Method, system and equipment for determining welding parameters of resistance spot welding

Abstract

The invention discloses a method, a system and equipment for determining welding parameters of resistance spot welding, and relates to the technical field of resistance spot welding. According to the method, workpiece attributes, welding parameters and welding quality information related to resistance spot welding are collected, a database of a distributed architecture is constructed, and the collected resistance spot welding related information is stored in a classified mode; establishing a material-parameter-quality mapping relation through a back propagation neural network model obtained through training; constructing an expert system rule base comprising a basic rule base, a parameter constraint rule base and a parameter optimization rule base; and input material and plate thickness information is received, and target welding parameters are determined through database retrieval or reasoning optimization in combination with a K-proximity algorithm and a genetic algorithm. According to the scheme, the problems that a traditional method is low in efficiency, poor in precision and weak in generalization ability are solved, efficient and accurate optimization of welding parameters is achieved, and the high requirement for resistance spot welding quality in the fields of automobile bodies in white and the like is met.

Description

A method, system and equipment for determining resistance spot welding parameters Technical Field

[0001] This invention relates to the field of resistance spot welding technology, and specifically to a method, system and equipment for determining resistance spot welding parameters. Background Technology

[0002] Welding is the primary material joining method in the automotive and aerospace industries. Due to its stable welding quality, high production efficiency, and ease of automation, resistance spot welding is the most important welding method in the automotive body-in-white welding process. It is widely used for connecting components such as the frame, roof, body panel, doors, body assemblies, and side panels. A typical car body contains thousands of resistance spot welds. Therefore, the quality of resistance spot welds is a key factor affecting the safety performance of the car body. For resistance spot welding, the design of welding parameters under different conditions is particularly important. In recent years, the development of automotive manufacturing technology has led to an increase in the number of welds on car bodies and a significant increase in process complexity. At the same time, the types of automotive body sheet materials are constantly diversifying, moving towards higher strength and coatings. Different sheet materials and combinations of thicknesses make it increasingly difficult to design efficient and high-quality resistance spot welding parameters.

[0003] Traditional resistance spot welding parameter design relies heavily on empirical formulas or experienced experts. This method is not only repetitive and inefficient, but also results in inconsistent parameter ranges for the same material combination from different engineers. Furthermore, the resistance spot welding process for materials used in entirely new vehicle models may not be fully documented in previous stable mass-produced vehicle welding data. With the increasing precision and automation in manufacturing, diverse material combinations and complex structural designs demand more adaptable welding parameters, rendering traditional welding manual-based parameter design systems inapplicable. On the other hand, increased production scale and accelerated product iteration require shorter parameter debugging cycles, and the efficiency of manual trial-and-error methods has become a production bottleneck.

[0004] Expert systems can combine expert knowledge and experience to transform discrete welding knowledge and experience into a reasoning-based rule engine through the construction of a structured knowledge database, achieving intelligent and standardized parameter design. Furthermore, with the development of intelligent manufacturing technology, the structure of expert systems is no longer limited to a single reasoning unit. Through the introduction of machine learning algorithms, the design of a "rule base + data model" rule knowledge base can link materials, parameters, and welding quality. Applying expert systems to the intelligent design of resistance spot welding parameters can provide a more scientific basis for welding parameters, meeting the design requirements of resistance spot welding parameters for automotive body-in-white.

[0005] Traditional resistance spot welding parameter design relies on the experience and knowledge of relevant experts and welding process manuals. However, with the continuous updates in automotive body materials and the constant changes in sheet thickness combinations, the previous method of designing welding parameters based solely on traditional experience and welding manuals can no longer meet the requirements. Summary of the Invention

[0006] To overcome the problems in existing technologies, namely that resistance spot welding parameter design relies on experience or welding manuals, resulting in low efficiency, poor accuracy, and weak generalization ability, this invention provides a method, system, and equipment for determining resistance spot welding parameters. By combining expert systems, machine learning algorithms, and optimization algorithms, efficient and accurate determination of welding parameters is achieved, improving welding quality and production efficiency.

