Methods for evaluating dynamic reference resistance curves as well as resistance welding methods and resistance welding equipment

Abstract

The invention relates to a method for evaluating dynamic reference resistance curves (RRC) for resistance welding processes, wherein, based on at least one operating parameter of at least one resistance welding device and at least one material parameter of at least two workpieces to be welded, operating values ​​for at least one dynamic reference resistance curve (RRC) are specified for carrying out a corresponding welding process, and the at least one dynamic reference resistance curve (RRC) is generated, wherein at least one welding process is carried out with the at least one dynamic reference resistance curve (RRC) and the resistance welding device is controlled in such a way that the operating values ​​specified in the at least one reference resistance curve (RRC) are at least approximately reflected in at least one corresponding dynamic resistance curve (RK).wherein at least one dynamic reference resistance curve (RRK) is evaluated and stored based on at least one corresponding dynamic resistance curve (RK) and at least one quality criterion (QC), as well as a corresponding resistance welding process and a corresponding resistance welding device.

Description

[0001] The invention relates to a method for evaluating dynamic reference resistance curves for a resistance welding process. Furthermore, the invention relates to a resistance welding process for producing a welded joint between at least two workpieces and to a resistance welding device for carrying out such a resistance welding process.

[0002] Resistance welding processes and resistance welding devices for producing a welded joint between at least two workpieces are known in numerous variations in vehicle construction.

[0003] Resistance welding allows workpieces to be joined together using a material-bonded process. For example, in automated body-in-white production, robot-guided welding guns weld different workpieces, such as sheet metal, together using resistance welding. When commissioning a new system, a reference resistance curve is created for each welding program, which the welding control system attempts to replicate in an adaptive mode during ongoing production. A new welding program is created for each material thickness combination on the system. The quality criterion for a weld spot is the indentation determined by ultrasound, which provides information about the lens size. Furthermore, weld spatter can occur during the welding process. This involves molten weld metal spraying out of the weld area and adhering to the workpiece, the electrodes, or system components.

[0004] Welding spatter, in particular, causes significant costs. It can disrupt the process and contaminate system components, leading to increased maintenance times and costs. Sharp-edged material residues on components can damage cable harnesses during assembly. Therefore, welding spatter is often manually removed, which is time-consuming and expensive. Furthermore, different settings are frequently used for the same material thickness combination, resulting in varying process and weld quality.

[0005] A resistance welding process is known from DE 10 2020 204 667 A1. In this process, welding processes are carried out in which welding electrodes are pressed against a weld point on the workpieces and energized with a welding current according to predefined welding parameters. During these welding processes, at least one characteristic value that characterizes the weld quality is determined. After a number of welding processes have been carried out, a statistical analysis of the determined characteristic values ​​is performed, and depending on the result of this statistical analysis, it is determined whether the predefined welding parameters should be adjusted.

[0006] From EP 3 539 713 A1, a method and a device for quality evaluation in resistance welding are known. The quality evaluation method characterizes resistance-welded components by evaluating at least one welding parameter from the group consisting of: welding current, welding voltage, contact force, resistance, and relative displacement. A recorded profile of the welding parameter to be evaluated is provided as a curve of the respective welding parameter. The curve is evaluated by reading at least one characteristic value from the respective curve as a function of time and optionally also as a function of at least one other welding parameter. The characteristic value is compared with at least one target value for the characteristic value, where the target value is defined based on the recorded profile or gradient of the welding parameter to be evaluated.Based on a comparison of the characteristic values, particularly with regard to the amount of a deviation of the characteristic value from the target value, the quality of the welded component is evaluated.

[0007] German patent DE 10 2015 225 050 A1 discloses a method for creating a database for carrying out welding processes. Production data, which describes different welding processes for workpieces to be welded together at a specific welding point, each with a specific workpiece thickness and / or material combination, are classified according to identical workpiece thickness and / or material combinations. For identical workpiece thickness and / or material combinations from the production data, a combination-specific data set is created for each welding point specific to the respective workpiece thickness and / or material combination. Welding processes for the specific welding points can then be carried out according to the respective welding point-specific data.Depending on parameters from each combination-specific data set, weld point-specific data for the respective workpiece thickness and / or material combination are selected and stored in a database.

