A method, system, device, and medium for determining welding parameters for a narrow lap welder

Abstract

The application discloses a welding parameter determination method, system, device and medium of a narrow lap welding machine, relates to the technical field of cold-rolled strip steel welding control, and the method comprises the following steps: obtaining a plurality of historical welding parameters generated in a historical welding process of the narrow lap welding machine and corresponding weld quality evaluation results of the historical welding parameters; based on the historical welding parameters and the historical quality evaluation results, a welding parameter sample library is constructed, and an excellent welding sample set meeting a preset condition is screened from the welding parameter sample library according to the weld quality evaluation results, wherein the excellent welding sample set comprises strip steel specification parameters associated with the screened historical welding parameters; the current specification parameters of the current strip steel to be welded are obtained, and the current specification parameters are matched with the strip steel specification parameters of the excellent welding sample set, so as to determine the target welding parameters corresponding to the matched strip steel specification parameters.

Description

Technical Field

[0001] This application relates to the field of welding control technology for cold-rolled strip steel, and in particular to a method, system, equipment and medium for determining welding parameters of a narrow lap welding machine. Background Technology

[0002] Currently, narrow-lap resistance welding machines are widely installed at the entrances of post-processing production lines such as cold rolling continuous annealing and hot-dip galvanizing. These machines are used to weld the tail of the previous coil of strip to the head of the next coil, ensuring continuous and stable operation of the production line. The quality of the weld directly affects the stability of the finished product, the yield rate, and production costs.

[0003] However, in current production, the setting of welding parameters mainly relies on the experience of operators or consulting process manuals. For combinations of strip steel with different thicknesses, materials, and carbon equivalents, a large number of offline welding tests need to be conducted manually to explore suitable parameters. The testing process is not only time-consuming and labor-intensive, but also poses certain safety risks. Summary of the Invention

[0004] The summary section introduces a series of simplified concepts, which will be further explained in detail in the detailed description section. This summary section is not intended to limit the key and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0005] In a first aspect, embodiments of this application provide a method for determining welding parameters of a narrow lap welding machine, the method comprising: Acquire multiple historical welding parameters generated during the historical welding process of the narrow lap welding machine, as well as the weld quality evaluation results corresponding to the historical welding parameters; Based on the historical welding parameters and the historical quality evaluation results, a welding parameter sample library is constructed, and an excellent welding sample set that meets the preset conditions is selected from the welding parameter sample library. The excellent welding sample set includes strip steel specification parameters associated with the selected historical welding parameters. Obtain the current specification parameters of the strip steel to be welded, and match the current specification parameters with the strip steel specification parameters of the excellent welding sample set to determine the target welding parameters corresponding to the matched strip steel specification parameters.

[0006] In one embodiment of the present invention, constructing a welding parameter sample library based on the historical welding parameters and the historical quality evaluation results includes: The historical welding parameters are subjected to outlier removal and normalization to generate standardized welding parameters; The standardized welding parameters, the weld quality evaluation results, and the steel coil identification codes corresponding to the standardized welding parameters are associated to obtain structured sample records. An index is created on the structured sample records based on the strip steel specification parameters corresponding to the standardized welding parameters, in order to construct the welding parameter sample library.

[0007] In one embodiment of the present invention, the step of selecting a set of excellent welding samples whose weld quality evaluation results meet preset conditions from the welding parameter sample library includes: Historical welding parameters whose weld quality evaluation results are qualified are selected from the welding parameter sample library; The historical welding parameters are grouped according to the corresponding strip steel specifications, and the standard deviation of each historical welding parameter in each group is calculated. Historical welding parameters whose standard deviation is less than a preset standard deviation threshold within each group are used as the excellent welding sample set.

[0008] In one embodiment of the present invention, after matching the strip steel specification parameters from the set of excellent welding samples based on the current specification parameters to determine the target welding parameters, the process includes: The narrow lap welding machine is controlled to perform welding operations according to the target welding parameters, and the actual welding parameters of the current welding process are obtained. The quality of the weld is determined by using a multi-model fusion algorithm based on the actual welding parameters, and the quality level of the current weld is obtained.

[0009] In one embodiment of the present invention, the step of determining the weld quality based on the actual welding parameters using a multi-model fusion algorithm to obtain the current weld quality level includes: The actual welding parameters are input into the decision tree model, the K-nearest neighbor model, the dynamic time warping model, the deep neural network model, and the XGBoost model, respectively, and the quality judgment sub-results output by each model are obtained. The multiple quality judgment sub-results are weighted and fused by voting to generate a fused judgment result; The quality grade of the current weld is determined based on the fusion determination result, and the quality grade includes one of excellent, good, qualified and poor.

[0010] In one embodiment of the present invention, after determining the quality level of the current weld based on the fusion determination result, the process includes: When the quality level is poor, the defect cause analysis results are determined based on the deviation between the actual welding parameters and the target welding parameters. The defect cause analysis results include at least one of abnormal current, insufficient pressure, speed fluctuation, lap deviation, and temperature exceeding the limit.

