Automobile radiator intelligent welding control method and system and intelligent welding production line

Abstract

The application is suitable for the field of intelligent manufacturing of automobile radiators, and particularly relates to an intelligent welding control method and system for automobile radiators and an intelligent welding production line. The intelligent welding control method comprises: continuously acquiring quantitative indicators reflecting the functional states of corresponding welding execution units through each state perception unit; constructing a hybrid benchmark based on the quantitative indicators of all welding execution units, and identifying the welding execution units with functional decay and the corresponding decay degrees based on the hybrid benchmark; determining the corresponding compensation strategy type based on the type of the welding position to which the welding execution unit with functional decay belongs; performing hierarchical compensation on the corresponding welding execution unit through the parameter adjustment unit based on the compensation strategy type and the decay degree; continuously monitoring the welding quality indicators of each welding point and iteratively adjusting until the welding quality indicators of all welding points return to the preset balanced range. The method improves the balanced consistency of the welding quality of all welding points.

Description

Technical Field

[0001] This application belongs to the field of intelligent manufacturing technology for automotive radiators, and in particular relates to an intelligent welding control method, system and intelligent welding production line for automotive radiators. Background Technology

[0002] The automotive radiator is a core component of the engine cooling system, typically composed of heat pipes, fins, and upper and lower manifolds connected by welding. Taking a mainstream all-aluminum radiator as an example, a single radiator usually contains dozens to hundreds of weld points, with resistance spot welding being a typical process where multiple welding units complete the welding of all weld points one by one. During resistance spot welding, the welding units operate in a high-temperature, high-pressure, and high-current environment for extended periods. The electrode faces undergo plastic deformation due to repeated pressure, increasing the contact area and causing a decrease in current density. Local alloying occurs between the aluminum and copper electrodes at high temperatures, forming brittle phases and a contamination layer on the electrode surface, increasing contact resistance. Scale buildup in the cooling water channels reduces electrode heat dissipation efficiency and accelerates electrode softening. Mechanical misalignment leads to uneven loading of the upper and lower electrodes, causing localized overheating. These degradations result in a non-uniform distribution of welding energy between different weld points.

[0003] Traditional welding control systems use uniform preset welding parameters for all weld points, implicitly assuming that the performance of all welding actuators remains constant throughout their service life. However, in reality, electrode performance continuously degrades with the accumulation of weld points, and the degradation rate varies at different locations due to differences in heat dissipation conditions and stress states. When individual welding actuators degrade, their corresponding weld points may experience incomplete welds or over-welds. Traditional systems cannot detect these individual differences in weld point condition and provide targeted compensation, leading to poor weld quality consistency, high product leakage rates, and increased scrap rates. Summary of the Invention

[0004] This application provides an intelligent welding control method, system, and intelligent welding production line for automotive radiators, which can improve the problem that traditional welding cannot detect the individual performance degradation of each welding execution unit during continuous operation.

[0005] In a first aspect, embodiments of this application provide an intelligent welding control method for automotive radiators, used to control a welding device having multiple independent welding execution units to weld multiple weld points on an automotive radiator, each of the welding execution units being configured with a state sensing unit and a parameter adjustment unit, the method comprising: During the welding process, each of the aforementioned state sensing units continuously acquires quantitative indicators reflecting the functional state of the corresponding welding execution unit; Based on the quantitative indicators of all the welding execution units, a hybrid benchmark is constructed, and the welding execution units that have experienced functional degradation and the corresponding degree of degradation are identified based on the hybrid benchmark. Based on the type of the welding part to which the welding execution unit that experienced the functional degradation belongs, determine the corresponding compensation strategy type; Based on the compensation strategy type and the attenuation level, the parameter adjustment unit performs graded compensation on the corresponding welding execution unit. Continuously monitor the welding quality indicators of each weld point and iteratively adjust them until the welding quality indicators of all weld points are restored to the preset equilibrium range.

[0006] The technical solutions described in this application embodiment have at least the following technical effects: The intelligent welding control method for automotive radiators provided in this application continuously acquires quantitative indicators reflecting the functional status of each welding execution unit through state sensing units configured in each welding execution unit during the welding process. This allows for the understanding of the performance degradation status of each unit, reducing the monitoring blind spot present in traditional solutions that only monitor power-side parameters, where the power supply is normal but the electrodes have degraded. Then, a hybrid benchmark is constructed based on the quantitative indicators of all welding execution units to identify welding execution units experiencing functional degradation and their corresponding degradation levels. Cross-validation using multiple benchmarks—group mutual calibration, historical trends, and absolute anchoring—overcomes the problem of missed detection by a single benchmark in scenarios of slow electrode degradation or collective drift. This accurately identifies individual welds with functional degradation and quantifies their degradation levels. Finally, based on the type of welding part to which the welding execution unit with functional degradation belongs, a corresponding compensation strategy type is determined, targeting different welds... The inherent differences in thermal conductivity at the joints are addressed by matching appropriate compensation methods to solve the problem of extremely low or even ineffective compensation efficiency of a uniform compensation strategy at thermally isolated parts. Then, based on the type of compensation strategy and the degree of attenuation, a parameter adjustment unit performs graded compensation on the corresponding welding execution units, adopting strength-adaptive compensation methods for different operating conditions. This compensates for insufficient energy without exacerbating the wear of the degraded electrode itself, ensuring that the compensation method is precisely matched to the physical characteristics and degree of degradation of the specific part. Finally, the welding quality indicators of each weld point are continuously monitored and iteratively adjusted until the welding quality indicators of all weld points are restored to the preset equilibrium range. The welding quality indicators are used as the final criterion for the compensation effect. After compensation, the evaluation is repeated, and the next iteration begins, thus forming a complete closed-loop control from perception, identification, compensation to verification, ensuring effective compensation and continuously improving the welding quality consistency of all weld points. This method, even under conditions where the performance of some welding execution units degrades, improves the uniformity of welding quality across all weld points through scenario-based and graded online closed-loop control, reducing the scrap rate and extending the effective life of the electrodes.

[0007] Secondly, embodiments of this application provide an intelligent welding control system for automotive radiators, used to control a welding device having multiple independent welding execution units to weld multiple weld points on an automotive radiator. Each welding execution unit is equipped with a status sensing unit and a parameter adjustment unit. The system includes: The sensing module is used to continuously acquire quantitative indicators reflecting the functional status of the corresponding welding execution unit through each of the state sensing units during the welding process. The identification module is used to construct a hybrid benchmark based on the quantitative indicators of all the welding execution units, and to identify the welding execution units that have experienced functional degradation and the corresponding degree of degradation based on the hybrid benchmark. The decision module is used to determine the corresponding compensation strategy type based on the type of the welding part to which the welding execution unit to which the functional degradation occurs; The compensation module is used to perform graded compensation on the corresponding welding execution unit through the parameter adjustment unit based on the compensation strategy type and the attenuation degree. The monitoring module is used to continuously monitor the welding quality indicators of each weld point and iteratively adjust them until the welding quality indicators of all weld points are restored to the preset equilibrium range.

[0008] Thirdly, embodiments of this application provide an intelligent welding production line for automotive radiators, comprising: A conveying device is used to sequentially transport automotive radiator workpieces to multiple welding stations; Multiple welding stations, each welding station is equipped with a welding device having multiple independent welding execution units, and each welding execution unit is equipped with a status sensing unit and a parameter adjustment unit; A control device electrically connected to the conveying device and the welding device, the control device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect above.

[0009] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0010] Fifthly, embodiments of this application provide a computer program product that, when run on a welding apparatus, causes the welding apparatus to execute the intelligent welding control method for automotive radiators described in the first aspect.