[0008] The invention is based on the objective of providing a method for evaluating dynamic reference resistance curves as well as resistance welding processes for producing a welded joint between at least two workpieces and a resistance welding device for carrying out such a resistance welding process, which increases the quality of the welded joint and reduces weld spatter and the rework caused by it.

[0009] This problem is solved by a method for evaluating dynamic reference resistance curves with the features of claim 1, by a resistance welding method with the features of claim 8, and by a resistance welding device with the features of claim 10. Advantageous embodiments with expedient further developments of the invention are specified in the dependent claims.

[0010] In order to provide a method for evaluating dynamic reference resistance curves for resistance welding processes, which increases the quality of the weld joint and reduces weld spatter and the resulting rework, operating values ​​for at least one dynamic reference resistance curve are specified based on at least one operating parameter of at least one resistance welding device and at least one material parameter of at least two workpieces to be welded, and the at least one dynamic reference resistance curve is generated.In this process, at least one welding operation is performed using at least one dynamic reference resistance curve, and the resistance welding device is controlled in such a way that the operating values ​​specified in the at least one reference resistance curve are at least approximately achieved in at least one corresponding dynamic resistance curve. The at least one dynamic reference resistance curve is evaluated and stored based on the at least one corresponding dynamic resistance curve and at least one quality criterion.

[0011] Furthermore, a resistance welding process is proposed for producing a welded joint between at least two workpieces using a resistance welding device. Dynamic reference resistance curves with operating values ​​for a welding process are provided, which are evaluated using this method. Depending on at least one operating parameter of the resistance welding device and at least one material parameter of the at least two workpieces to be welded, one of the evaluated dynamic reference resistance curves is selected. If several suitable dynamic reference resistance curves are available, the one with the highest rating is selected.The welding process is carried out using the selected dynamic reference resistance curve, and the resistance welding device is controlled in such a way that the operating values ​​specified in the selected reference resistance curve are at least approximately reflected in a corresponding current dynamic resistance curve.

[0012] Furthermore, a resistance welding device is proposed, which includes an evaluation and control unit and a welding gun and is designed to carry out such a resistance welding process.

[0013] Implementations of the present resistance spot welding process can preferably be used as a joining technology for the production of body-in-white in large automotive companies. During the welding process, two or more overlapping sheets can be clamped together by the electrodes of a welding gun, and an electric current is induced. The high electrical resistance between the sheets generates heat, leading to local melting and the formation of the weld, which can, for example, have a lens shape. Sufficient flow is required to form adequate coalescence. A compromise exists between achieving an optimal weld size or weld lens and avoiding spatter, which is a welding defect in which molten metal escapes from the weld pool.

[0014] When generating dynamic reference resistance curves, data sets from previous welding processes and dynamic process conditions can be considered, which can be acquired from multiple production lines at different manufacturing plants. The individual data sets are each taken from welding points on resistance welding equipment, which may differ in design, type of industrial robot, or type of welding gun. At least one material parameter of the at least two workpieces to be welded relates, for example, to a material-thickness combination of the at least two workpieces. Different material-thickness combinations are welded at the various production locations.For each recorded data point of a corresponding weld spot, an electrical current and voltage profile can be recorded over the welding time, for example, in 1.0 millisecond time increments. A timestamp and unique characteristics, such as information about the material thickness combination and material types of the at least two workpieces to be welded, the resistance welding equipment (e.g., type of robot and welding gun), a predicted penetration value, and information about the presence of weld spatter, can be recorded for each data point. Each recorded data point is linked to a dynamic reference resistance curve. Each combination of resistance welding equipment and material thickness combination can have a unique dynamic reference resistance curve.