[0011] In one embodiment of the present invention, after controlling the narrow lap welding machine to perform welding operations according to the target welding parameters, the process includes: The target welding parameters are input into a preset target temperature prediction model to obtain a predicted weld temperature curve. The target temperature prediction model is obtained by training the network to be trained using historical welding parameters. Obtain the actual weld temperature curve when the narrow lap welding machine is performing welding operation, and determine the temperature difference between the actual weld temperature curve and the predicted weld temperature curve. If the temperature difference value is greater than a preset temperature difference threshold, at least one parameter in the target welding parameters is adjusted so that the temperature difference value is less than or equal to the preset temperature difference threshold, and the adjusted parameter is used for the next matching.

[0012] Secondly, this application proposes a welding parameter determination system for a narrow lap welding machine, the system comprising: a data acquisition module, a sample library construction module, and a data matching module; The data acquisition module is configured to acquire multiple historical welding parameters generated during the historical welding process of the narrow lap welding machine, as well as the weld quality evaluation results corresponding to the historical welding parameters. The sample library construction module is configured to: construct a welding parameter sample library based on the historical welding parameters and the historical quality evaluation results, and select an excellent welding sample set from the welding parameter sample library whose weld quality evaluation results meet preset conditions. The excellent welding sample set includes strip steel specification parameters associated with the selected historical welding parameters. The data matching module is configured to: obtain the current specification parameters of the strip steel to be welded, and match the current specification parameters with the strip steel specification parameters of the excellent welding sample set, so as to determine the target welding parameters corresponding to the matched strip steel specification parameters.

[0013] Thirdly, an electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program stored in the memory to implement the steps of a method for determining welding parameters of a narrow lap welding machine as described in any of the first aspects above.

[0014] Fourthly, this application also proposes a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the steps of a method for determining welding parameters of a narrow lap welding machine according to any one of the first aspects.

[0015] In summary, the welding parameter determination method for a narrow lap welding machine according to this application embodiment constructs a welding parameter sample library by acquiring historical welding parameters and corresponding weld quality evaluation results generated from historical welding processes. It then selects a set of excellent welding samples that meet the quality conditions, allowing the extraction of successful welding experience implicit in historical data and overcoming the shortcomings of traditional methods that rely on manual memory and experience review. Based on this, once the specification parameters of the strip steel to be welded are obtained, the corresponding strip steel specification parameters are directly matched from the set of excellent welding samples, thereby quickly determining the target welding parameters. This eliminates the need for extensive offline welding tests for each new specification strip steel, significantly reducing testing workload and safety risks.

[0016] The method for determining welding parameters of the narrow lap welding machine proposed in this application, and other advantages, objectives and features of this application will be partly apparent from the following description, and partly understood by those skilled in the art through study and practice of this application. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit this specification. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A flowchart illustrating a method for determining welding parameters of a narrow lap welding machine, provided in an embodiment of this application; Figure 2 A schematic diagram of a welding parameter determination system for a narrow lap welding machine provided in this application embodiment; Figure 3 This is a schematic diagram of an electronic device for determining welding parameters of a narrow lap welding machine, provided in an embodiment of this application. Detailed Implementation

[0018] To better understand the technical solutions provided in the embodiments of this specification, the technical solutions of the embodiments of this specification will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of this specification and the specific features in the embodiments are detailed descriptions of the technical solutions of the embodiments of this specification, rather than limitations on the technical solutions of this specification. In the absence of conflict, the embodiments of this specification and the technical features in the embodiments can be combined with each other.

[0019] In this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, without necessarily requiring or implying any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The term "two or more" includes two or more cases.

[0020] Please see Figure 1 This is a flowchart illustrating a method for determining welding parameters of a narrow lap welding machine according to an embodiment of this application, specifically including: S110. Obtain multiple historical welding parameters generated during the historical welding process of the narrow lap welding machine and the weld quality evaluation results corresponding to the historical welding parameters. For example, the system first acquires multiple historical welding parameters generated during the narrow lap welding process, along with the weld quality evaluation results for each weld. These historical welding parameters include key process variables such as welding current, welding wheel pressure, welding speed, lap length, compensation amount, front and rear strip thickness and carbon equivalent, and weld temperature. These parameters reflect the actual operating status of the welding machine under different conditions. The weld quality evaluation results are derived from manual cupping tests, quantifying the welding quality of each weld according to preset grading standards (e.g., 0 for poor, 1 for good, and 2 for excellent). By simultaneously collecting these two types of data, a raw data foundation with both input features and output labels is provided for the subsequent construction of a welding parameter sample library. This allows the system to learn the intrinsic correlation between welding quality and process parameters from historical successes and failures.

[0021] S120. Based on the historical welding parameters and the historical quality evaluation results, a welding parameter sample library is constructed, and an excellent welding sample set that meets the preset conditions is selected from the welding parameter sample library. The excellent welding sample set includes strip steel specification parameters associated with the selected historical welding parameters. For example, based on the historical welding parameters and the historical quality evaluation results, the collected historical welding parameters are first cleaned and standardized to remove outliers and unify the dimensions. Then, the standardized welding parameters, corresponding weld quality evaluation results, and key information such as steel coil identification codes are associated to form structured sample records. An index is then created according to the strip steel specifications, thereby constructing a welding parameter sample library that can be quickly retrieved and iteratively updated. Based on this welding parameter sample library, qualified candidate samples are extracted based on the weld quality evaluation results meeting preset conditions. The candidate samples are then grouped according to the strip steel specifications, and the statistical consistency index (such as standard deviation) of each welding parameter in each group is calculated. Samples with small parameter fluctuations and high consistency are retained, ultimately forming an excellent welding sample set. This excellent welding sample set not only contains the selected excellent welding parameters but also associates them with the corresponding strip steel specifications, thus providing a reliable data foundation for subsequent matching of target welding parameters according to specifications.