[0011] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating the intelligent welding control method for automotive radiators provided in an embodiment of this application. Figure 2 This is a schematic diagram of the distribution of heat sink welding point types and adjacent load sharing compensation provided in the embodiments of this application; Figure 3 This is a schematic diagram of the intelligent welding control system for automotive radiators provided in an embodiment of this application; Figure 4 This is a schematic diagram of the control device provided in the embodiments of this application. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0015] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0016] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0017] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determination" or "if the described condition or event is detected" may be interpreted, depending on the context, as "once determination," "in response to determination," "once the described condition or event is detected," or "in response to the detection of the described condition or event."

[0018] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0019] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0020] Traditional welding control systems use uniform preset welding parameters for all weld points. However, electrode performance continuously degrades with the accumulation of weld points, and the degradation rate of electrodes at different locations varies due to differences in heat dissipation conditions and stress states. When individual welding execution units degrade, their corresponding weld points may experience incomplete welds or over-welds. Traditional systems cannot detect the individual state differences of each weld point and perform targeted compensation, resulting in poor weld quality consistency, high product leakage rate, and increased scrap rate.

[0021] While some existing welding quality monitoring solutions exist, such as using the effective value of welding current to determine whether the power supply side is normal, or using post-weld visual inspection to remove defective products, these solutions have the following shortcomings: First, they only monitor power supply parameters and cannot reflect the actual performance degradation of the electrode side, resulting in a monitoring blind spot where the power supply is normal but the electrode has degraded; second, most of them are offline inspections or post-weld removals, lacking online closed-loop compensation capabilities for the welding process; third, they do not consider the essential differences in heat conduction characteristics between different welding parts of the heat sink (such as pipe-manifold connection and fin-heat pipe connection), and adopt a uniform compensation strategy, resulting in extremely low or even ineffective compensation efficiency at thermal isolation parts.

[0022] To address the aforementioned issues, this application provides an intelligent welding control method, system, and intelligent welding production line for automotive radiators. In this method, during the welding process, a state sensing unit configured in each welding execution unit continuously acquires quantitative indicators reflecting the functional state of the corresponding unit, thereby understanding the performance degradation state of each unit and reducing the monitoring blind spot present in traditional solutions that only monitor power-side parameters, where the power supply is normal but the electrodes have degraded. Then, a hybrid benchmark is constructed based on the quantitative indicators of all welding execution units, and welding execution units experiencing functional degradation and their corresponding degradation levels are identified accordingly. Cross-validation using multiple benchmarks—group mutual calibration, historical trends, and absolute anchoring—overcomes the problem of missed detection by a single benchmark in scenarios of slow electrode degradation or collective drift, thus accurately identifying functionally degraded individuals and quantifying their degradation levels from all weld points. Finally, based on the type of welding part to which the functionally degraded welding execution unit belongs, a corresponding compensation strategy type is determined, addressing the different welding parts in terms of heat conduction. The method employs appropriate compensation methods to address the issue of extremely low or even ineffective compensation efficiency of a uniform compensation strategy at thermally isolated areas, based on the inherent differences in characteristics. Then, based on the type and degree of attenuation of the compensation strategy, a parameter adjustment unit performs tiered compensation on the corresponding welding execution units, adopting strength-adaptive compensation methods for different operating conditions. This compensates for insufficient energy without exacerbating the wear and tear on the degraded electrode itself, ensuring a precise match between the compensation method and the physical characteristics and degree of degradation of the specific location. Finally, the welding quality indicators of each weld point are continuously monitored and iteratively adjusted until all weld points' welding quality indicators return to a preset equilibrium range. The welding quality indicators serve as the final criterion for the compensation effect. After compensation, a re-evaluation is conducted, and the process enters the next iteration, thus forming a complete closed-loop control from perception, identification, compensation to verification. This ensures effective compensation and continuously improves the consistency of welding quality across all weld points. Even under conditions where some welding execution units experience performance degradation, this method improves the uniformity of welding quality across all weld points through scenario-specific and tiered online closed-loop control, reducing the scrap rate and extending the effective lifespan of the electrodes.

[0023] The intelligent welding control method for automotive radiators provided in this application embodiment can be applied to a welding device. In this case, the welding device is the executing entity of the intelligent welding control method for automotive radiators provided in this application embodiment. This application embodiment does not impose any restrictions on the specific type of welding device.

[0024] For example, a welding apparatus may include multiple independent welding execution units and a control device, with the control device electrically connected to the welding execution units. Each welding execution unit includes a welding electrode (or welding head), a status sensing unit configured within that unit, and a parameter adjustment unit. In resistance spot welding scenarios, the status sensing unit preferably includes: a welding current transformer (Rogowski coil or Hall sensor), an inter-electrode voltage acquisition unit, and a displacement sensor (optical scale or inductive displacement sensor) built into the servo welding torch; the parameter adjustment unit is an inverter for the welding power supply and a servo pressure regulating mechanism, used to independently adjust the welding current, energizing time, and electrode pressure of that unit. The control device monitors and controls the entire welding process. For example, the control device may be a PLC, a microcontroller, an IoT terminal, a computer, etc.

[0025] To better understand the intelligent welding control method for automotive radiators provided in the embodiments of this application, the specific implementation process of the intelligent welding control method for automotive radiators provided in the embodiments of this application will be described by way of example below.

[0026] Figure 1 This illustration shows a schematic flowchart of an intelligent welding control method for automotive radiators provided in an embodiment of this application. The intelligent welding control method for automotive radiators includes: S100 continuously acquires quantitative indicators reflecting the functional status of the corresponding welding execution unit through each state sensing unit during the welding process.

[0027] It can be understood that quantitative indicators are a set of physical quantities that can be directly measured or indirectly calculated to reflect the performance changes of welding execution units during continuous operation. After repeated high-temperature and high-pressure welding operations, the welding execution unit's ability to transfer welding energy will gradually degrade. This degradation will be directly reflected in the characteristic parameters of the welding process. Therefore, by continuously acquiring quantitative indicators, the functional status of each unit can be monitored in real time.

[0028] For example, quantitative indicators can be constructed by extracting feature parameters from the dynamic resistance curve during the welding process, which can directly reflect the changes in the contact state between the electrode and the workpiece; quantitative indicators can also be constructed by collecting the displacement of the electrode end face, which can reflect the changes in the geometric dimensions after electrode wear; quantitative indicators can also be constructed by the visual features of the weld nugget image after welding, and so on, but not limited to these.

[0029] In one possible implementation, the quantification index includes characteristic parameters extracted from the dynamic resistance curve of the welding process; wherein, the dynamic resistance curve is a curve showing the change of resistance value with energization time during each welding energization period, obtained by synchronously collecting the voltage between electrodes and the welding current; the characteristic parameters include at least one of initial peak contact resistance, steady-state resistance, and resistance decrease rate; the initial peak contact resistance is used to reflect the degree of increase in contact resistance caused by electrode surface contamination and alloying; the steady-state resistance is used to reflect the degree of decrease in current density caused by mushrooming of the electrode end face due to plastic deformation; and the resistance decrease rate is used to reflect the dynamic change rate of the contact state from initial impact to steady state.

[0030] The dynamic resistance curve is understood to be a curve showing the ratio of inter-electrode voltage to welding current changing over time during resistance spot welding. Its morphological characteristics are strongly correlated with electrode condition and welding quality, making it a core data source characterizing the functional state of the welding execution unit. The initial peak contact resistance is a local maximum value of the dynamic resistance curve during the initial stage of energization (typically 0-5 milliseconds), and its magnitude is mainly affected by the thickness of the contamination layer on the electrode surface. The steady-state resistance is the average value of the dynamic resistance curve during the melting plateau period, and its value is mainly affected by the effective contact area of ​​the electrode end faces. The resistance drop rate is the slope from the initial peak value to the steady-state value, and its value reflects the intensity of melting. The rise time is the time taken from energization to reaching the initial peak contact resistance.

