When adjustment or optimization is needed, the parameters and structure of the model are updated incrementally and adjusted adaptively to adapt to new knowledge distribution.

[0133] As an optional embodiment, the method further comprises: displaying the optimal STC curve on a visual graphical user interface.

[0134] Specifically, the visual graphical user interface is used for interaction with the user, and can provide the user with an intuitive and convenient interface. The user can input or select real-time spot welding conditions in the graphical user interface (refer to Figure 3 ) to obtain the predicted optimal STC curve (refer to Figure 4 ). The vertical coordinate of the optimal STC curve is current, and the horizontal coordinate is time. By observing the curve, the rising, stable, and falling stages of the current during welding can be understood, and whether there is abnormal fluctuation can be determined, so as to evaluate and control the welding quality.

[0135] Referring to Figure 5 , a flowchart of a spot welding STC curve recommendation method provided by an embodiment of the present application.

[0136] In the embodiment of the present application, the spot welding STC curve recommendation method first collects and preprocesses the data in the spot welding field, and further inputs the processed training data set into the basic language model for training to obtain a pre-constructed STC curve recommendation model.

[0137] When the user inputs the real-time spot welding conditions of the current spot welding operation into the pre-constructed STC curve recommendation model, the real-time spot welding conditions are input into the pre-constructed STC curve recommendation model to obtain the optimal STC curve, which is further displayed to the user by the visual graphical user interface, and at the same time, the current optimal real-time spot welding conditions and the STC curve are integrated into incremental data and input into the pre-constructed STC curve recommendation model again, so as to dynamically optimize and update the pre-constructed STC curve recommendation model to obtain a new pre-constructed STC curve recommendation model for the next prediction.

[0138] According to the spot welding STC curve recommendation method provided in the present application, the base language model is trained in the field of spot welding knowledge, so that the model can comprehensively understand the spot welding process, and at the same time, the model can be updated and optimized by learning and analyzing new materials, further effectively improving the prediction accuracy of the spot welding STC curve, obtaining a more optimal STC spot welding curve that meets the actual demand, and performing spot welding operation according to the optimal STC curve, which can effectively meet the diversified welding demand, ensure the welding quality, and improve the spot welding production efficiency.

[0139] Reference Figure 6 The present application provides a spot welding STC curve recommendation device.

[0140] Based on the same concept, the present application also provides a spot welding STC curve recommendation device corresponding to the spot welding STC curve recommendation method provided in any of the above embodiments.

[0141] The spot welding STC curve recommendation device comprises an acquisition module 610, a prediction module 620, and an output module 630.

[0142] The acquisition module 610 is configured to acquire real-time spot welding conditions of a current spot welding operation; the prediction module 620 is configured to input the real-time spot welding conditions into a pre-constructed STC curve recommendation model to obtain an optimal STC curve; wherein the optimal STC curve describes an ideal output current value of a spot welding device in a spot welding operation process; and the output module 630 is configured to perform spot welding operation according to the optimal STC curve.

[0143] In some embodiments, the acquisition module 610 is further configured to: acquire relevant data in the spot welding field, classify all the relevant data according to attributes to obtain a plurality of spot welding data reference tables; wherein the relevant data includes professional knowledge in the spot welding field and business data of historical spot welding operations, for each spot welding data reference table, each data in the spot welding data reference table is semantized to obtain corresponding semantic spot welding data; wherein the semantic spot welding data represents the business data of the historical spot welding operations and the context information between each piece of business data; construct a training data set including the semantic spot welding data, train the base language model according to the training data set to obtain the pre-constructed STC curve recommendation model.

[0144] In some embodiments, the acquisition module 610 is further configured to: acquire non-inferential welding data, and add the non-inferential welding data to the training data set.

[0145] In some embodiments, the real-time spot welding conditions include at least one of the total thickness of the welding components of the current spot welding operation, the hierarchical structure between the welding components, the material parameters of each welding material, and the specifications of the spot welding equipment.

[0146] In some embodiments, the prediction module 620 is further configured to: determine a mapping relationship between the real-time spot welding conditions and the ideal output current value of the spot welding equipment; sort the mapping relationship in the time dimension to obtain a target mapping sequence; wherein the time dimension is aligned with the operation time of the spot welding operation; determine the optimal STC curve according to the target mapping sequence

[0147] In some embodiments, the real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation are integrated, and the integrated real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation are used as incremental data; the pre-constructed STC curve recommendation model is incrementally trained using the incremental data to obtain an updated STC curve recommendation model.

[0148] In some embodiments, the output module 630 is further configured to: display the optimal STC curve on a visual graphical user interface.