[0022] S130. Obtain the current specification parameters of the strip steel to be welded, and match the current specification parameters with the strip steel specification parameters of the excellent welding sample set to determine the target welding parameters corresponding to the matched strip steel specification parameters.

[0023] For example, after constructing the excellent welding sample set, the system obtains the current specifications of the strip to be welded, specifically including key attributes such as the thickness and carbon equivalent of the preceding and following strips. Subsequently, the system compares and matches the current specifications with the strip specifications corresponding to each sample stored in the excellent welding sample set. During the matching process, the system can determine the closest sample entry based on the similarity or complete consistency of the specifications. Once a strip specification parameter that matches or is closest to the current specifications is found, the system extracts the associated historical welding parameters from that sample entry and uses these parameters as the target welding parameters to be used in the current welding operation. The entire matching process relies on a pre-established index structure in the welding parameter sample library, enabling efficient searching and matching. This ensures that a set of historically validated and reliable welding parameter combinations is automatically obtained without manual intervention, providing an accurate basis for setting subsequent welding operations.

[0024] In summary, the welding parameter determination method for narrow lap welding machines proposed in this application constructs a welding parameter sample library by acquiring historical welding parameters and corresponding weld quality evaluation results generated from historical welding processes. It then selects a set of excellent welding samples that meet the quality requirements, allowing the extraction of successful welding experience implicit in historical data and overcoming the shortcomings of traditional methods that rely on manual memory and experience review. Based on this, once the specifications of the strip steel to be welded are obtained, the corresponding strip steel specifications are directly matched from the set of excellent welding samples, thereby quickly determining the target welding parameters. This eliminates the need for extensive offline welding tests for each new specification of strip steel, significantly reducing testing workload and safety risks.

[0025] In some examples, the construction of a welding parameter sample library based on the historical welding parameters and the historical quality evaluation results includes: The historical welding parameters are subjected to outlier removal and normalization to generate standardized welding parameters; The standardized welding parameters, the weld quality evaluation results, and the steel coil identification codes corresponding to the standardized welding parameters are associated to obtain structured sample records. An index is created on the structured sample records based on the strip steel specification parameters corresponding to the standardized welding parameters, in order to construct the welding parameter sample library.

[0026] For example, outlier removal and normalization are performed on the historical welding parameters to generate standardized welding parameters. Specifically, outlier removal is based on preset outlier judgment criteria. For example, records exceeding the physical upper or lower limits of each welding parameter, as well as sampling points deviating from the mean by more than three times the standard deviation, are identified as outliers and removed to avoid data contamination caused by sensor failure, communication interference, or manual input errors affecting the accuracy of subsequent analysis. Normalization is performed by using preset normalization intervals (e.g., linear mapping to the [0,1] interval) for welding parameters of different dimensions (e.g., current in kA, pressure in kN, and speed in m / min) to unify the scale, eliminating the interference of dimensional differences on statistical calculations and model training, thereby generating comparable standardized welding parameters. On this basis, the standardized welding parameters, the weld quality evaluation results, and the steel coil identification codes corresponding to the standardized welding parameters are associated to obtain structured sample records. The weld quality evaluation results are derived from manual cupping tests and quantified according to preset levels (0-poor, 1-good, 2-very good). The steel coil identification code serves as a unique traceability identifier to associate the specific steel coil material information corresponding to each weld. By mapping the above three types of information one-to-one according to timestamps or welding event IDs and combining them into record entries with a fixed format, each structured sample record completely includes input features (standardized welding parameters), output labels (quality evaluation results), and material traceability information, thus providing standardized data units for the subsequent establishment of a searchable and analyzable sample library.

[0027] An index is built on the structured sample records based on the strip steel specifications corresponding to the standardized welding parameters to construct the welding parameter sample library. Specifically, the associated strip steel specifications are extracted from each structured sample record. These specifications include at least the preceding strip steel thickness, the following strip steel thickness, the preceding strip steel carbon equivalent, and the following strip steel carbon equivalent. These parameters characterize the material property boundary conditions of the current welding condition. Using the combination of these strip steel specifications as the index key, a fast search path from the specifications to the record storage location is established for each structured sample record according to a preset index structure (such as a B-tree or hash mapping). After the index is established, all structured sample records and their indexes together constitute the welding parameter sample library. This welding parameter sample library supports direct location of sample record sets with the same or similar specifications based on the input target specifications, thereby achieving efficient retrieval. At the same time, the sample library retains the original standardized welding parameters and quality evaluation results, allowing for further mining of excellent welding sample sets according to preset quality conditions (such as selecting quality evaluation results greater than or equal to good level) and statistical consistency conditions (such as standard deviation less than a preset threshold). By establishing an index, the welding parameter sample library not only achieves structured data storage, but more importantly, it provides the technical foundation for rapid matching by specification parameters, avoiding the inefficient operation of traversing the entire library.