[0031] This setup, by clearly corresponding the specific composition of the quantitative indicators with the physical meaning of the dynamic resistance curve and the degradation causes reflected by each characteristic parameter, enables the subsequent calculation of individual deviation based on the hybrid benchmark to accurately distinguish different degradation modes such as electrode surface contamination and end-face mushrooming, providing a reliable quantitative basis for differentiated graded compensation.

[0032] In one possible implementation, the quantitative indicator also includes the displacement of the electrode end face acquired by a displacement sensor; wherein, the displacement of the electrode end face is the actual axial displacement change of the electrode rod after the welding electrode is closed, which is used to reflect the electrode wear deformation.

[0033] It can be understood that the electrode end face displacement refers to the actual displacement of the electrode under servo drive during the welding process. This displacement will undergo systematic shift after the electrode wears out, and can be used as a direct measure of the electrode's geometric state, complementing the characteristics of the dynamic resistance curve.

[0034] For example, the displacement of the electrode end face can be acquired by the grating ruler built into the servo welding gun; the absolute position change of the electrode rod can be acquired by the inductive displacement sensor; the displacement can also be estimated by measuring the change in the shape of the electrode end face through machine vision, and so on, but not limited to these.

[0035] In this embodiment, the displacement of the electrode end face is read in real time during each welding pressure holding stage by the grating ruler built into the servo welding gun. Its value shows a monotonically increasing relationship with the cumulative number of welding times of the electrode, which can be used as evidence of geometric degradation independent of electrical characteristics. When the increment of the displacement relative to the initial value of the new electrode exceeds the preset wear threshold (0.5 mm in this embodiment), it serves as an auxiliary criterion for triggering electrode regrinding.

[0036] This setup, through the fusion and acquisition of multi-source quantitative indicators, can comprehensively characterize the functional status of the welding execution unit from multiple dimensions, including electrical and geometric features, providing a sufficient data foundation for subsequent attenuation identification.

[0037] S200 constructs a hybrid benchmark based on the quantitative indicators of all welding execution units, and identifies welding execution units that have experienced functional degradation and the corresponding degree of degradation based on the hybrid benchmark.

[0038] Hybrid benchmarks refer to a reference standard for determining the normal state through multiple independent benchmark construction methods, in order to overcome the problem of missed detection by a single benchmark in scenarios of slow electrode degradation or collective drift. If only a single benchmark is used (such as a fixed factory benchmark or a simple sliding window), the system may misjudge degradation as normal when degradation accumulates slowly or all units drift collectively. If only a group mutual calibration benchmark (single layer) or a group mutual calibration plus historical EWMA (double layer) scheme is used, when a cooling system failure occurs on the production line causing all electrodes to degrade synchronously (collective drift), the group mutual calibration benchmark becomes invalid due to the synchronous offset of all units, and cannot identify degradation; when the electrode undergoes slow cumulative degradation, the EWMA of the double layer scheme will follow the degradation trend and miss early anomalies. Hybrid benchmarks use the physical dimensions of the electrode end face (independently measured by the vision system) as a correction anchor point that does not drift with degradation, and form cross-validation with group benchmarks and historical benchmarks in both collective drift and slow degradation scenarios, thereby eliminating the blind spots of a single benchmark.

[0039] For example, a primary benchmark can be constructed by the distribution center of the quantitative indicators of each welding execution unit within the same welding cycle, and abnormal individuals can be identified by the relative differences within the group; an auxiliary benchmark can also be constructed by the exponentially weighted moving average of the historical quantitative indicators of each welding execution unit outside the primary benchmark, and slow degradation can be captured by the historical trend of a single point; an absolute anchoring benchmark can also be constructed by periodically visually inspecting the electrode end face size, and fallback calibration can be performed by using the absolute measurement value of the physical size, and so on, but not limited to these.

[0040] In one possible implementation, in step S200, a hybrid benchmark is constructed based on the quantitative indicators of all welding execution units, including: S210 uses the distribution center of the quantitative index of each welding execution unit in each feature component dimension within the same welding cycle as the center value of the main benchmark in the corresponding feature component dimension. The distribution center is the robust statistic after removing extreme values, and the robust statistic is the truncated mean or median.

[0041] It is understandable that the master benchmark is used for horizontal comparison of welding execution units within the same welding cycle. Its core basis is that, under the premise of consistent nominal process conditions, the base material, assembly, and pressurization states experienced by each weld point within the same welding cycle are highly similar. The quantitative indicators of a healthy welding execution unit should cluster around a common center of normality across all characteristic component dimensions. Therefore, taking the distribution center of the quantitative indicators of all welding execution units within the same cycle on a certain characteristic component dimension as the center value of the master benchmark for that dimension can reflect the typical health level of the current cycle group. This application uses robust statistics instead of the arithmetic mean to characterize this distribution center because: if the arithmetic mean is used directly, individual welding execution units that have already experienced functional decline and whose quantitative indicators have deviated significantly will skew the average value, causing the master benchmark itself to be contaminated and weakening its ability to identify declining individuals; therefore, extreme values ​​in the set are first removed, and then the truncated mean or median of the remaining data is taken as the distribution center. The truncated mean is the average value obtained after removing the extreme samples at the beginning and end of the sample by a preset proportion. The median is the sample value in the middle position after sorting. Both are insensitive to extreme values ​​and can stably characterize the center of the healthy group even in the presence of some degenerate units, providing a reliable lateral reference for subsequent individual deviation calculations.

[0042] S220 uses the exponentially weighted moving average of historical quantitative indicators of each welding execution unit as an auxiliary benchmark.

[0043] It is understandable that auxiliary benchmarks are used for longitudinal tracking of individual welding execution units, and their significance lies in revealing the evolution trend of the same execution unit over time. Unlike the primary benchmark, which focuses on the lateral comparison of multiple units within the same cycle, the auxiliary benchmark focuses on the longitudinal comparison of the same unit across cycles: for any welding execution unit, the quantitative indicators of the same characteristic components corresponding to several historical welding cycles before entering the current cycle are taken as an exponentially weighted moving average in chronological order to obtain the auxiliary benchmark value of the unit in that dimension. The exponentially weighted moving average assigns greater weight to recent history and decreasing weight to distant history, thus smoothing out the random noise introduced by a single measurement and timely tracking the trend deviation caused by the slow degradation of unit performance, reducing the possibility of misjudging occasional fluctuations as degradation. By using the auxiliary benchmark in conjunction with the primary benchmark, this application can distinguish between two different types of deviations: the unit's own variation compared to its history and the unit's variation compared to the same cycle group, thereby improving the accuracy of attenuation identification.

[0044] S230 uses the periodic inspection results of the electrode end face dimensions of each welding execution unit as the absolute anchoring reference.

[0045] It is understandable that an absolute anchoring reference is an absolute reference derived directly from physical measurements, independent of welding electrical signals, used to objectively calibrate the actual geometric state of the electrode end faces of each welding execution unit. During long-term use, the electrode end faces of welding execution units undergo plastic deformation due to repeated pressure and high temperatures, i.e., mushrooming of the end faces, accompanied by changes in the effective contact area. This change is the most direct and fundamental physical cause of functional degradation. This application measures the electrode end face dimensions of each welding execution unit according to a preset cycle (e.g., after completing a preset number of welds, or before the start of each shift), and uses the measurement results as the absolute anchoring reference. Because this reference directly reflects the physical wear state of the electrodes and does not rely on relative comparisons of quantitative indicators, the absolute anchoring reference can still provide an independent physical correction anchor point when the main reference is distorted due to collective drift, or when the auxiliary reference is inaccurate due to historical data bias.