[0007] To achieve the aforementioned objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides a method for determining resistance spot welding parameters, the method comprising:

[0009] Collect resistance spot welding related information; the resistance spot welding related information includes: workpiece attribute information, welding parameter information, and welding quality information;

[0010] A distributed architecture is used to build a resistance spot welding information database, and the collected resistance spot welding related information is classified and stored.

[0011] After preprocessing the welding parameter information and welding quality information, they are input into a BP neural network for training to obtain a backpropagation neural network model, which determines the mapping relationship between workpiece attributes, welding parameters and welding quality.

[0012] Based on the mapping relationship between the workpiece attributes, welding parameters and welding quality, an expert system rule base is established.

[0013] The system receives information on the material and thickness of the target workpiece, retrieves matching welding parameters, or performs inference and optimization through an expert system database to obtain the target welding parameters.

[0014] Optionally, the workpiece attribute information includes material type and plate thickness information;

[0015] The welding parameter information includes: electrode pressure, welding current, and welding time information;

[0016] The welding quality information includes tensile strength and weld nugget diameter information.

[0017] Optionally, the resistance spot welding related information may also include: supplementing the information corresponding to the missing plate thickness combinations through welding experiments.

[0018] Optionally, the step of classifying and storing the collected resistance spot welding related information includes: storing workpiece attribute information, welding parameter information, and welding quality information in the workpiece attribute database, welding parameter database, and welding quality database included in the resistance spot welding information database, respectively; establishing database tables through data definition statements and data manipulation statements, and performing information addition and deletion operations on the database tables; storing string-type resistance spot welding related information in a character data format; and storing numeric-type resistance spot welding related information in a numeric data format.

[0019] The numerical types include integer and floating-point types.

[0020] Optionally, the step of preprocessing the welding parameter information and welding quality information and then inputting them into a BP neural network for training to obtain a backpropagation neural network model includes: dividing the collected resistance spot welding related information into a training set and a test set according to a preset ratio; training the BP neural network using the training set; and verifying the generalization ability of the BP neural network using the test set.

[0021] The training results of the BP neural network are evaluated using a loss function. The initial weights and biases of the BP neural network are iteratively updated until the loss function meets the preset convergence condition, thereby obtaining the backpropagation neural network model.

[0022] In this model, the input layer neurons correspond to workpiece attribute information and welding parameter information, while the output layer neurons correspond to welding quality information.

[0023] Optionally, the expert system database includes: a basic rule base for storing the basic process principles and material-parameter matching benchmarks for resistance spot welding; a parameter constraint rule base for storing the range of resistance spot welding parameters corresponding to different materials and plate thicknesses; and a parameter optimization rule base that uses a trained BP neural network as a parameter optimization model to determine the initial welding parameter range and perform optimization.

[0024] Optionally, the step of receiving the material and thickness information of the target workpiece, retrieving matching welding parameters, or performing reasoning optimization through an expert system database to obtain the target welding parameters includes: retrieving the resistance spot welding information database based on the received material and thickness information of the target workpiece; if a completely matching case exists, the corresponding welding parameters are directly output; if no completely matching case exists, the similarity between the target workpiece and the case in the expert system rule base is calculated using the K-nearest neighbor algorithm, similar cases are selected to obtain the initial welding parameter range, and then a genetic algorithm is used to perform multi-objective optimization in the parameter optimization rule base to obtain the target welding parameters.

[0025] Optionally, the step of calculating the similarity between the target workpiece and the cases in the expert system rule base using the K-nearest neighbor algorithm includes: calculating the Euclidean distance between the material matching degree and plate thickness of the target workpiece and the cases in the expert system rule base, converting the Euclidean distance into a similarity, and filtering a preset number of similar cases based on the similarity.

[0026] In a second aspect, the present invention provides a resistance spot welding parameter determination system for implementing the method as described in any of the first aspects, comprising:

[0027] The information collection module is used to collect resistance spot welding related information, including: workpiece attribute information, welding parameter information, and welding quality information.