[0028] In some examples, the step of selecting a set of excellent welding samples from the welding parameter sample library whose weld quality evaluation results meet preset conditions includes: Historical welding parameters whose weld quality evaluation results are qualified are selected from the welding parameter sample library; The historical welding parameters are grouped according to the corresponding strip steel specifications, and the standard deviation of each historical welding parameter in each group is calculated. Historical welding parameters whose standard deviation is less than a preset standard deviation threshold within each group are used as the excellent welding sample set.

[0029] For example, historical welding parameters with qualified weld quality evaluation results are selected from the welding parameter sample library. Specifically, in the constructed welding parameter sample library, each structured sample record is associated with a corresponding weld quality evaluation result. This evaluation result is derived from the artificial cupping test and is divided into three categories: "poor," "good," and "very good" according to preset levels. "Good" and "very good" are defined as qualified quality levels. The system traverses all sample records in the welding parameter sample library, reads the weld quality evaluation result stored in each record, and compares it with preset qualified quality conditions: if the evaluation result is "good" or "very good," the historical welding parameter corresponding to the sample is determined to meet the qualified quality conditions and is retained; if the evaluation result is "poor," the historical welding parameter corresponding to the sample is determined to not meet the qualified quality conditions and is removed. After the above selection operation, the system obtains a set of qualified historical welding parameters. Each historical welding parameter in this set has been verified by actual welding quality, eliminating the interference of unqualified welding samples on subsequent statistical analysis, thereby providing reliable input data for extracting highly consistent excellent samples.

[0030] The historical welding parameters are grouped according to their corresponding strip steel specifications. The standard deviation of each historical welding parameter within each group is calculated, and historical welding parameters whose standard deviation is less than a preset standard deviation threshold are designated as the excellent welding sample set. Specifically, from the selected set of qualified historical welding parameters, the system extracts the strip steel specifications associated with each historical welding parameter. These specifications include at least the preceding strip steel thickness, the following strip steel thickness, the preceding strip steel carbon equivalent, and the following strip steel carbon equivalent. The system uses combinations of these specifications as the grouping criterion, classifying historical welding parameters with identical combinations into the same group, thus forming multiple subsets separated by specifications. For each subset, the system selects key welding parameters (e.g., welding current, welding wheel pressure, welding speed) as statistical objects and calculates the standard deviation of all historical welding parameters within that subset on the key welding parameters. This standard deviation is used to quantify the dispersion of welding parameter values ​​under the same specifications. The system compares the calculated standard deviation with a preset standard deviation threshold, which is a pre-calibrated positive number used to define the acceptable upper limit of parameter consistency. If the standard deviation is less than the threshold, it indicates that the parameter values ​​of each weld within the group are concentrated and have small fluctuations, showing high consistency. If the standard deviation is greater than or equal to the threshold, it indicates that the parameters within the group are dispersed and have large fluctuations, showing poor consistency. The system retains only all historical welding parameters in the entire subset where the standard deviation is less than the preset threshold, extracts these historical welding parameters from the quality-compliant set, and finally forms the excellent welding sample set. Each historical welding parameter in this excellent welding sample set not only meets the welding quality requirements but also exhibits good reproducibility and stability under the same specifications, serving as a reliable benchmark for subsequent parameter matching.

[0031] In some examples, after matching the strip specification parameters from the set of excellent weld samples based on the current specification parameters to determine the target weld parameters, the process includes: The narrow lap welding machine is controlled to perform welding operations according to the target welding parameters, and the actual welding parameters of the current welding process are obtained. The quality of the weld is determined by using a multi-model fusion algorithm based on the actual welding parameters, and the quality level of the current weld is obtained.

[0032] For example, after matching the target welding parameters from the set of excellent welding samples based on the current specification parameters, the system controls the narrow lap welding machine to perform welding operations according to the target welding parameters. Specifically, the system writes the various set values ​​of the target welding parameters, including welding current, welding wheel pressure, welding speed, lap amount, compensation amount, and rolling wheel pressure, into the corresponding address units of the programmable logic controller of the narrow lap welding machine through an industrial communication protocol, driving the welding machine to perform narrow lap resistance welding according to the above set values. Simultaneously with the welding operation, the system collects the actual welding parameters generated during the current welding process in real time at a preset sampling frequency. The actual welding parameters include at least the actual welding current, actual welding wheel pressure, actual welding speed, actual weld temperature, and actual rolling wheel pressure. These actual welding parameters are directly obtained from sensor measurements or feedback signals in the welding machine control system, reflecting the actual dynamic response of the welding machine during the welding process rather than relying solely on the preset target values. The system synchronously stores the collected actual welding parameters according to timestamps and associates them with the strip steel specification parameters and target welding parameters corresponding to the current welding operation, providing input data under real working conditions for subsequent quality judgment steps. After obtaining the actual welding parameters of the current welding process, the system uses a multi-model fusion algorithm to determine the weld quality based on these parameters, thus obtaining the current weld quality level. Through this multi-model fusion algorithm, the system can integrate the advantages of different models, overcoming the low accuracy of single threshold or single model judgments, and achieving a reliable assessment of the current weld quality.