[0046] S240, the quantitative indicators of each welding execution unit within the same welding cycle are weighted and fused according to confidence level based on the main reference, auxiliary reference and absolute anchoring reference to obtain a hybrid reference.

[0047] It can be understood that a hybrid benchmark is a confidence-weighted fusion result of the primary benchmark, auxiliary benchmark, and absolute anchor benchmark on the same feature component dimension, which combines the advantages of the three and suppresses their respective blind spots. This application dynamically assigns a confidence weight to each type of benchmark in each fusion: when healthy units constitute the majority within the same cycle and the group distribution is concentrated, the horizontal comparison of the primary benchmark is reliable, and its weight is increased accordingly; when a unit has sufficient and stable historical data and a clear vertical trend, the weight of the auxiliary benchmark is increased; when the electrode end face size has just undergone a fresh detection and its physical state is clear, the weight of the absolute anchor benchmark is increased. Conversely, when the confidence of a certain type of benchmark decreases due to data anomalies (such as too many degenerate units in the current cycle causing contamination of the primary benchmark, or a missing detection window causing the absolute anchor benchmark to fail), its weight is reduced accordingly, and the fusion result relies more on the other two types of benchmarks. Through this adaptive weighted fusion based on confidence level, the hybrid benchmark can obtain stable and physically interpretable reference values ​​under different operating conditions such as slow degradation, collective drift, and single-point step changes, providing a unified and reliable judgment baseline for subsequent identification of functionally degraded individuals and the degree of degradation based on the hybrid benchmark. In the embodiments of this application, the confidence weights of various benchmarks are determined as follows: Main baseline confidence level:

[0048] Auxiliary baseline confidence level: Where historical data completeness = accumulated historical cycle number / preset window number, trend significance = |EWMA recent slope| / EWMA noise standard deviation; Absolute anchorage baseline confidence level: The shorter the time since the last electrode end face detection, the higher the weight; if the time exceeds the calibration cycle, the weight is set to zero.

[0049] The three types of weights are normalized and then used for weighted fusion: This configuration, through the integration of the three-layer structure of the main reference, auxiliary reference, and absolute anchoring reference, enables the hybrid reference to possess both the sensitivity of group lateral comparison and the continuity of single-point longitudinal tracking, while also maintaining physically interpretable correction capabilities when the electrode state undergoes a step change. This effectively avoids the problem of missed detection by a single reference in scenarios of slow degradation or collective drift.

[0050] In one possible implementation, step S200 involves identifying the welding execution unit experiencing functional degradation and its corresponding degree of degradation based on a hybrid reference, including: S250, calculate the individual deviation of each welding execution unit. The individual deviation is the weighted sum of the normalized deviations of each feature component included in the quantitative index relative to the corresponding dimension of the main reference.

[0051] As can be understood, a characteristic component refers to the value of each specific component contained in a quantitative indicator in a single dimension. For example, the initial peak contact resistance, steady-state resistance, and resistance decrease rate extracted from the dynamic resistance curve, as well as the displacement of the electrode end face collected by the displacement sensor, are each characteristic components. The corresponding dimension of the main reference refers to the distribution center of the main reference in the same dimension. Individual deviation is the bridge that maps multi-dimensional characteristic components to a single scalar indicator, enabling features of different dimensions and units to participate in degradation judgment in a unified manner, thereby achieving classification with an interpretable threshold.

[0052] Individual deviation can be calculated by weighted summation of normalized features relative to the main baseline; Mahalanobis distance can be used to measure the statistical distance between the feature vectors of each unit and the population center; fuzzy comprehensive evaluation method can be used to fuse the deviation of multiple features, etc., but is not limited to these.

[0053] In this embodiment of the application, individual deviation degree The calculation formula is: Where: i is the index of the welding execution unit, i=1,2,…,N, N is the total number of welding execution units; k is the index of the feature component, k=1,2,…,K, K is the total number of feature components involved in the determination, K=5 in this embodiment, corresponding to the initial peak contact resistance, steady state resistance, resistance decrease rate, rise time and electrode end face displacement, respectively. The actual measured value of the kth characteristic component of the i-th welding execution unit is acquired by step S100; σ_k is the truncated mean of the kth characteristic component of all welding execution units within the same welding cycle (i.e., the center value corresponding to the master reference); σ_k is the standard deviation of the kth characteristic component of all welding execution units within the same welding cycle, used for dimensional normalization. The weight coefficient for the k-th feature component is learned by the degradation pattern classifier based on historical welding data. =1, and in this embodiment, the initial value is taken as an equal weight of 0.2. The specific method for determining the weights may include: initially assigning equal weights to each feature; updating the weights offline in each maintenance cycle based on the discriminative power (such as information gain or F-value) of each feature component for normal / degenerate judgment; or obtaining the weights through supervised learning by a degradation mode classifier based on historical labeled data, but is not limited to this. In this embodiment, the formula in... That is, the reference value of the hybrid benchmark in this dimension is obtained by fusing the central value of the primary benchmark with confidence, and is used to replace the single central value of the primary benchmark.

[0054] S260 determines whether an individual's deviation is normal, slightly degraded, or severely degraded based on a comparison of the individual's deviation with a preset threshold.

[0055] It is understandable that the judgment rule is: when The condition is considered normal when the lower limit of the preset threshold is less than or equal to the lower limit of the preset threshold; when the lower limit of the preset threshold is less than or equal to the lower limit of the preset threshold, the condition is considered normal. When the value is less than the upper limit of the preset threshold, it is considered a slight degradation; when IDI_i is greater than or equal to the upper limit of the preset threshold, it is considered a severe degradation. The upper and lower limits of the preset threshold can be manually input, and their specific values ​​can be optimized based on the actual false positive rate during the process verification stage.

[0056] This configuration maps multi-dimensional feature components to a single scalarized individual deviation, and uses a threshold to achieve a three-level judgment of normal, slight degradation, and severe degradation, providing a unified and interpretable decision basis for subsequent scenario-based graded compensation.

[0057] In one possible implementation, the method also includes: Based on the characteristic parameters of the quantitative indicators, the degradation mode of the welding execution unit with functional decline is classified. The degradation mode includes at least one of the following: electrode end face mushrooming, surface contamination or alloying, insufficient cooling and centering misalignment; graded compensation is adaptively adjusted according to the degradation mode.

[0058] It is understandable that different degradation patterns may exhibit similar overall deviations in quantitative indicators, but their underlying physical mechanisms are different, and the appropriate compensation methods also differ. Therefore, classifying them before compensation can improve the targeting of compensation.

[0059] For example, patterns can be distinguished by the combined characteristics of initial peak contact resistance and steady-state resistance: a significant increase in initial peak contact resistance while the steady-state resistance remains essentially normal corresponds to surface contamination or alloying; a significant decrease in steady-state resistance while the initial peak contact resistance remains essentially normal corresponds to mushrooming of the electrode end face; a monotonically increasing steady-state resistance over time corresponds to insufficient cooling; centering offset patterns can be identified through the asymmetric characteristics of the melt core; degradation patterns can also be automatically identified based on multi-feature combinations using machine learning classifiers such as support vector machines or random forests, and so on, but are not limited to these.