[0015] In an advantageous embodiment of the method, the operating parameters of the resistance welding device can include an effective electrode area and / or an adjustable effective electrode force and / or an adjustable electrical voltage curve and / or an adjustable current curve. The current dynamic resistance curve can be determined by dividing the electrical voltage curve by the electrical current curve. The adjustable effective electrode force can correspond to a force with which a corresponding electrode is pressed against the corresponding workpiece in the area of ​​the weld. The at least one material parameter of the at least two workpieces to be welded can include a material thickness combination and / or a material combination.To improve the selection of the dynamic reference resistance curve, the existing dynamic reference resistance curves can be grouped according to the material thickness combination and according to identical process conditions, such as effective electrode area.

[0016] In a further advantageous embodiment of the method, at least one quality criterion can correspond to the presence of at least one weld spatter or a number of weld spatters and / or a welding time and / or a distance between the current dynamic resistance curve and the selected dynamic reference resistance curve and / or a penetration depth of a penetration. Three of the four quality criteria mentioned are based on the resistance signal. For example, the presence of a weld spatter can be detected by a characteristic sharp drop in the dynamic resistance curve. The ejected metal trapped between the sheets increases the effective contact area, while the mechanical collapse around the molten weld spot reduces the effective thickness of the material, leading to a drop in resistance. Here, the ejection or...The formation of weld spatter and the time of its emission can be detected, for example, by evaluating the dynamic resistance curve, preferably by the evaluation and control unit. A quality criterion could be, for example, the percentage of the dynamic resistance curve that welds against the corresponding dynamic reference resistance curve and leads to the emission or formation of weld spatter.

[0017] Welding time in serial production lines, especially high-speed production lines, is often limited by the required production rate. Therefore, it is not possible to simply compensate for lower current intensities by increasing the welding time. The welding time can be automatically extended in response to detected discharges or weld spatter, or if an internally calculated quality metric falls below a threshold. The latter correlates strongly with the energy input. Automatic extension of the welding time is problematic because it can lead to unpredictability and fluctuating production times. In embodiments of the resistance welding process according to the invention, the welding time can be represented as the length of the dynamic reference resistance curve. In this case, a shorter welding time can lead to a better overall evaluation.

[0018] The distance or deviation of the current dynamic resistance curve from the selected dynamic reference resistance curve can be used, for example, to draw conclusions about process stability. For instance, groups can first be generated in the datasets for each combination of material thickness and resistance welding device. For each group, the rate at which successive resistance values ​​in the set of dynamic resistance curves deviate from the corresponding dynamic reference resistance curve can be calculated, based on a predefined threshold. This threshold determines the percentage of welds that must deviate from the dynamic reference resistance curve for the difference to be considered significant. This threshold can, for example, be set to a value of 96%.The dynamic resistance curves are aligned to the length of the corresponding dynamic reference resistance curve. This alignment can be adjusted by truncation or extension to ensure all curves have the same length. Furthermore, the distance or deviation of the current dynamic resistance curve from the selected dynamic reference resistance curve—that is, how much the shape of the current dynamic resistance curve deviates from the selected dynamic reference resistance curve—can be calculated using methods such as dynamic time warping, Frechet distance, or wavelet transforms (time-frequency transformations). The distance or deviation of the current dynamic resistance curve from the selected dynamic reference resistance curve can also be measured using lower-dimensional representations in latent space.Using methods such as Time Series to Vec TS2Vec, PCA, and Autoencoder, one could train an anomaly detection model on the dynamic resistance curves and see how much the representation of the selected dynamic reference resistance curve deviates from the representation of the current dynamic resistance curve in latent space.

[0019] The penetration depth can be measured, for example, using a non-destructive testing method with ultrasonic pulse-echo and can be used to indirectly infer the size of the weld joint. The penetration depth is calculated as the difference between the initial thickness or total thickness of the workpieces to be welded and the remaining thickness or residual thickness of the welded workpieces at the weld joint, expressed as a percentage of the initial thickness. Measurements are only available for a subset of the welds produced daily. Therefore, the ultrasonic pulse-echo results were used to train a regression model and predict the penetration depth for all welds. This means that the regression model can predict the penetration depth for all welds. Such a regression model has a mean effective error (MAE) of 1.67%.This level of error is considered acceptable, and the penetration depth values ​​can be assumed to be correct. The datasets can contain welds with varying penetration depth requirements. For each requirement, upper and lower thresholds can be defined, and penalties can be assigned to weld points that fall outside these limits. The further a weld point deviates from the limit, the higher the penalty, resulting in a lower overall score.