[0033] In some examples, the method of using a multi-model fusion algorithm to determine the weld quality based on the actual welding parameters and obtain the current weld quality level includes: The actual welding parameters are input into the decision tree model, the K-nearest neighbor model, the dynamic time warping model, the deep neural network model, and the XGBoost model, respectively, and the quality judgment sub-results output by each model are obtained. The multiple quality judgment sub-results are weighted and fused by voting to generate a fused judgment result; The quality grade of the current weld is determined based on the fusion determination result, and the quality grade includes one of excellent, good, qualified and poor.

[0034] For example, after the actual welding parameters are input into the decision tree model, the decision tree model performs layer-by-layer conditional judgments on various features of the actual welding parameters, such as welding current, welding wheel pressure, welding speed, overlap, compensation, rolling wheel pressure, and weld temperature, according to preset node splitting rules. Each internal node represents a comparison between a welding parameter and its corresponding threshold. Each path from the root node to the leaf node corresponds to a set of parameter value range combinations. The leaf node stores pre-labeled quality category labels. The decision tree model passes the input actual welding parameters down the tree structure, eventually landing on a certain leaf node, and outputs the quality category label stored in the leaf node as the first quality judgment sub-result. The first quality judgment sub-result includes one of excellent, good, qualified, or poor. When the actual welding parameters are input into the K-nearest neighbor model, the K-nearest neighbor model calculates the Euclidean distance or Manhattan distance between the feature vector of the current actual welding parameters and the feature vector of historical welding parameters in the welding parameter sample library, and selects the K nearest neighbor samples that are closest to all historical welding parameters; it counts the number of samples belonging to each quality level among these K nearest neighbor samples, and outputs the quality level that appears most frequently as the second quality judgment sub-result. When the actual welding parameters are input into the dynamic time warping model, the dynamic time warping model extracts the temperature curve of the weld temperature changing with time during the actual welding process as the sequence to be matched, and reads several temperature curves of historical welding parameters that are similar to the current strip steel specifications from the welding parameter sample library as reference sequences; the dynamic time warping model calculates the minimum cumulative distance between the sequence to be matched and each reference sequence through a dynamic programming algorithm. This distance can eliminate the nonlinear scaling effect of the time axis caused by the difference in welding speed, find the reference sequence with the most similar shape to the sequence to be matched, and output the quality level corresponding to the reference sequence as the third quality judgment sub-result.

[0035] When the actual welding parameters are input into a deep neural network model, the model consists of an input layer, multiple hidden layers, and an output layer. The input layer receives the feature vector of the actual welding parameters after normalization. Each hidden layer extracts deep abstract features layer by layer through linear transformation and nonlinear activation functions. The output layer uses the Softmax activation function to map the features of the last layer to a probability distribution of the current welding parameters belonging to four quality levels: excellent, good, acceptable, and poor. The quality level with the highest probability is output as the fourth quality judgment sub-result. When the actual welding parameters are input into an XGBoost (Extreme Gradient Boosting) model, the XGBoost model is based on a gradient boosting tree framework and is composed of multiple classification and regression trees. Each classification and regression tree outputs a leaf node score based on different feature dimensions of the actual welding parameters and their splitting thresholds. The XGBoost model adds the leaf node scores of all classification and regression trees and transforms them through a logistic function to obtain the probability value of the current actual welding parameters belonging to each quality level. The quality level with the highest probability value is selected as the fifth quality judgment sub-result. After obtaining the first, second, third, fourth, and fifth quality decision sub-results output by the decision tree model, K-nearest neighbor model, dynamic time warping model, deep neural network model, and XGBoost model, respectively, a weighted voting fusion is performed on the multiple quality decision sub-results: Each model is pre-assigned a weight coefficient, which is calibrated based on the accuracy of each model on the validation set; models with higher accuracy are assigned greater weight. During fusion, the quality levels output by each model are summed according to their weight values, and the total weight scores for the four levels (Excellent, Good, Acceptable, and Poor) are calculated. The quality level with the highest total weight score is selected as the fusion judgment result. Finally, the quality level of the current weld is determined based on the fusion judgment result, and this quality level is output as one of the four levels: Excellent, Good, Acceptable, and Poor.

[0036] By inputting actual welding parameters into five heterogeneous models—decision tree, K-nearest neighbor, dynamic time warping, deep neural network, and XGBoost—independent quality judgment sub-results are obtained from each model. These sub-results are then weighted and fused to generate the final fused judgment result. This approach leverages the strengths of different models: the decision tree model offers strong rule interpretability and fast computation; the K-nearest neighbor model utilizes historical experience based on similarity matching; the dynamic time warping model focuses on the temporal similarity analysis of weld temperature curve shapes; the deep neural network model excels at capturing nonlinear high-order interactions between welding parameters; and the XGBoost model possesses high generalization ability and anti-overfitting characteristics through gradient boosting ensemble learning. The weighted voting fusion mechanism further overcomes the risk of misjudgment that may occur with a single model under specific working conditions. This results in a weld quality grade that significantly outperforms traditional single-threshold or single-model judgment methods in both accuracy and recall, achieving more reliable automatic weld quality judgment and reducing missed detections of weld tears or false positives of qualified welds due to judgment errors.