[0060] In this embodiment, the degradation mode classification result can be used to guide the parameter selection of the graded compensation strategy in step S400: for the mushrooming mode of the electrode end face, it is preferred to increase the welding current or share the compensation with neighboring areas; for the surface contamination or alloying mode, it is preferred to increase the electrode pressure or perform electrode grinding; for the insufficient cooling mode, it is preferred to trigger the increase of cooling water flow or shutdown for maintenance; for the centering offset mode, it is triggered to adjust the mechanical centering rather than parameter compensation.

[0061] This setup, by identifying degradation patterns in advance and guiding the selection of compensation parameters accordingly, enables tiered compensation to take the most appropriate intervention measures for different underlying physical mechanisms, reducing the failure of uniform compensation indiscriminately for specific degradation patterns. Degradation patterns are classified as optional enhancement steps; even if the classification is not performed, step S400 can still perform default tiered compensation based on the degree of attenuation.

[0062] In one possible implementation, the method further includes: detecting the position of the electrode end face by an electrode displacement sensor when the circuit is closed under no-load conditions; when the position change exceeds a preset threshold, determining that an electrode replacement or electrode regrinding event has occurred, and automatically resetting the hybrid reference.

[0063] It is understandable that replacing or grinding electrodes can cause a step change in the characteristics of the welding execution unit. If the reference is not reset, the system may misjudge the normal low deviation of the new electrode as abnormal, or misjudge the normal state of the worn electrode after it has been replaced as a sudden improvement. Therefore, the reference needs to be reset in time after the event occurs.

[0064] For example, the position of the electrode end face can be detected by performing an unloaded closing action and reading the displacement sensor reading before the start of each welding cycle; the event signal can also be directly obtained by setting a position switch or RFID tag on the electrode grinding machine or electrode changing mechanism; the event can also be inferred by monitoring whether the quantitative index shows a step change exceeding a preset amplitude during multiple consecutive welding operations, and so on, but not limited to these.

[0065] In this embodiment, when an electrode replacement or grinding event is detected, the system automatically performs the following benchmark reset operation: clears the auxiliary benchmark EWMA history sequence of the welding execution unit and reinitializes it; marks the absolute anchoring benchmark as pending update and refreshes it after the next visual measurement; the first N_w weld points after reset (N_w=30 in this embodiment) enter the break-in period, during which only data is collected to establish the initial benchmark and does not participate in the degradation judgment, so as to avoid misjudgment caused by the change of characteristics during the break-in period.

[0066] With this setting, when the characteristics of the welding execution unit change abruptly due to electrode replacement or electrode grinding, the reference can be automatically reset and a break-in period can be set, avoiding misjudging the normal state of a brand-new electrode as abnormal, or misjudging the normal state of a worn electrode after replacement as a sudden improvement.

[0067] S300 determines the corresponding compensation strategy type based on the type of welding part to which the welding execution unit that experienced functional degradation belongs.

[0068] Understandable, such as Figure 2 As shown, different welded parts of a heat sink exhibit fundamental differences in heat conduction characteristics: in the tube-manifold connection, adjacent weld points are connected by a continuous aluminum alloy base material, creating a clear heat conduction path; while in the fin-heat pipe connection, adjacent weld points are typically separated by air gaps or non-metallic materials, essentially blocking the heat conduction path. This difference directly determines the effectiveness of proximity-based compensation strategies in different parts; therefore, it is necessary to determine the part type before compensation to select a matching compensation strategy.

[0069] For example, the compensation strategy type can be determined by querying a pre-established radiator welding point-part type mapping table, which is automatically loaded by the MES system when importing product models; the part type can also be automatically determined by visually recognizing the geometric layout and adjacency relationship of the welding points; the specific part of the current welding can also be determined by the sensor signals of the workpiece clamping station, and so on, but not limited to these.

[0070] In one possible implementation, the types of welded parts include pipe-manifold connections and fin-heat pipe connections. Pipe-manifold connections are where there is a heat conduction path between adjacent weld points, while fin-heat pipe connections are where adjacent weld points are thermally isolated. In step S300, based on the type of welded part to which the welding execution unit experiencing functional degradation belongs, a corresponding compensation strategy type is determined, including: S310, if the welding execution unit with degraded function belongs to the pipe-manifold connection, a first compensation strategy including proximity sharing compensation is determined, wherein proximity sharing compensation is an indirect energy supplement that utilizes a portion of the heat generated by the increased energy output of adjacent healthy solder joints to conduct to the degraded solder joints. If the welding execution unit with degraded function belongs to the heat sink-heat pipe connection, a second compensation strategy of direct compensation at this point is determined.

[0071] It is understandable that the first compensation strategy is applicable to the pipe-manifold connection because there is a heat conduction path formed by aluminum alloy base material between adjacent solder joints in this part. After increasing the energy output of the adjacent healthy solder joints, some heat can be conducted to the degraded solder joint area through the base material, thus achieving indirect heat sharing and compensation. The second compensation strategy is applicable to the heat sink-heat pipe connection because the heat conduction path in this part is blocked, and the efficiency of adjacent heat sharing and compensation is extremely low. It is necessary to directly compensate the parameters of the degraded solder joint itself.

[0072] For example, in the first compensation strategy, the solder joints participating in the adjacent sharing compensation can be selected from the 1 to 2 healthy solder joints on the left and right sides of the degraded solder joint; in the second compensation strategy, the local point direct compensation can only adjust the welding current or energizing time of the degraded solder joint itself; it can also adjust multiple parameters at the same time according to the degree of degradation, etc., but not limited to these.

[0073] In this embodiment, the correspondence between part type and compensation strategy type can be stored in a strategy mapping table. The table entries include part type identifier, spacing between adjacent solder joints, and calibration value of heat conduction efficiency coefficient η. The heat conduction efficiency coefficient η is obtained through CAE thermal simulation combined with process test calibration. For a typical pipe-manifold connection (spacing 15~30 mm, 6061 aluminum alloy), the value range of η is 0.05~0.15. The calibration method for the heat transfer efficiency coefficient η includes: Step a, establishing a finite element heat transfer model of the pipe-manifold connection, inputting actual welding process parameters (base welding current I_base, energizing time t_base, base material grade and thermal conductivity, and distance d between adjacent weld points), simulating to obtain the ratio of the temperature rise ΔT_comp in the degraded weld point region under the influence of the energy increment ΔI_neighbor of the adjacent healthy weld point to the temperature rise ΔT_self required for direct compensation at this point, and using this ratio as the calibration value of η; Step b, in the first trial welding, measuring the actual temperature rise of the degraded weld point region after the adjacent energy increase with an infrared thermal imager, correcting the calibration value of step a, so that the calibration error is controlled within ±0.02; For a typical pipe-manifold connection (spacing 15~30 mm, 6061 aluminum alloy), the calibrated η value range is 0.05~0.15.

[0074] This setup establishes a clear mapping between the welding location type and the compensation strategy type, enabling the pipe-manifold connection to utilize the heat conduction path of the aluminum alloy base material for adjacent heat sharing compensation, while allowing the heat-isolated radiator-heat pipe connection to directly adopt local point compensation, ensuring that the compensation method is precisely matched with the physical characteristics of the specific location.

[0075] S400 performs graded compensation on the corresponding welding execution unit according to the type of compensation strategy and the degree of attenuation.

[0076] It is understandable that graded compensation refers to adopting different compensation methods based on the degree of attenuation: for slight degradation, mild compensation methods that cause less disturbance to the system and do not aggravate electrode wear are preferred; for severe degradation, stronger point parameter adjustment methods are required, and multiple welding parameters are linked to maintain the stability of total heat input.