[0028] The database module is used to build a resistance spot welding information database using a distributed architecture, and to classify and store the collected resistance spot welding related information.

[0029] The mapping relationship acquisition module is used to preprocess the welding parameter information and welding quality information, input them into the BP neural network for training, obtain the backpropagation neural network model, and determine the mapping relationship between workpiece attributes, welding parameters and welding quality.

[0030] The rule base construction module is used to establish an expert system rule base based on the mapping relationship between the workpiece attributes, welding parameters and welding quality;

[0031] The target welding parameter determination module is used to receive the material and thickness information of the target workpiece, retrieve matching welding parameters, or perform reasoning optimization through an expert system database to obtain the target welding parameters.

[0032] Thirdly, the present invention provides an electronic device, the electronic device comprising:

[0033] At least one processor; and

[0034] A memory communicatively connected to the at least one processor; wherein,

[0035] The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the method described in any of the first aspects.

[0036] Compared with the closest existing technology, the present invention has the following advantages:

[0037] This invention proposes a method, system, and equipment for determining resistance spot welding parameters. By constructing a distributed resistance spot welding information database, it achieves systematic storage and rapid retrieval of resistance spot welding-related information, avoiding the tedious process of repeatedly consulting manuals and relying on manual experience in traditional methods. In the case of no perfectly matching cases, the K-nearest neighbor algorithm is used to quickly screen similar cases to obtain the initial parameter range, and then combined with a genetic algorithm for optimization, which greatly shortens the parameter design and debugging cycle.

[0038] Secondly, by utilizing a BP neural network to establish a nonlinear mapping relationship between workpiece attributes, welding parameters, and welding quality, the welding quality under different parameter combinations can be accurately predicted. Simultaneously, by combining the constraints and optimization functions of the expert system rule base, the designed welding parameters are ensured to meet process requirements and quality standards, effectively improving the design accuracy and reliability of welding parameters. By constructing an expert system rule base, the process principles and parameter constraints of resistance spot welding are standardized and structured, avoiding inconsistencies in parameter design caused by differences in the experience of different engineers. This achieves standardization and normalization of resistance spot welding parameter design, which is beneficial for ensuring the stability of welding quality.

[0039] This invention combines case-based reasoning, rule-based reasoning, and intelligent algorithms into a hybrid optimization method. It can not only handle parameter design problems corresponding to common material and plate thickness combinations, but also supplement missing data through welding experiments and optimize parameters for novel material and plate thickness combinations through intelligent algorithms. This solves the problem of weak generalization ability of traditional methods and is applicable to resistance spot welding parameter design for various material and plate thickness combinations. In particular, it can meet the actual production needs of fields such as automotive body-in-white and has broad application prospects. Attached Figure Description

[0040] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0041] Figure 1 is a flowchart of a method for determining resistance spot welding parameters provided by the present invention;

[0042] Figure 2 is a schematic diagram of the backpropagation neural network model in an embodiment of the present invention;

[0043] Figure 3 is a flowchart of the method for determining target welding parameters in an embodiment of the present invention;

[0044] Figure 4 is a schematic diagram of the similarity calculation process of the K-nearest neighbor algorithm in an embodiment of the present invention;

[0045] Figure 5 is a structural block diagram of a resistance spot welding parameter determination system according to an embodiment of the present invention;

[0046] Figure 6 is an internal structure diagram of the electronic device provided by the present invention. Detailed Implementation

[0047] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of the present invention and are therefore merely examples, and should not be construed as limiting the scope of protection of the present invention.

[0048] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0049] This invention provides a method, system, and equipment for determining resistance spot welding parameters, applicable to scenarios requiring resistance spot welding in the automotive, aerospace, and other fields, and particularly suitable for designing resistance spot welding parameters for automotive body-in-white.

[0050] Example 1: Please refer to Figure 1. Figure 1 shows a method for determining resistance spot welding parameters provided in Example 1 of the present invention. The method specifically includes the following steps:

[0051] S101, Collect resistance spot welding related information; the resistance spot welding related information includes: workpiece attribute information, welding parameter information and welding quality information;

[0052] S102, a distributed architecture is used to build a resistance spot welding information database, and the collected resistance spot welding related information is classified and stored.