[0037] In some examples, after determining the quality level of the current weld based on the fusion determination result, the process includes: When the quality level is poor, the defect cause analysis results are determined based on the deviation between the actual welding parameters and the target welding parameters. The defect cause analysis results include at least one of abnormal current, insufficient pressure, speed fluctuation, lap deviation, and temperature exceeding the limit.

[0038] For example, if the quality grade of the current weld is determined to be poor based on the fusion judgment result, the system first reads the actual welding parameters collected during the current welding operation and the pre-set target welding parameters. The actual welding parameters include at least the actual welding current, actual welding wheel pressure, actual welding speed, actual overlap, actual compensation, and actual weld temperature. The target welding parameters are set values ​​for controlling the welding machine operation, matched from a set of excellent welding samples. The system calculates the absolute value of the current deviation between the actual welding current and the target welding current, the absolute value of the pressure deviation between the actual welding wheel pressure and the target welding wheel pressure, the absolute value of the speed deviation between the actual welding speed and the target welding speed, and the absolute value of the overlap deviation between the actual overlap and the target overlap, and obtains the absolute value of the actual weld temperature.

[0039] For each of the above deviations, the system compares it with the corresponding pre-stored defect judgment thresholds: a preset current anomaly threshold defines the acceptable range of current deviation; a preset pressure deficiency threshold defines the lower limit tolerance of pressure below the target value; a preset speed fluctuation threshold defines the allowable fluctuation range of speed deviation; and a preset overlap deviation threshold defines the allowable error of overlap deviation. When the absolute value of the current deviation is greater than the preset current anomaly threshold, the system determines that there is a current anomaly defect; when the actual welding wheel pressure is less than the target welding wheel pressure and the absolute value of the pressure deviation is greater than the preset pressure deficiency threshold, the system determines that there is a pressure deficiency defect; when the absolute value of the speed deviation is greater than the preset speed fluctuation threshold, the system determines that there is a speed fluctuation defect; when the absolute value of the overlap deviation is greater than the preset overlap deviation threshold, the system determines that there is an overlap deviation defect; and when the actual weld temperature is greater than the preset upper temperature threshold, the system determines that there is a temperature over-limit defect.

[0040] After independently determining each defect, the system summarizes all identified defect causes and generates a defect cause analysis result. This result includes at least one or more combinations of abnormal current, insufficient pressure, speed fluctuation, lap deviation, and excessive temperature. If only one defect condition is met, the result contains only that single defect cause; if multiple conditions are met, the result includes all met defects in a set. If the quality grade of the weld is determined to be poor based on the fusion assessment result, various deviations are calculated based on the deviations between the actual and target welding parameters and compared with preset defect thresholds. This determines whether defects such as abnormal current, insufficient pressure, speed fluctuation, lap deviation, or excessive temperature exist. All found defects are then summarized to generate a structured defect cause analysis result. This allows the system to automatically determine weld quality as unqualified while providing specific root causes, overcoming the limitation of traditional quality assessment methods that only output a binary "qualified" or "unqualified" result without explaining the cause of failure.

[0041] In some examples, after controlling the narrow lap welding machine to perform welding operations according to the target welding parameters, the process includes: The target welding parameters are input into a preset target temperature prediction model to obtain a predicted weld temperature curve. The target temperature prediction model is obtained by training the network to be trained using historical welding parameters. Obtain the actual weld temperature curve when the narrow lap welding machine is performing welding operation, and determine the temperature difference between the actual weld temperature curve and the predicted weld temperature curve. If the temperature difference value is greater than a preset temperature difference threshold, at least one parameter in the target welding parameters is adjusted so that the temperature difference value is less than or equal to the preset temperature difference threshold, and the adjusted parameter is used for the next matching.

[0042] For example, after controlling the narrow lap welding machine to perform welding operations according to the target welding parameters, the system first inputs the target welding parameters used in this welding operation into a preset target temperature prediction model to obtain the predicted weld temperature curve. The target temperature prediction model is obtained by training the network to be trained using historical welding parameters. Specifically, the training process uses historical welding parameters collected during historical welding processes as input features and the actual temperature curve of the corresponding historical weld as a supervision label. The weight parameters of the network to be trained are iteratively updated by minimizing the mean square error or mean absolute error between the predicted temperature curve and the actual temperature curve until the model converges. The trained target temperature prediction model can, based on the input welding current, welding wheel pressure, welding speed, lap length, compensation amount, and the thickness and carbon equivalent of the preceding and following strips, pre-calculate the expected trajectory of the weld temperature change over time under the given parameter conditions, i.e., the predicted weld temperature curve. Meanwhile, during the welding operation, the system collects the actual weld temperature value output by the infrared temperature sensor of the narrow lap welding machine in real time at a preset sampling frequency, and records it in the order of timestamps to form the actual weld temperature curve.