[0077] For example, for minor degradation, compensation can be achieved by increasing the energy output of adjacent healthy solder joints, or by directly and slightly increasing the current or time of the degraded solder joint itself. For severe degradation, the total heat input can be maintained by reducing the current of the degraded solder joint and extending the power-on time in conjunction with it, or by triggering an electrode replacement warning combined with temporary parameter adjustments to maintain production continuity, and so on, but not limited to these.

[0078] In one possible implementation, based on the compensation strategy type and the attenuation level, the parameter adjustment unit performs graded compensation on the corresponding welding execution unit, including: S410, when the attenuation level is slight degradation, if the first compensation strategy is adopted, the energy output of at least one healthy welding execution unit adjacent to the degraded welding execution unit is increased, and the degradation part is compensated by heat conduction between the metal base material of adjacent weld points.

[0079] It is understandable that in the pipe-manifold connection, increasing the welding energy of the adjacent healthy solder joint will increase the size of the weld nugget and raise the local temperature. Some of the heat will be conducted along the aluminum alloy base material to the degraded solder joint area, thereby compensating to some extent for the insufficient energy of the degraded solder joint caused by the decline in electrode performance.

[0080] For example, the energy output of the nearest healthy solder joint on one side of the degraded solder joint can be increased; the energy output of the nearest healthy solder joint on both sides of the degraded solder joint can be increased simultaneously to balance the heat field distribution; the set of solder joints participating in the heat sharing can be dynamically selected based on the distance between the degraded solder joint and each neighboring solder joint and the current state, etc., but not limited to these.

[0081] In this embodiment, the energy increment near a healthy solder joint Calculate using the following formula: in: This represents the upper limit of the current increment between adjacent solder joints; in this embodiment, it is taken as +15% × It is used to prevent over-welding of adjacent solder joints due to excessive energy; η is the heat transfer efficiency coefficient, with a typical value of 0.05~0.15; This is the reference welding current corresponding to this product model; The threshold for determining severe degradation is defined in step S260; the min(·) function is used to constrain the increment from not exceeding the upper limit. This is to prevent adjacent solder joints from being soldered.

[0082] It should be noted that the thermal efficiency of proximity-based compensation is limited by Fourier's law of heat conduction, and its contribution to the temperature rise of 15 mm pitch aluminum alloys is limited (approximately 15%~25%) within a 200-millisecond welding time. Therefore, this strategy is only used as an auxiliary compensation method for slight degradation and is not relied upon for severe degradation.

[0083] With this setup, the adjacent area is used to share the compensation for minor degradation at the pipe-manifold connection. This allows for partial indirect supplementation of the energy-deficient area by utilizing the thermal conductivity of the aluminum alloy base material without exacerbating the wear of the degraded electrode itself. The parameter fine-tuning at this point undertakes the main compensation share.

[0084] S420, when the attenuation level is slight degradation, if the second compensation strategy is adopted, the energy output of the degradation welding execution unit itself will be directly increased.

[0085] It is understandable that in the heat sink-heat pipe connection, due to the thermal isolation between adjacent solder joints, the adjacent energy sharing compensation is basically ineffective. Therefore, for slight degradation, a small amount of energy compensation is directly applied to the degraded solder joint itself to make up for its energy deficiency in the most direct way.

[0086] For example, the welding current of the degraded weld joint can be increased directly (e.g., 3%~8%); the energizing time of the degraded weld joint can be extended directly (e.g., 15%~30%); the current and time parameters can be adjusted slightly at the same time to match the material characteristics of the part, etc., but not limited to these.

[0087] In this embodiment of the application, the energy increment at this point under the second compensation strategy Calculate using the following formula: in: To determine the upper limit of the current increment at the degraded solder joint, in this embodiment, it is taken as +10% × β is the local compensation efficiency coefficient. Since there is no heat conduction loss in the heat sink-heat pipe connection, the value of β is greater than η. In the embodiments of this application, it is taken as 0.3~0.5.

[0088] This configuration allows for direct compensation at the heatsink-heat pipe connection point for minor degradation, overcoming the limitation of ineffective adjacent heat sharing under thermal isolation conditions, thus ensuring that degraded solder joints in this area are also effectively compensated.

[0089] S430, when the attenuation level is severe degradation, the welding current is reduced and the welding time is extended in conjunction with the degradation welding execution unit to maintain the total heat input unchanged.

[0090] It is understandable that the electrode transfer efficiency has decreased significantly when the electrode is severely degraded. Continuing to increase the current may exceed the rated output of the power supply or cause further damage to the electrode. Therefore, a strategy of derating (reducing the current) and extending the energizing time is adopted. By using the relationship Q=I²Rt, the total heat input Q is kept constant while the current is reduced by extending the time, thereby ensuring the welding quality without aggravating electrode wear.

[0091] For example, the current can be reduced to 70% to 90% of the reference value and the energizing time can be extended accordingly; the derating ratio and delay can be dynamically calculated based on the measured degree of degradation; the electrode pressure can be slightly increased to improve contact while operating under derating conditions, etc., but not limited to these.

[0092] In this embodiment of the application, the derating current and extend time It can be calculated using the following formula: Wherein: α is the reduction coefficient, which is taken as 0.1~0.3 in this embodiment of the application to limit the maximum reduction range; The upper limit of the range for individual deviation can be determined by process verification; The reference welding current; The reference power-on time is provided by the process parameter table; The additional power-on time required beyond the baseline power-on time, so that the total heat input Q = ×R×( + )≥Q target (Target heat input). In this embodiment, maintaining a constant total heat input Q requires that the extension time after reducing the current satisfy a quadratic relationship: ,in For the derating welding current, For reference power-on time; for example, when hour, , making The total heat input Q remains constant, rather than being extended nonlinearly.

[0093] This setup employs a strategy of derating operation and interlocking delay for severe degradation. Even when electrode performance has significantly declined, the total heat input is maintained through time-dimensional compensation, reducing the vicious cycle of accelerated electrode failure caused by forcibly increasing the current.

[0094] The S500 continuously monitors the welding quality indicators of each weld point and iteratively adjusts them until the welding quality indicators of all weld points are restored to the preset equilibrium range.

[0095] It is understandable that welding quality indicators are the ultimate criterion for measuring the effectiveness of compensation. By continuously monitoring this indicator and re-evaluating it after compensation, we can verify whether the compensation is effective. If it still does not meet the standard, we will enter the next iteration to form a closed-loop control.

[0096] For example, the welding quality indicator can be the diameter of the weld nugget after welding, which can be measured online using a vision system; or it can be the shear strength of the weld joint, which can be determined using an empirical formula based on the weld nugget diameter. =k× It can be deduced; the quality of melt nucleus formation can also be indirectly evaluated through the closed characteristics of the dynamic resistance curve, etc., but is not limited to this.

[0097] In an optional preferred embodiment, the method further includes: When the required compensation exceeds the rated output capacity of the welding power supply, or when the diameter of the weld nugget after compensation exceeds the acceptable range, it is determined that it cannot be compensated, an alarm is triggered, and electrode replacement is recommended; when the frequency of triggering severe degradation judgment by the same welding execution unit within a preset cycle exceeds the threshold, the electrode life is determined to be terminated, and it is forcibly replaced.

[0098] It is understandable that not all electrode degradation can be resolved through parameter compensation. When the degree of degradation exceeds the upper limit of the welding power supply's adjustment capability, or when the compensation action itself leads to new quality problems such as over-welding, continued compensation is meaningless and should be deemed uncompensable, prompting manual intervention. When an electrode frequently triggers severe degradation in a short period of time, it indicates that it has reached the end of its lifespan and should be forcibly replaced to avoid the continuous production of defective products.