[0053] S103, After preprocessing the welding parameter information and welding quality information, the data are input into a BP neural network for training to obtain a backpropagation neural network model and determine the mapping relationship between workpiece attributes, welding parameters and welding quality.

[0054] S104. Based on the mapping relationship between the workpiece attributes, welding parameters and welding quality, establish an expert system rule base;

[0055] S105 receives the material and thickness information of the target workpiece, retrieves matching welding parameters, or performs reasoning optimization through the expert system database to obtain the target welding parameters.

[0056] In step S101 above, collecting resistance spot welding related information specifically includes workpiece attribute information, welding parameter information, and welding quality information. Workpiece attribute information includes material type (e.g., high-strength steel, aluminum alloy, galvanized steel, etc.) and plate thickness; welding parameter information includes electrode pressure, welding current, and welding time; and welding quality information includes tensile strength and weld nugget diameter.

[0057] During the data collection process, welding parameter information is collected based on the plate thickness at preset intervals. For welding information of the corresponding plate thickness combination that has not been collected, welding parameter information and welding quality information under the plate thickness combination are obtained by conducting welding experiments and supplemented into the collected resistance spot welding related information to ensure the integrity and comprehensiveness of the data.

[0058] For example, if the plate thickness interval is set to 0.2mm, when welding information for a certain material with plate thicknesses of 1.0mm and 1.4mm is collected, but welding information for a plate thickness of 1.2mm is missing, welding experiments are carried out for the 1.2mm plate thickness of that material, and the tensile strength and weld nugget diameter are recorded under different combinations of electrode pressure, welding current and welding time. The optimal combination of welding parameters and the corresponding welding quality information are then selected and added to the dataset.

[0059] In step S102 above, the construction of the resistance spot welding information database specifically includes:

[0060] A distributed architecture is adopted, and a resistance spot welding information database is built through database management tools, including database management software (such as MySQL) and graphical visualization tools (such as DataGrip). Database tables are created and information can be added, deleted, modified and retrieved through data definition statements (DDL) and data manipulation statements (DML).

[0061] The distributed architecture database includes a workpiece attribute database, a welding parameter database, and a welding quality database, which store corresponding resistance spot welding related information. When creating database tables, string-type resistance spot welding related information (such as material type, data source, etc.) is stored using a character data format (such as VARCHAR), while numeric-type resistance spot welding related information (such as plate thickness, welding current, tensile strength, etc.) is stored using a numeric data format, including integer (INT) and floating-point (FLOAT) types.

[0062] The specific database table fields and data types are shown in Table 1:

[0063] Table 1

[0064]

[0065] The database tables with the above structure are created using DDL statements, and the collected resistance spot welding information is entered into the workpiece attribute database, welding parameter database, and welding quality database using DML statements, thereby achieving data classification, storage, and efficient management.

[0066] In step S103 above, training the BP neural network includes:

[0067] A backpropagation neural network model was designed, which includes an input layer, a hidden layer, and an output layer, as shown in Figure 2. The input layer neurons correspond to workpiece attribute information and welding parameter information, with a total of 5 input neurons, corresponding to material type (converted to numerical values ​​through encoding), plate thickness, welding current, welding time, and electrode pressure, respectively. The output layer neurons correspond to welding quality information, with a total of 2 output neurons, corresponding to weld nugget diameter and tensile strength, respectively.

[0068] The collected welding parameter and welding quality information is preprocessed as follows: First, the data is cleaned to remove outliers (such as parameter values ​​and quality index values ​​that are significantly outside the reasonable range); then, the data is normalized to convert data of different magnitudes to the same interval (such as [0,1]) to improve the training efficiency and convergence speed of the neural network; finally, the preprocessed resistance spot welding information is divided into training and testing sets in an 8:2 ratio. The training set is used to train the BP neural network, and the testing set is used to verify the generalization ability of the model.