[0020] In a further advantageous embodiment of the method, at least two of the listed quality criteria can be evaluated when assessing the dynamic reference resistance curves. These two quality parameters can be weighted and combined to form an overall quality assessment. For example, the quality criteria welding time and penetration depth can be standardized to values ​​between 0.0 and 1.0 using a min-max scaling approach. A weight of 1.0 can be assigned to the penetration depth criterion. A weight of 2.0 can be assigned to both the deviation criterion and the ejection or weld spatter criterion. A weight of 0.01 can be assigned to the welding time criterion to compensate for the influence of welding time relative to the other criteria. Subsequently, a weighted average value can be calculated for each dynamic reference resistance curve.Based on the weighted average, a standardized, weighted sum of the factors involved can then be calculated, ensuring that the contribution of each individual factor is scaled according to its relative importance, which is determined by the weights. This means that dynamic reference resistance curves with a low rating may exhibit a high ejection rate, a high deviation from the dynamic resistance curves, long welding times, and penetration penalties. Conversely, dynamic reference resistance curves with a high rating may exhibit no ejection, a low deviation from the dynamic resistance curves, short welding times, and no penetration penalties.

[0021] In a further advantageous embodiment of the method, one of the dynamic reference resistance curves, which leads to at least one weld spatter in the corresponding dynamic resistance curve, can be replaced by a newly generated and evaluated dynamic reference resistance curve, whose corresponding current curve has lower current values ​​than the corresponding current curve of the original dynamic reference resistance curve or a current pause in a time range in which the at least one weld spatter occurs.

[0022] In an advantageous embodiment of the resistance welding process, the generated and evaluated dynamic reference resistance curves can be stored in a database and individually transmitted to the corresponding resistance welding device via a communication link. For this purpose, the data of the evaluated dynamic reference resistance curves can be analyzed in the central database, transformed into a format suitable for welding control, and made available to the resistance welding device via edge technologies. This allows the new parameters to be accepted by a user upon manual input and thus transferred to the evaluation and control unit. Alternatively, the new parameters can be transferred directly to the evaluation and control unit without manual confirmation by the user.

[0023] The features and combinations of features mentioned above in the description, as well as those subsequently mentioned in the figure description and / or shown in the figures alone, can be used not only in the combinations specified, but also in other combinations or on their own, without departing from the scope of the invention. Thus, embodiments that are not explicitly shown or explained in the figures, but which can be derived and generated from the explained embodiments by separate combinations of features, are also to be considered as encompassed and disclosed by the invention.

[0024] Exemplary embodiments of the invention are shown in the drawings and are explained in more detail in the following description. In the drawings, identical reference numerals denote components or elements that perform the same or analogous functions. Here, the following are shown: Fig. 1 a schematic flowchart of an embodiment of a method according to the invention for evaluating dynamic reference resistance curves; Fig. 2 a schematic flowchart of an embodiment of a resistance welding process according to the invention for producing a welded joint between at least two workpieces, which is made with the process according to the invention from Fig. 1 weighted dynamic reference resistance curves used; Fig. 3 a schematic representation of an embodiment of a resistance welding device according to the invention, which is configured to perform the resistance welding process according to the invention. Fig. 2 to be carried out; Fig. 4 a detailed representation IV from Fig. 3; Fig. 5 a schematic representation of a welding electrode for the resistance welding device according to the invention made of Fig. 3 and Fig. 4; Fig. 6 a schematic characteristic curve diagram with a dynamic reference resistance curve, a corresponding current curve and a corresponding dynamic resistance curve; Fig. 7 a schematic characteristic curve diagram with a dynamic reference resistance curve, a corresponding current curve and a corresponding dynamic resistance curve; and Fig. 8 a schematic characteristic curve diagram with a dynamic reference resistance curve, a corresponding current curve and a corresponding dynamic resistance curve.