[0099] For example, the non-compensable determination can be triggered based on whether the single compensation requirement exceeds 120% of the rated output capacity of the power supply; it can also be triggered based on whether the diameter of the weld nugget after compensation exceeds the upper limit of the process (e.g., greater than 1.3 times the standard diameter); the electrode life termination determination can be triggered based on whether the number of severe degradation triggers in the most recent 1000 welding operations exceeds a preset number (e.g., 20 times), and so on, but is not limited to these.

[0100] In this embodiment of the application, the triggering condition for the non-compensable determination is: the amount requiring compensation. ,in The rated output current of the welding power source, or the calculated weld nugget diameter after compensation. ,in The standard weld nugget diameter is the upper limit; the trigger condition for electrode life termination is: in the most recent 1000 welding operations, the number of severe degradation triggers exceeds 20. If either of the above two conditions is met, a corresponding alarm signal or forced replacement command will be generated, and the automatic compensation logic of the station where the welding execution unit is located will be suspended until manual confirmation.

[0101] This setup, through dual protection of non-compensable determination and electrode lifespan termination determination, can promptly switch to the hardware replacement path when parameter compensation methods fail, ensuring production continuity and preventing defective products from flowing into the next process.

[0102] In one possible implementation, tiered compensation also includes: After identifying a welding execution unit with degraded function, the position of the corresponding weld point in the welding sequence is postponed to the later part of the sequence, and the heat accumulated by the previous weld point is used to indirectly compensate the weld point for energy.

[0103] It is understandable that the welding sequence refers to the preset welding order of all weld points on the same workpiece. During the spot welding of a heat sink, the earlier welded positions have better heat dissipation conditions, while the later welded positions receive a larger actual heat input due to the overall temperature rise of the workpiece. Actively relocating degraded weld points to the later stages of the welding sequence can utilize the heat accumulated by the preceding weld points to provide additional indirect energy compensation. This is a zero-cost compensation path that does not exacerbate electrode wear.

[0104] For example, degraded solder joints can be moved directly to the last position of the current welding cycle; the number of solder joints that need to be postponed can be calculated based on the degree of degradation and moved to a position in the sequence where heat accumulation is sufficient; robot path planning can also be taken into account during sequence scheduling to avoid large amounts of idle travel, etc., but not limited to these.

[0105] In this embodiment, the welding sequence rearrangement follows the following constraint: the spacing between adjacent weld points in a single welding loop after rearrangement is not less than the minimum spacing to prevent current shunting effect. (In the embodiments of this application) =12 mm); rearrangement does not disrupt the clamping stability of the radiator structure; the robot trajectory change caused by rearrangement is within the kinematically achievable range. This minimum spacing Based on the empirical value of the critical shunting distance for aluminum alloy resistance spot welding, when the distance between adjacent weld points is less than this value, the preceding weld point will form a significant shunting channel, resulting in insufficient heat input to the subsequent weld point.

[0106] This setup, without increasing any hardware power consumption, utilizes the heat accumulation effect inherent in the welding sequence itself to provide additional indirect compensation for degraded solder joints, further enhancing the compensation margin of the solution.

[0107] To facilitate understanding of the overall control logic of this application, the following explanation will be based on a specific application scenario.

[0108] A certain all-aluminum radiator production line uses resistance spot welding. Each workpiece contains 96 weld points distributed across the pipe-manifold connection (64 points) and the fin-heat pipe connection (32 points), completed collaboratively by 8 independent welding execution units. During the continuous production of the 500th workpiece, the reference identification module detected the individual deviation of execution unit #3 (responsible for the 8 weld points in the pipe-manifold area) using a hybrid reference. The energy level rose to 2.8, indicating slight degradation with the degradation mode being mushrooming of the electrode face. Based on the unit's location type (pipe-manifold connection), the strategy decision module determined to adopt the first compensation strategy (proximity sharing compensation). The compensation execution module calculated the energy increment of adjacent healthy solder joints. =min(15%× 0.12× ×2.8 / 4.0)≈8.4%× The current of the healthy execution units adjacent to each other on both sides of unit #3 is increased by approximately 8.4%. Simultaneously, the sequence scheduling module postpones the degraded solder joints among the eight solder joints in this area to the later part of the current cycle sequence. The next welding cycle is then retested after compensation. The temperature dropped back to 1.6, entering the normal range, verifying the effectiveness of the compensation. The entire process was conducted online without downtime, ensuring consistent welding quality for this batch of workpieces.

[0109] To quantitatively verify the technical effect of the method of this application, under the conditions of the same workpiece model (all-aluminum heat sink, 96 solder points), the same electrode brand and the number of sampled solder points (500 units), the traditional unified parameter group was set as the control group and the method group of this application was set as the experimental group. The comparison results are shown in Table 1 below (the following values ​​are illustrative examples and are not limiting measured limits): Table 1 It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0110] Corresponding to the intelligent welding control method for automotive radiators described in the above embodiments, this application also provides an intelligent welding control system for automotive radiators, wherein each module of the system can implement each step of the intelligent welding control method for automotive radiators. Figure 3 This paper shows a structural block diagram of an intelligent welding control system for automotive radiators provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0111] Reference Figure 3 The intelligent welding control system for automotive radiators includes: The sensing module is used to continuously acquire quantitative indicators reflecting the functional status of the corresponding welding execution unit through various state sensing units during the welding process.

[0112] The identification module is used to construct a hybrid benchmark based on the quantitative indicators of all welding execution units, and to identify welding execution units that have experienced functional degradation and the corresponding degree of degradation based on the hybrid benchmark.

[0113] The decision module is used to determine the corresponding compensation strategy type based on the type of welding part to which the welding execution unit that has experienced functional degradation belongs.

[0114] The compensation module is used to perform graded compensation on the corresponding welding execution unit through the parameter adjustment unit based on the compensation strategy type and attenuation level.

[0115] The monitoring module is used to continuously monitor the welding quality indicators of each weld point and iteratively adjust them until the welding quality indicators of all weld points are restored to the preset equilibrium range.

[0116] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0117] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described module division is merely an example. In practical applications, the functions described above can be assigned to different modules as needed, that is, the internal structure of the device can be divided into different modules to complete all or part of the functions described above. The modules in the embodiments can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the modules in the above device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0118] This application embodiment also provides an intelligent welding production line for automotive radiators, including a conveying device, multiple welding stations, and a control device, wherein: The conveying device is used to sequentially transport automotive radiator workpieces to multiple welding stations.

[0119] Each welding station is equipped with a welding device with multiple independent welding execution units, and each welding execution unit is equipped with a status sensing unit and a parameter adjustment unit.

[0120] The control device is electrically connected to the conveying device and the welding device. Figure 4 This is a schematic diagram of the structure of the control device 4 provided in one embodiment of this application. Figure 4 As shown, the control device 4 in this embodiment includes: at least one processor 40 ( Figure 4 Only one is shown in the image), at least one memory 41 ( Figure 4 (Only one is shown in the image) and a computer program 42 stored in the at least one memory 41 and executable on the at least one processor 40. When the processor 40 executes the computer program 42, it causes the control device 4 to perform the steps in any of the above embodiments of the intelligent welding control method for automotive radiators, or causes the control device 4 to perform the functions of each module in the above embodiments of the device.

[0121] For example, the computer program 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 42 in the control device 4.

[0122] The control device 4 can be a microcontroller, PLC, or other computing device. The control device 4 may include, but is not limited to, a processor 40 and a memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of control device 4 and does not constitute a limitation on control device 4. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0123] The processor 40 can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0124] In some embodiments, the memory 41 may be an internal storage unit of the control device 4, such as a hard disk or memory of the control device 4. In other embodiments, the memory 41 may be an external storage device of the control device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the control device 4. Furthermore, the memory 41 may include both internal storage units and external storage devices of the control device 4. The memory 41 is used to store operating systems, applications, bootloaders, data, and other programs, such as the program code of computer programs. The memory 41 can also be used to temporarily store data that has been output or will be output.