[0069] The training set data is input into the backpropagation neural network model for training. The mean squared error function (MSE) is used as the loss function to evaluate the training effect of the neural network model. The formula for the MSE is:

[0070]

[0071] Where n is the number of samples, V i y is the model's predicted value. i These are actual measured values. The initial weights and biases of the neural network model are iteratively updated using the backpropagation algorithm. The loss function value is calculated after each iteration until it is less than a preset threshold (e.g., 0.001), indicating that the model meets the preset convergence condition. At this point, the trained BP neural network is obtained, resulting in the backpropagation neural network model. This model can establish a mapping relationship between workpiece attributes, welding parameters, and welding quality.

[0072] The trained BP neural network is validated using test set data. The prediction accuracy of the model is calculated. If the prediction accuracy meets the preset requirements (e.g., greater than 90%), the model is qualified. If the prediction accuracy does not meet the preset requirements, the parameters such as the number of hidden layer neurons and the learning rate of the neural network are adjusted, and the model is retrained until it is qualified.

[0073] In step S104 above, the construction of the expert system rule base includes:

[0074] Construct an expert system rule base that includes a basic rule base, a parameter constraint rule base, and a parameter optimization rule base:

[0075] Basic rule base: Stores the basic process principles and material-parameter matching benchmarks for resistance spot welding, such as "the welding current of resistance spot welding is positively correlated with the resistivity of the material, the greater the resistivity, the greater the welding current required" and "for low carbon steel materials, the welding time is positively correlated with the plate thickness, the greater the plate thickness, the longer the welding time", etc., providing a theoretical basis for the preliminary design of welding parameters.

[0076] Parameter constraint rule library: Stores the value range of resistance spot welding parameters corresponding to different materials and plate thicknesses. For example, "the welding current value range of high-strength steel materials is 8-12KA" and "the electrode pressure value range of aluminum alloy materials with a plate thickness of 1.0-2.0mm is 3-5KN". These rules are used to constrain the value of welding parameters and avoid welding quality problems caused by parameters exceeding the reasonable range.

[0077] Parameter optimization rule base: This base uses a trained BP neural network as the parameter optimization model to determine and optimize the initial welding parameter range. It includes model invocation rules and parameter optimization target rules, such as rules like "when there is no perfectly matching case, invoke the BP neural network model to optimize the initial welding parameter range" and "the goal of parameter optimization is to ensure the weld nugget diameter meets design requirements and maximizes tensile strength."

[0078] As shown in Figure 3, in step S105, the material and thickness information of the target workpiece are received, and matching welding parameters are retrieved, or the target welding parameters are obtained through reasoning and optimization using an expert system database. This is specifically implemented through the following workflow:

[0079] Input the material and thickness information of the target workpiece, and determine the welding parameters using the following methods:

[0080] Database retrieval: Input the material and thickness information of the target workpiece into the resistance spot welding information database and search for a completely matching case (i.e., a case with the same material type and thickness). If a completely matching case exists, the corresponding welding current, welding time, and electrode pressure information are directly extracted from the welding parameter database as the target welding parameters.

[0081] Inference optimization: If no perfect match is found, the parameter optimization rule base is invoked, and inference optimization is performed according to the following process:

[0082] a. Calculate the similarity between the target workpiece and cases in the database based on the K-nearest neighbor algorithm, as shown in Figure 4: First, calculate the material matching degree (if the materials are the same, the matching degree is 1; if the materials are different, the matching degree is determined according to the differences in material properties, such as 0.8, 0.6, etc.); then calculate the Euclidean distance of the plate thickness. The Euclidean distance formula is:

[0083]

[0084] in, t represents the thickness of the target workpiece. i Let the plate thickness be the value of the i-th case in the database.

[0085] Finally, the material matching degree and the Euclidean distance of the plate thickness are converted into a comprehensive similarity score. The conversion formula is as follows:

[0086]

[0087] Where S is the overall similarity, S m For material matching degree, α is the weighting coefficient (ranging from 0 to 1, which can be adjusted according to actual conditions, such as 0.5), and d max The preset maximum plate thickness Euclidean distance.