[0149] According to the spot welding STC curve recommendation device provided in the present application, the base language model is trained with spot welding field knowledge, so that the model can comprehensively understand the spot welding process, and at the same time, the model can be updated and optimized through learning and analysis of new materials, further effectively improving the prediction accuracy of the spot welding STC curve, obtaining a more optimal STC spot welding curve that meets the actual demand, and performing spot welding operation according to the optimal STC curve, which can effectively meet the diversified welding demand, while ensuring the welding quality and improving the spot welding production efficiency.

[0150] Based on the same concept, corresponding to the spot welding STC curve recommendation method provided by any of the above embodiments, the present application also provides an electronic device, including a processor and a memory, the memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the steps of the spot welding STC curve recommendation method of the first aspect.

[0151] Figure 7 A more specific electronic device hardware structure diagram provided by the embodiments of the present application is shown, which can include: a processor 710, a memory 720, an input / output interface 730, a communication interface 740 and a bus 750. The processor 710, the memory 720, the input / output interface 730 and the communication interface 740 are connected to each other through the bus 750 for communication within the device.

[0152] The processor 710 can be implemented by a general CPU (Central Processing Unit, central processor), a microprocessor, an application specific integrated circuit (ASIC), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the embodiments of the present application.

[0153] The memory 720 can be implemented by ROM (Read Only Memory, read-only memory), RAM (Random Access Memory, random access memory), static storage device, dynamic storage device, etc. The memory 720 can store the operating system and other application programs, and when the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are saved in the memory 720 and executed by the processor 710.

[0154] The input / output interface 730 is used to connect the input / output module to realize information input and output. The input / output module can be configured as a component in the device (not shown in the figure), or can be externally connected to the device to provide corresponding functions. The input device can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device can include a display, a speaker, a vibrator, an indicator light, etc.

[0155] The communication interface 740 is used to connect the communication module (not shown in the figure) to realize the communication interaction between the device and other devices. The communication module can realize communication through wired mode (such as USB, network cable, etc.), or through wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0156] The bus 750 includes a path for transferring information between the various components (for example, the processor 710, the memory 720, the input / output interface 730, and the communication interface 740) of the device.

[0157] It should be noted that although the above device only shows the processor 710, the memory 720, the input / output interface 730, the communication interface 740, and the bus 750, in the specific implementation process, the device can also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device can also only contain the components necessary to implement the embodiments of the present application, and does not necessarily contain all the components shown in the figure.

[0158] The electronic device of the above embodiment is used to implement the corresponding spot welding STC curve recommendation method in any of the preceding embodiments, and has the beneficial effects of the corresponding spot welding STC curve recommendation method embodiments, which are not repeated here.

[0159] Based on the same concept, corresponding to the spot welding STC curve recommendation method provided by any of the above embodiments, the present application also provides a computer readable storage medium, and the above readable storage medium stores programs or instructions, and the programs or instructions are executed by the processor to implement the steps of the spot welding STC curve recommendation method of the first aspect.

[0160] The above computer readable storage medium can be any available medium or data storage device accessible by a computer, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid state disk (SSD), etc.).

[0161] The computer instructions stored in the storage medium of the above embodiment are used to make the computer execute the corresponding spot welding STC curve recommendation method in any of the preceding embodiments, and have the beneficial effects of the corresponding spot welding STC curve recommendation method embodiments, which are not repeated here.

[0162] Based on the same concept, corresponding to the spot welding STC curve recommendation method provided by any of the above embodiments, the present application also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer to make the computer and / or the processor execute the spot welding STC curve recommendation method. Corresponding to the execution subject of each step in each embodiment of the spot welding STC curve recommendation method, the processor executing the corresponding step can belong to the corresponding execution subject.

[0163] In particular, the computer program code for carrying out operations of the present application can be written in any of various programming languages, including object-oriented programming languages such as Java, Smalltalk, C++, as well as conventional procedural programming languages such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0164] It should be understood that each block of the flowchart and / or block diagram illustrations, and combinations of blocks in the flowchart and / or block diagram illustrations, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0165] These computer program instructions can also be stored in a computer- readable medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable medium produce an article of manufacture including instructions which implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0166] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0167] The computer program product of the above-mentioned embodiments is used for enabling a computer and / or processor to perform the spot welding STC curve recommendation method of any of the above-mentioned embodiments, and has the beneficial effects of the corresponding spot welding STC curve recommendation method embodiments, which are not repeated here.