[0043] After the welding operation is completed, the system aligns the actual weld temperature curve and the predicted weld temperature curve on the same time axis point by point, calculates the temperature difference between the two curves at the same time point, and takes the maximum, minimum, or arithmetic mean of the absolute values ​​of the temperature differences at all time points as the temperature difference value. This temperature difference value quantitatively reflects the degree of deviation between the actual thermal response and the ideal thermal response of the welding machine during the actual welding process: the smaller the temperature difference value, the closer the actual welding process is to the predicted ideal state; the larger the temperature difference value, the more likely there are interference factors not captured by the model or the target welding parameters themselves deviate from the optimal setting range for this specification of strip steel.

[0044] After obtaining the temperature difference value, the system compares it with a preset temperature difference threshold. The preset temperature difference threshold is a pre-calibrated positive number, typically determined based on the maximum allowable temperature difference corresponding to qualified welds in historical data, used to define the acceptable upper limit of deviation between the predicted and actual temperatures. If the temperature difference value is less than or equal to the preset temperature difference threshold, it indicates that the actual thermal response of this weld is basically consistent with the predicted thermal response, and the target welding parameters have good applicability to this specification of strip steel. The system does not adjust the parameters and stores the target welding parameters and their corresponding actual weld quality level as positive samples in the welding parameter sample library for subsequent iterative optimization. If the temperature difference value is greater than the preset temperature difference threshold, it indicates that a significant abnormal thermal response occurred during the actual welding process, and the target welding parameters failed to make the weld temperature change according to the expected trajectory. In this case, the system triggers the parameter adjustment process.

[0045] Specifically, the system determines the type and direction of parameter adjustment based on the deviation direction and degree of the actual weld temperature curve from the predicted weld temperature curve: if the actual weld temperature curve is generally higher than the predicted weld temperature curve, it indicates that the welding heat input is too large. The system reduces the welding current setting or welding wheel pressure setting in the target welding parameters according to a preset step size, or increases the welding speed setting according to a preset step size, to reduce the total heat input. If the actual weld temperature curve is generally lower than the predicted weld temperature curve, it indicates that the welding heat input is too small. The system increases the welding current setting or welding wheel pressure setting, or decreases the welding speed setting according to a preset step size, to increase the total heat input. After each parameter adjustment, the system re-inputs the adjusted parameters into the target temperature prediction model, recalculates the new predicted weld temperature curve, and compares the temperature difference with the actual weld temperature curve again. This adjustment process is repeated until the temperature difference is less than or equal to the preset temperature difference threshold. After adjustment, the system uses the adjusted welding parameters as the new target welding parameters and associates these new target welding parameters with the strip steel specification parameters corresponding to the current weld. This is used to match the system with the set of excellent welding samples the next time it encounters the same or similar strip steel specification parameters. Thus, this feedback adjustment mechanism allows the parameter combinations in the set of excellent welding samples to be corrected based on deviations from actual welding results, thereby gradually approaching the optimal welding parameters for that strip steel specification and achieving adaptive iteration of the sample library.

[0046] like Figure 2 As shown, this application proposes a welding parameter determination system for a narrow lap welding machine. The system includes: a data acquisition module 21, a sample library construction module 22, and a data matching module 23. The data acquisition module 21 is configured to acquire multiple historical welding parameters generated during the historical welding process of the narrow lap welding machine and the weld quality evaluation results corresponding to the historical welding parameters. The sample library construction module 22 is configured to: construct a welding parameter sample library based on the historical welding parameters and the historical quality evaluation results, and select an excellent welding sample set from the welding parameter sample library whose weld quality evaluation results meet preset conditions. The excellent welding sample set includes strip steel specification parameters associated with the selected historical welding parameters. The data matching module 23 is configured to: obtain the current specification parameters of the strip steel to be welded, and match the current specification parameters with the strip steel specification parameters of the excellent welding sample set to determine the target welding parameters corresponding to the matched strip steel specification parameters.

[0047] The effects of applying the aforementioned method in the above system can be found in the description of the aforementioned method embodiments, and will not be repeated here.

[0048] like Figure 3 As shown, this application embodiment also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, it implements the steps of any of the above-described methods for determining the welding parameters of the narrow lap welding machine.

[0049] Since the electronic device described in this embodiment is the device used to implement the welding parameter determination device of a narrow lap welding machine in the embodiments of this application, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the method described in the embodiments of this application. Therefore, how the electronic device implements the method in the embodiments of this application will not be described in detail here. Any device used by those skilled in the art to implement the method in the embodiments of this application is within the scope of protection of this application.

[0050] In practical implementation, when the computer program 311 is executed by the processor, it can achieve the following: Figure 1 Any of the corresponding implementation methods in the embodiments.

[0051] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0052] 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-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-readable program code.

[0053] 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 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 a machine for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0054] 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, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0055] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0056] This application also provides a computer program product, which includes computer software instructions that, when executed on a processing device, cause the processing device to execute the LDPC decoding method of a solid-state drive controller.

[0057] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0058] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0059] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0060] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0061] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0062] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0063] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

[0064] Although preferred embodiments have been described in this specification, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this specification.

[0065] Obviously, those skilled in the art can make various modifications and variations to this specification without departing from its spirit and scope. Therefore, if such modifications and variations fall within the scope of the claims and their equivalents, this specification is also intended to include such modifications and variations.