[0088] b. Based on the comprehensive similarity, select a preset number (e.g., 3) of similar cases, extract the welding parameters corresponding to these similar cases from the welding parameter database, and determine the initial welding parameter range.

[0089] c. Input the initial welding parameter range into the BP neural network in the parameter optimization rule base, and combine it with a genetic algorithm for multi-objective optimization: take meeting the design requirements for weld nugget diameter and maximizing tensile strength as the optimization objectives, and use the selection, crossover, and mutation operations of the genetic algorithm to iteratively optimize the initial welding parameters. After each iteration, use the BP neural network to predict the welding quality and determine whether the optimization objectives are met; if not, continue iterating; if met, stop iterating to obtain the Pareto optimal solution of the welding parameters, i.e., the target welding parameters.

[0090] Example 2: Based on the same inventive concept, this application also provides a resistance spot welding parameter determination system for implementing the above method. The solution provided by this system is similar to the solution described in the above embodiments. Therefore, the specific limitations in one or more system embodiments provided below can be found in the limitations of the method above, and will not be repeated here.

[0091] In one embodiment, the system provided by this invention, as shown in FIG5, includes: an information collection module 210, a database module 220, a mapping relationship acquisition module 230, a rule base construction module 240, and a target welding parameter determination module 250, wherein:

[0092] Information collection module 210 is used to collect resistance spot welding related information; the resistance spot welding related information includes: workpiece attribute information, welding parameter information and welding quality information;

[0093] Database module 220 is used to build a resistance spot welding information database using a distributed architecture, and to classify and store the collected resistance spot welding related information.

[0094] The mapping relationship acquisition module 230 is used to preprocess the welding parameter information and welding quality information, input them into the BP neural network for training, obtain the backpropagation neural network model, and determine the mapping relationship between workpiece attributes, welding parameters and welding quality.

[0095] The rule base construction module 240 is used to establish an expert system rule base based on the mapping relationship between the workpiece attributes, welding parameters and welding quality;

[0096] The target welding parameter determination module 250 is used to receive the material and plate thickness information of the target workpiece, retrieve matching welding parameters, or perform reasoning optimization through an expert system database to obtain the target welding parameters.

[0097] In the above embodiment, the various modules of the system work together. The information collection module provides data support for the database module, the database module provides the data retrieval basis for the target welding parameter determination module, the mapping relationship acquisition module and the rule base construction module provide technical support for the parameter determination module, and the target welding parameter determination module outputs the target welding parameters. The entire system has a reasonable structure, is easy to operate, and can efficiently and accurately determine the resistance spot welding parameters.

[0098] In one embodiment, an electronic device is provided, which may be a terminal, and its internal structure diagram may be as shown in Figure 6. The electronic device includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with an external terminal; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements the resistance spot welding parameter determination method described in any one of steps S101 to S105. The display screen of the electronic device may be a liquid crystal display (LCD) or an electronic ink display. The input device may be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the outer casing of the electronic device, or an external keyboard, touchpad, or mouse, etc.

[0099] Those skilled in the art will understand that the structure shown in Figure 6 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0100] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0101] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, 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, embedded processor, 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 specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0102] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0103] These computer program instructions may 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 specified in one or more flowcharts and / or one or more block diagrams.

[0104] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval.

Claims

1. A method for determining resistance spot welding parameters, characterized in that, The method includes: collecting resistance spot welding related information; the resistance spot welding related information includes: workpiece attribute information, welding parameter information, and welding quality information; constructing a resistance spot welding information database using a distributed architecture, and classifying and storing the collected resistance spot welding related information; after preprocessing the welding parameter information and welding quality information, inputting them into a BP neural network for training to obtain a backpropagation neural network model, and determining the mapping relationship between workpiece attributes, welding parameters, and welding quality; establishing an expert system rule base based on the mapping relationship between workpiece attributes, welding parameters, and welding quality; receiving the material and plate thickness information of the target workpiece, retrieving matching welding parameters, or performing reasoning optimization through the expert system database to obtain the target welding parameters.