[0025] As from Fig. As can be seen in 1, the illustrated embodiment of a method 100 according to the invention for evaluating dynamic in Fig. Reference resistance curves RRK for resistance welding processes 200 shown in Figures 6 to 8, one embodiment of which is in Fig. The figure shown is a step S100, in which, based on at least one operating parameter, at least one in Fig. The resistance welding device 10, as depicted in Figures 3 to 5, and at least one material parameter of at least two workpieces to be welded, are used to define operating values ​​for at least one dynamic reference resistance curve RRK for carrying out a corresponding welding process. In step S110, the at least one dynamic reference resistance curve RRK is generated. In step S120, at least one welding process is carried out using the at least one dynamic reference resistance curve RRK, and the resistance welding device 10 is controlled such that the operating values ​​specified in the at least one reference resistance curve RRK are at least approximately achieved in at least one corresponding dynamic resistance curve RK.In step S130, the at least one dynamic reference resistance curve RRK is evaluated based on the at least one corresponding dynamic resistance curve RK and at least one quality criterion QK and stored in step S140.

[0026] As from Fig. As can be seen further in Figure 2, the illustrated embodiment of the resistance welding process 200 according to the invention for producing a welded joint 3 between at least two workpieces 1 with a resistance welding device 10 comprises a step S100 in which dynamic reference resistance curves RRK with operating values ​​for a welding process are provided, which are compared with the one shown in Figure 2. Fig. The procedures described in Figure 1 were evaluated. In step S210, one of the evaluated dynamic reference resistance curves RRK is selected based on at least one operating parameter of the resistance welding device 10 and at least one material parameter of the at least two workpieces 1 to be welded. If several suitable dynamic reference resistance curves RRK are available, the one with the highest rating is selected in step S210. In step S220, the welding process is carried out with the selected dynamic reference resistance curve RRK, and the resistance welding device 10 is controlled such that the operating values ​​specified in the selected reference resistance curve RRK are at least approximately reflected in a corresponding current dynamic resistance curve RK.

[0027] As from Fig. As can be seen further in Figures 3 to 5, the illustrated embodiment of the resistance welding device 10 according to the invention comprises an evaluation and control unit 12 and a welding gun 14. The resistance welding device 10 is designed to carry out the resistance welding process 200 according to the invention.

[0028] The dynamic reference resistance curves (RRK) generated and evaluated by method 100 are presented in a Fig. The data shown in the database 20 are stored and individually transmitted via a communication link 22 to the corresponding resistance welding device 10 or to the evaluation and control unit 12 of the resistance welding device 10.

[0029] During the welding process, the electrodes 16 are pressed against an adjustable force F.

[0030] As from Fig. As can be seen in Figures 3 to 5, the welding gun 14 comprises two arms 15, each connected to an electrode 16. The electrodes 16 can be pressed against the corresponding workpiece 1 in the area of ​​the weld point with an effective electrode force F, which is adjustable by the evaluation and control unit 12. A first arm 15A of the welding gun 14 is connected to a first electrode 16A, which is pressed against a first, here upper, workpiece 1A during the welding process with the electrode force F. A second arm 15B of the welding gun 14 is connected to a second electrode 16B, which is pressed against a second, here lower, workpiece 1B during the welding process with the electrode force F.During the welding process, the evaluation and control unit 12 uses the electrodes 16 to set an electrical voltage curve and an electrical current curve in order to generate a corresponding weld connection 3 at the welding point.

[0031] As especially from Fig. As can be seen further in Figure 5, each electrode 16 has a cylindrical base body with a first diameter D1 and a rounded tip with a second diameter D2 and a first radius R1, which forms an effective electrode surface 18. The tip tapers from the first diameter D1 of the base body to the second diameter D2. In the illustrated embodiment, the taper is at an angle B of approximately 60°. Furthermore, a second radius R2, which is significantly smaller than the first radius R1, is formed between the first radius R1 and the slope of the taper.