Claims

1. A method for determining welding parameters of a narrow lap welding machine, characterized in that, The method includes: Acquire multiple historical welding parameters generated during the historical welding process of the narrow lap welding machine, as well as the weld quality evaluation results corresponding to the historical welding parameters; Based on the historical welding parameters and the historical quality evaluation results, a welding parameter sample library is constructed, and an excellent welding sample set that meets the preset conditions is selected from the welding parameter sample library. The excellent welding sample set includes strip steel specification parameters associated with the selected historical welding parameters. Obtain the current specification parameters of the strip steel to be welded, and match the current specification parameters with the strip steel specification parameters of the excellent welding sample set to determine the target welding parameters corresponding to the matched strip steel specification parameters.

2. The method for determining welding parameters of a narrow lap welding machine according to claim 1, characterized in that, The construction of a welding parameter sample library based on the historical welding parameters and the historical quality evaluation results includes: The historical welding parameters are subjected to outlier removal and normalization to generate standardized welding parameters; The standardized welding parameters, the weld quality evaluation results, and the steel coil identification codes corresponding to the standardized welding parameters are associated to obtain structured sample records. An index is created on the structured sample records based on the strip steel specification parameters corresponding to the standardized welding parameters, in order to construct the welding parameter sample library.

3. The method for determining welding parameters of a narrow lap welding machine according to claim 1, characterized in that, The step of selecting a set of excellent welding samples whose weld quality evaluation results meet preset conditions from the welding parameter sample library includes: Historical welding parameters whose weld quality evaluation results are qualified are selected from the welding parameter sample library; The historical welding parameters are grouped according to the corresponding strip steel specifications, and the standard deviation of each historical welding parameter in each group is calculated. Historical welding parameters whose standard deviation is less than a preset standard deviation threshold within each group are used as the excellent welding sample set.

4. The method for determining welding parameters of a narrow lap welding machine according to claim 1, characterized in that, After determining the target welding parameters by matching the strip steel specification parameters from the set of excellent welding samples based on the current specification parameters, the process includes: The narrow lap welding machine is controlled to perform welding operations according to the target welding parameters, and the actual welding parameters of the current welding process are obtained. The quality of the weld is determined by using a multi-model fusion algorithm based on the actual welding parameters, and the quality level of the current weld is obtained.

5. The method for determining welding parameters of a narrow lap welding machine according to claim 4, characterized in that, The process of determining the weld quality based on the multi-model fusion algorithm using the actual welding parameters to obtain the current weld quality level includes: The actual welding parameters are input into the decision tree model, the K-nearest neighbor model, the dynamic time warping model, the deep neural network model, and the XGBoost model, respectively, and the quality judgment sub-results output by each model are obtained. The multiple quality judgment sub-results are weighted and fused by voting to generate a fused judgment result; The quality grade of the current weld is determined based on the fusion determination result, and the quality grade includes one of excellent, good, qualified and poor.

6. The method for determining welding parameters of a narrow lap welding machine according to claim 1, characterized in that, After determining the quality level of the current weld based on the fusion determination result, the process includes: When the quality level is poor, the defect cause analysis results are determined based on the deviation between the actual welding parameters and the target welding parameters. The defect cause analysis results include at least one of abnormal current, insufficient pressure, speed fluctuation, lap deviation, and temperature exceeding the limit.

7. The method for determining welding parameters of a narrow lap welding machine according to claim 4, characterized in that, After controlling the narrow lap welding machine to perform welding operations according to the target welding parameters, the process includes: The target welding parameters are input into a preset target temperature prediction model to obtain a predicted weld temperature curve. The target temperature prediction model is obtained by training the network to be trained using historical welding parameters. Obtain the actual weld temperature curve when the narrow lap welding machine is performing welding operation, and determine the temperature difference between the actual weld temperature curve and the predicted weld temperature curve. If the temperature difference value is greater than a preset temperature difference threshold, at least one parameter in the target welding parameters is adjusted so that the temperature difference value is less than or equal to the preset temperature difference threshold, and the adjusted parameter is used for the next matching.

8. A welding parameter determination system for a narrow lap welding machine, characterized in that, The system includes: a data acquisition module, a sample library construction module, and a data matching module; The data acquisition module is configured to acquire multiple historical welding parameters generated during the historical welding process of the narrow lap welding machine, as well as the weld quality evaluation results corresponding to the historical welding parameters. The sample library construction module is configured to: construct a welding parameter sample library based on the historical welding parameters and the historical quality evaluation results, and select an excellent welding sample set from the welding parameter sample library whose weld quality evaluation results meet preset conditions. The excellent welding sample set includes strip steel specification parameters associated with the selected historical welding parameters. The data matching module is configured to: obtain the current specification parameters of the strip steel to be welded, and match the current specification parameters with the strip steel specification parameters of the excellent welding sample set, so as to determine the target welding parameters corresponding to the matched strip steel specification parameters.

9. An electronic device, comprising: The memory and processor are characterized in that the processor, when executing a computer program stored in the memory, implements the steps of a method for determining welding parameters of a narrow lap welding machine as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for determining welding parameters of a narrow lap welding machine as described in any one of claims 1-7.

Images