[0168] It should be noted that, in the present document, 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 also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a", "comprising", or the like does not, without more constraints, preclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element. Furthermore, it is to be understood that the methods and apparatuses of the present application can be carried out by specific hardware, by software, or by a combination of hardware and software. It is therefore, contemplated to this patent to cover any and all modifications, variations, or equivalents that fall within the scope of the present application. Accordingly, where a concept can have been illustrated in only one of the exemplary embodiments, various aspects of the concept can be modified and / or combined to produce a variety of other embodiments that are not specifically illustrated. Thus, for purposes of describing the present application, certain aspects of the application can be presented in terms of sequences of actions, but it should be appreciated that these sequences are examples and are not limiting. The sequences of actions could be changed, and other sequences could be implemented. Moreover, it should be appreciated that sometimes it is easier to describe one aspect of the application in terms of another aspect of the application. Therefore, the description herein of one aspect of the application in terms of another aspect of the application is used merely to more particularly exemplify the application. Such a description should not be limiting since other aspects of the application could be used to describe the one aspect of the application.

[0169] From the above description of the embodiments, it is clear that the above-mentioned method can be realized by means of a computer software product and a general hardware platform, of course, it can also be realized by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disc, optical disc, etc.), and includes a plurality of instructions for making the terminal or network side device execute the method described in each embodiment of the present application.

[0170] The embodiments of the present application are described above in conjunction with the drawings, but the present application is not limited to the above-mentioned specific embodiments, and the above-mentioned specific embodiments are only illustrative, not limiting. Those skilled in the art can make many forms of embodiments under the inspiration of the present application without departing from the scope of the present application and the scope protected by the claims.

Claims

1. A spot welding STC curve recommendation method, characterized in that: include: Get the real-time spot welding conditions of the current spot welding operation; Inputting the real-time spot welding conditions into a pre-built STC curve recommendation model to obtain an optimal STC curve; wherein the optimal STC curve describes an ideal output current value of the spot welding equipment during the spot welding operation; Spot welding is performed according to the optimal STC curve.

2. The spot welding STC curve recommendation method according to claim 1, characterized in that: The method further comprises: Acquire relevant data in the field of spot welding, classify and organize all relevant data according to attributes, and obtain multiple spot welding data reference tables; wherein the relevant data includes professional knowledge in the field of spot welding and business data of historical spot welding operations; For each of the spot welding data reference tables, semantically transform each piece of data in the spot welding data reference table to obtain corresponding semantic spot welding data; wherein the semantic spot welding data represents the business data of the historical spot welding operation and the context information between each piece of business data; A training data set including the semantic spot welding data is constructed, and a basic language model is trained according to the training data set to obtain the pre-constructed STC curve recommendation model.

3. The spot welding STC curve recommendation method according to claim 2, characterized in that: The method further comprises: Non-inferential welding data is acquired and added to the training data set.

4. The spot welding STC curve recommendation method according to claim 3, characterized in that: The real-time spot welding conditions include at least one of the total thickness of welded parts in the current spot welding operation, the hierarchical structure between welded parts, material parameters of each welding material, and specifications of spot welding equipment.

5. The spot welding STC curve recommendation method according to claim 4, characterized in that: Inputting the real-time spot welding conditions into a pre-built STC curve recommendation model to obtain an optimal STC curve includes: Determining a mapping relationship between the real-time spot welding condition and an ideal output current value of the spot welding device; Sorting the mapping relationships in a time dimension to obtain a target mapping sequence; wherein the time dimension is aligned with the operation time of the spot welding operation; The optimal STC curve is determined according to the target mapping sequence.

6. The spot welding STC curve recommendation method according to claim 5, characterized in that: The method further comprises: Integrating the real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation, and using the integrated real-time spot welding conditions of the current spot welding operation and the optimal STC curve of the current spot welding operation as incremental data; The pre-built STC curve recommendation model is incrementally trained using the incremental data to obtain an updated STC curve recommendation model.

7. The spot welding STC curve recommendation method according to claim 5, characterized in that: The method further comprises: The optimal STC curve is displayed on a visual graphical user interface.

8. A spot welding STC curve recommendation device, characterized in that: include: an acquisition module configured to acquire real-time spot welding conditions of a current spot welding operation; A prediction module is configured to input the real-time spot welding conditions into a pre-built STC curve recommendation model to obtain an optimal STC curve; wherein the optimal STC curve describes an ideal output current value of the spot welding equipment during the spot welding operation; The output module is configured to perform spot welding according to the optimal STC curve.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the spot welding STC curve recommendation method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the spot welding STC curve recommendation method according to any one of claims 1 to 7 are implemented.

Images