2. The method according to claim 1, characterized in that, The workpiece attribute information includes material type and plate thickness information; The welding parameter information includes: electrode pressure, welding current, and welding time information; the welding quality information includes: tensile strength and weld nugget diameter information.

3. The method according to claim 2, characterized in that, The resistance spot welding related information also includes: supplementing the missing plate thickness combinations with information through welding experiments.

4. The method according to claim 1, characterized in that, The process of classifying and storing the collected resistance spot welding related information includes: storing workpiece attribute information, welding parameter information, and welding quality information in the workpiece attribute database, welding parameter database, and welding quality database included in the resistance spot welding information database, respectively; establishing database tables through data definition statements and data manipulation statements, and performing information addition and deletion operations on the database tables; storing string-type resistance spot welding related information in a character data format; and storing numeric-type resistance spot welding related information in a numeric data format; wherein, the numeric type includes integer and floating-point types.

5. The method according to claim 1, characterized in that, The process of preprocessing welding parameter information and welding quality information, and then inputting them into a backpropagation neural network for training to obtain a backpropagation neural network model includes: dividing the collected resistance spot welding related information into a training set and a test set according to a preset ratio; training the BP neural network using the training set; verifying the generalization ability of the BP neural network using the test set; evaluating the training results of the BP neural network using a loss function, iteratively updating the initial weights and biases of the BP neural network until the loss function meets a preset convergence condition, thereby obtaining the backpropagation neural network model; wherein, the input layer neurons of the backpropagation neural network model correspond to workpiece attribute information and welding parameter information, and the output layer neurons correspond to welding quality information.

6. The method according to claim 1, characterized in that, The expert system database includes: a basic rule base for storing the basic process principles and material-parameter matching benchmarks for resistance spot welding; a parameter constraint rule base for storing the range of resistance spot welding parameters for different materials and plate thicknesses; and a parameter optimization rule base that uses a trained BP neural network as the parameter optimization model to determine the initial welding parameter range and perform optimization.

7. The method according to claim 6, characterized in that, The process of receiving the material and thickness information of the target workpiece, retrieving matching welding parameters, or performing reasoning optimization through an expert system database to obtain the target welding parameters includes: retrieving the resistance spot welding information database based on the received material and thickness information of the target workpiece; if a completely matching case exists, the corresponding welding parameters are directly output; if no completely matching case exists, the similarity between the target workpiece and the case in the expert system rule base is calculated using the K-nearest neighbor algorithm, similar cases are selected to obtain the initial welding parameter range, and then a genetic algorithm is used to perform multi-objective optimization in the parameter optimization rule base to obtain the target welding parameters.

8. The method according to claim 7, characterized in that, The calculation of the similarity between the target workpiece and the cases in the expert system rule base using the K-nearest neighbor algorithm includes: calculating the Euclidean distance between the material matching degree and plate thickness of the target workpiece and the cases in the expert system rule base, converting the Euclidean distance into a similarity, and filtering a preset number of similar cases based on the similarity.

9. A resistance spot welding parameter determination system for implementing the method as described in any one of claims 1-8, characterized in that, include: The information collection module is used to collect information related to resistance spot welding; The resistance spot welding related information includes: workpiece attribute information, welding parameter information, and welding quality information; a database module, used to construct a resistance spot welding information database using a distributed architecture, and to classify and store the collected resistance spot welding related information; a mapping relationship acquisition module, used to preprocess the welding parameter information and welding quality information, input them into a BP neural network for training, obtain a backpropagation neural network model, and determine the mapping relationship between workpiece attributes, welding parameters, and welding quality; a rule base construction module, used to establish an expert system rule base based on the mapping relationship between workpiece attributes, welding parameters, and welding quality; and a target welding parameter determination module, used to receive the material and plate thickness information of the target workpiece, retrieve matching welding parameters, or perform reasoning optimization through the expert system database to obtain the target welding parameters.

10. An electronic device, characterized in that, The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1-8.

Images