[0125] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0126] This application provides a computer program product that, when run on a welding apparatus, causes the welding apparatus to perform the steps in any of the above-described method embodiments.

[0127] 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, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying the computer program code to the welding apparatus, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, such as a USB flash drive, a portable hard drive, a magnetic disk, or an optical disk.

[0128] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0129] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0130] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the above-described embodiments of the intelligent welding control system for automotive radiators are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules 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 indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0131] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0132] The above-described 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, and should all be included within the protection scope of this application.

Claims

1. A smart welding control method for automotive radiators, characterized in that, A method for controlling a welding apparatus having multiple independent welding execution units to weld multiple weld points on an automotive radiator, each of the welding execution units being configured with a status sensing unit and a parameter adjustment unit, the method comprising: During the welding process, each of the aforementioned state sensing units continuously acquires quantitative indicators reflecting the functional state of the corresponding welding execution unit; Based on the quantitative indicators of all the welding execution units, a hybrid benchmark is constructed, and the welding execution units that have experienced functional degradation and the corresponding degree of degradation are identified based on the hybrid benchmark. Based on the type of the welding part to which the welding execution unit that experienced the functional degradation belongs, determine the corresponding compensation strategy type; Based on the compensation strategy type and the attenuation level, the parameter adjustment unit performs graded compensation on the corresponding welding execution unit. Continuously monitor the welding quality indicators of each weld point and iteratively adjust them until the welding quality indicators of all weld points are restored to the preset equilibrium range.

2. The intelligent welding control method for automotive radiators as described in claim 1, characterized in that, The quantitative indicators include characteristic parameters extracted from the dynamic resistance curve of the welding process; wherein, the dynamic resistance curve is a curve showing the change of resistance value with energization time, obtained by synchronously collecting the voltage between electrodes and the welding current during each welding energization period; the characteristic parameters include at least one of initial peak contact resistance, steady-state resistance, and resistance decrease rate; the initial peak contact resistance is used to reflect the degree of increase in contact resistance caused by electrode surface contamination and alloying; the steady-state resistance is used to reflect the degree of decrease in current density caused by mushrooming of the electrode end face due to plastic deformation; and the resistance decrease rate is used to reflect the dynamic change rate of the contact state from initial impact to steady state.

3. The intelligent welding control method for automotive radiators as described in claim 1 or 2, characterized in that, The quantitative indicators also include the displacement of the electrode end face collected by the displacement sensor; wherein, the displacement of the electrode end face is the actual axial displacement change of the electrode rod after the welding electrode is closed, which is used to reflect the electrode wear deformation.

4. The intelligent welding control method for automotive radiators as described in claim 1, characterized in that, The construction of a hybrid benchmark based on the quantitative indicators of all the welding execution units includes: The distribution center of the quantitative index of each welding execution unit in each feature component dimension within the same welding cycle is used as the center value of the main reference in the corresponding feature component dimension. The distribution center is a robust statistic after removing extreme values. The robust statistic is the truncated mean or median. The exponentially weighted moving average of the historical quantitative indicators of each welding execution unit is used as an auxiliary benchmark. The periodic inspection results of the electrode end face dimensions of each welding execution unit are used as the absolute anchoring reference; The quantitative indicators of each welding execution unit within the same welding cycle are weighted and fused according to confidence level based on the main reference, the auxiliary reference, and the absolute anchoring reference to obtain a hybrid reference.

5. The intelligent welding control method for automotive radiators as described in claim 4, characterized in that, The identification of the welding execution unit with functional degradation and the corresponding degree of degradation based on the hybrid reference includes: Calculate the individual deviation of each of the welding execution units; wherein the individual deviation is the weighted sum of the normalized deviations of each feature component included in the quantitative index relative to the corresponding dimension of the main reference. Based on the comparison between the individual deviation and the preset threshold, the welding execution unit that has experienced functional degradation and the corresponding degree of degradation are identified; wherein, the degree of degradation includes normal, slight degradation or severe degradation.

6. The intelligent welding control method for automotive radiators as described in claim 1, characterized in that, The types of welding locations include pipe-manifold connections and fin-heat pipe connections. Pipe-manifold connections are locations where there is a heat conduction path between adjacent weld points, while fin-heat pipe connections are locations where adjacent weld points are thermally isolated. The determination of the corresponding compensation strategy type based on the type of welding location to which the welding execution unit experiencing the functional degradation belongs includes: If the welding execution unit that experiences the aforementioned functional degradation is located at the pipe-manifold connection, a first compensation strategy including proximity sharing compensation is determined; wherein, the proximity sharing compensation is an indirect energy supplement that utilizes a portion of the heat generated by the increased energy output of adjacent healthy welds to conduct to the degraded welds. If the welding execution unit that experiences the aforementioned functional degradation is located at the heat sink-heat pipe connection point, the second compensation strategy of direct compensation at this point is determined to be adopted.

7. The intelligent welding control method for automotive radiators as described in claim 6, characterized in that, The step of performing graded compensation on the corresponding welding execution unit through the parameter adjustment unit based on the compensation strategy type and the attenuation degree includes: When the attenuation level is slight degradation, if the first compensation strategy is adopted, the energy output of at least one healthy welding execution unit adjacent to the degraded welding execution unit is increased, and the degradation part is compensated by heat conduction between the metal base material of adjacent weld points; if the second compensation strategy is adopted, the energy output of the degraded welding execution unit itself is directly increased. When the attenuation level is severe degradation, the welding current is reduced and the welding time is extended in conjunction with the degradation of the welding execution unit to maintain the total heat input unchanged.

8. The intelligent welding control method for automotive radiators as described in claim 1, characterized in that, The tiered compensation also includes: After identifying a welding execution unit with degraded function, the position of the weld point corresponding to the welding execution unit in the welding sequence is postponed to the later part of the sequence, and the heat accumulated by the previous weld point is used to indirectly compensate the weld point for energy.

9. An intelligent welding control system for automotive radiators, characterized in that, A welding apparatus having multiple independent welding execution units is used to weld multiple weld points on an automotive radiator. Each welding execution unit is equipped with a status sensing unit and a parameter adjustment unit. The system includes: The sensing module is used to continuously acquire quantitative indicators reflecting the functional status of the corresponding welding execution unit through each of the state sensing units during the welding process. The identification module is used to construct a hybrid benchmark based on the quantitative indicators of all the welding execution units, and to identify the welding execution units that have experienced functional degradation and the corresponding degree of degradation based on the hybrid benchmark. The decision module is used to determine the corresponding compensation strategy type based on the type of the welding part to which the welding execution unit to which the functional degradation occurs; The compensation module is used to perform graded compensation on the corresponding welding execution unit through the parameter adjustment unit based on the compensation strategy type and the attenuation degree. The monitoring module is used to continuously monitor the welding quality indicators of each weld point and iteratively adjust them until the welding quality indicators of all weld points are restored to the preset equilibrium range.

10. An intelligent welding production line for automotive radiators, characterized in that, include: A conveying device is used to sequentially transport automotive radiator workpieces to multiple welding stations; Multiple welding stations, each welding station is equipped with a welding device having multiple independent welding execution units, and each welding execution unit is equipped with a status sensing unit and a parameter adjustment unit; A control device electrically connected to the conveying device and the welding device, the control device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method as described in any one of claims 1 to 8.

Images