[0032] In the illustrated embodiments, the at least one operating parameter of the at least one resistance welding device 10 corresponds to the effective electrode area 18 and / or the effective electrode force F and / or the adjustable electrical voltage curve and / or the adjustable current curve. Furthermore, the at least one dynamic resistance curve RK is determined by dividing the corresponding electrical voltage curve by the corresponding electrical current curve. The at least one material parameter of the at least two workpieces 1 to be welded corresponds, in the illustrated embodiments, to a material thickness combination and / or a material combination.

[0033] In the illustrated embodiments, at least one quality criterion (QC) corresponds to the presence of at least one in Fig. 6 to 8 depicted weld spatters SSp or a number of weld spatters SSp and / or one in Fig. 6 to 8 shown welding time Ts and / or one in Fig. 6 to 8 shown distance dA of the at least one dynamic reference resistance curve RKK to the corresponding at least one dynamic resistance curve RK and / or one in Fig. 4. The penetration depth d of a penetration is shown in Figure 5. In this case, at least two quality criteria QC are evaluated when assessing the at least one dynamic reference resistance curve RRK. Furthermore, the at least two quality parameters QC are weighted and combined to form an overall quality assessment.

[0034] As especially from Fig. As can be seen further in Figure 4, the first electrode 16A creates a first penetration 5A with a first penetration depth d1 in the first workpiece 1A, and the second electrode 16B creates a second penetration 5B with a second penetration depth d2 in the second workpiece 1B, wherein the two penetration depths d1 and d2 are identical in the illustrated embodiment. As can be seen from Figure 4, the first electrode 16A creates a first penetration 5A with a first penetration depth d2 in the second workpiece 1B, where the two penetration depths d1 and d2 are identical in the illustrated embodiment. Fig. As can be seen further in Figure 4, the stack of the two workpieces to be welded 1 has a total thickness t1 outside the weld joint 3 and a residual thickness t2 in the area of ​​the weld joint.

[0035] As from Fig. 7 and Fig. As can be seen further in Figure 8, the dynamic reference resistance curve RRK shown, which leads to at least one weld spatter (SSp) in the corresponding dynamic resistance curve RK, can be replaced by a newly generated and evaluated dynamic reference resistance curve.

[0036] As from Fig. As can be seen further in Figure 7, a corresponding first current curve SK1 shown for the new, not shown, dynamic reference resistance curve exhibits lower current values ​​in a time range in which at least one weld spatter SSp occurs than the corresponding curve in Figure 7. Fig. Figure 6 shows the current curve SK of the original dynamic reference resistance curve RRK.

[0037] As from Fig. As can be seen further in Figure 8, a corresponding second current curve SK2 shown for the new, not shown, dynamic reference resistance curve has a current pause in a time range in which at least one weld spatter SSp occurs.

[0038] By lowering the current values ​​of the current curve or by suspending the current values ​​of the current curve in the time range in which at least one weld spatter SSp occurs, its formation can be effectively prevented. REFERENCE MARK LIST 1, 1A, 1B Workpiece 3 welded joints 5, 5A, 5B Penetration t1 Total thickness t2 residual thickness QK quality criterion d, d1, d2 Penetration depth Ts welding time dA distance SSp weld spatter 10 Resistance welding device 12 Evaluation and control unit 14 Welding pliers 16, 16A, 16B Welding electrode 18, 18A, 18B effective electrode area D1, D2 diameter R1, R2 Radius B angle RRK dynamic reference resistance curve RK dynamic resistance curve SK1, SK2, SK3 Flow curve 100 methods for evaluating dynamic reference resistance curves for resistance welding processes S100 to S140 process step 200 resistance welding processes S200 to S220 process step QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2020 204 667 A1

[0005] EP 3 539 713 A1

[0006] DE 10 2015 225 050 A1

[0007]

Images