Machine tool welding method and system for metal material welding

Abstract

The invention relates to the technical field of machine tool welding, in particular to a machine tool welding method and system for metal material welding, and the method comprises the following steps: injecting detection current which does not cause metal melting into a welding loop in a pre-pressing stage of pressing a to-be-welded metal part by an electrode of a welding machine tool; measuring the initial contact resistance of the welding loop; establishing and tracking an electrode state reference according to the initial contact resistance; after main welding current is applied to a welding machine tool, disturbance caused by electrode state changes in a main welding signal is recognized and decoupled through the electrode state reference; according to the recognition and decoupling result, differential adjustment is conducted on welding parameters of the welding machine tool; by introducing an electrode state reference and signal identification decoupling mechanism, deeper understanding and more accurate control on the welding process are realized, and the limitation that material differences and electrode state changes are difficult to effectively deal with in a complex production environment in the prior art is overcome.

Description

Technical Field

[0001] This invention relates to the field of machine tool welding, and more particularly to a machine tool welding method and system for welding metal materials. Background Technology

[0002] In modern industrial production, welding machine tools are core equipment for achieving efficient and automated joining of metal parts. Resistance spot welding, as a commonly used joining process, uses welding machine tools to precisely control parameters such as electrode pressure, welding current, and time, causing localized melting at the contact point of multi-layer metal sheets to form a weld nugget, thus achieving a reliable connection.

[0003] However, ensuring the stability of weld quality remains a significant challenge in the large-scale continuous production of welding machine tools. This stems primarily from two coupled factors: First, the inherent fluctuations in the material of the workpiece itself. Even with metallic materials conforming to the same standards, microscopic differences in chemical composition, grain structure, and surface condition between different batches can lead to variations in key physical parameters such as electrical and thermal conductivity, thereby affecting heat input and weld nugget formation during the welding process. When welding machine tools employ fixed preset parameters, these material fluctuations directly result in inconsistencies in weld nugget size and quality.

[0004] Secondly, there are dynamic changes in the state of the electrodes, the actuators of the welding machine tool. During long-term continuous welding operations, the ends of the welding electrodes inevitably wear down. After repeated contact with high-temperature steel plates and under enormous pressure, the contact surface of copper alloy electrodes gradually flattens, expands, and may even become concave. Simultaneously, zinc layers from galvanized steel plates or oxides on the steel plate surface may form an adhesion layer on the electrode surface, leading to contamination. These changes in electrode state directly alter the contact resistance and heat dissipation conditions between the electrode and the workpiece. For example, electrode wear leading to an increased contact area may reduce the current density per unit area, thereby reducing localized heating; while electrode surface contamination may increase contact resistance, leading to localized overheating or even spatter. These changes in electrode state can also independently cause anomalies in real-time signals such as electrode voltage and current waveforms.

[0005] Currently, welding machine tools generally rely on real-time monitoring of welding current, voltage, and other signals to achieve quality control and adaptive parameter adjustment. However, the key issue is that both material fluctuations and electrode condition changes affect the welding electrical signal, making the root cause of signal variations unclear. The control system of the welding machine tool struggles to distinguish whether a signal anomaly stems from changes in workpiece material requirements or deterioration of the electrode condition. This misjudgment leads to incorrect parameter compensation: for example, misinterpreting insufficient energy caused by electrode wear as material demand and blindly increasing the current may exacerbate electrode burn-out and cause spatter; conversely, it may lead to incomplete penetration. This kind of adaptive adjustment with unclear root causes amplifies production fluctuations and reduces the overall yield. Therefore, there is an urgent need for a welding method that can be used in welding machine tools to accurately identify the root cause of signal disturbances and perform correct compensation. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by proposing a machine tool welding method and system for welding metal materials.

[0007] In a first aspect, the present invention provides a machine tool welding method for welding metal materials, the method being implemented by the execution system of the welding machine tool, the method comprising the following steps: During the pre-pressing stage of electrode pressing of the metal parts to be welded on the welding machine tool, a probe current that does not cause metal melting is injected into the welding circuit, and the initial contact resistance of the welding circuit is measured. Based on the initial contact resistance, establish and track the electrode state reference; After the main welding current is applied to the welding machine tool, the disturbance caused by the change in electrode state in the main welding signal is identified and decoupled using the electrode state reference. Based on the identification and decoupling results, the welding parameters of the welding machine tool are adjusted in a differentiated manner.

[0008] Secondly, a machine tool welding system for welding metal materials is provided, the system comprising: The detection module is used to inject a detection current into the welding circuit without causing the metal to melt during the pre-pressing stage of the electrode pressing of the welding machine tool onto the metal parts to be welded, and to measure the initial contact resistance of the welding circuit. The reference establishment and tracking module is used to establish and track the electrode state reference based on the initial contact resistance; The signal recognition and decoupling module is used to identify and decouple the disturbances caused by changes in electrode state in the main welding signal after the main welding current is applied to the welding machine tool, using the electrode state reference. The parameter adjustment module is used to make differentiated adjustments to the welding parameters of the welding machine tool based on the identification and decoupling results.

[0009] Compared with the prior art, the present invention has the following beneficial effects: By injecting a probe current and measuring the initial contact resistance during the pre-pressing stage of electrode bonding of the metal parts to be welded, an electrode state reference can be established and tracked. After the main welding current is applied, this electrode state reference is used to identify and decouple disturbances in the main welding signal caused by changes in electrode state, and welding parameters are adjusted differentially based on the identification and decoupling results. This method effectively solves the problem of welding signal ambiguity caused by batch material differences and electrode wear or contamination in the prior art, which makes it difficult for the system to accurately diagnose the root cause of the problem and make effective adaptive adjustments. By distinguishing disturbances caused by changes in electrode state, this application can avoid misjudging electrode problems as material problems, thereby avoiding incorrect parameter adjustments and effectively reducing spatter and weld defects. Attached Figure Description

[0010] Figure 1 This is a flowchart of the method of the present invention.

[0011] Figure 2 This is a schematic diagram of the system structure of the present invention.

[0012] In the diagram: 201, detection module; 202, benchmark establishment and tracking module; 203, signal identification and decoupling module; 204, parameter adjustment module. Detailed Implementation

[0013] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0014] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0015] Key terms involved in metal welding include: "Electrode": Typically made of copper alloy, it applies pressure and current to the metal parts to be welded, causing localized melting to form a weld joint. The electrode's geometry, material properties, and surface condition directly affect the weld quality. "Metal Part to be Welded": Refers to the metal component to be welded; its material type, thickness, and surface treatment (such as plating) significantly influence the welding process. "Welding Circuit": A closed circuit consisting of a power source, electrode, metal part to be welded, and connecting wires. The main welding current flows through this circuit, generating resistance heat. "Probe Current": A low-intensity current applied during the pre-pressing stage to measure contact resistance without causing metal melting, used to assess the initial contact state between the electrode and the metal part to be welded. "Initial Contact Resistance": The resistance value between the electrode and the metal part to be welded, measured during the pre-pressing stage, reflecting initial conditions such as the cleanliness and degree of pressure at the contact interface. "Electrode Condition Reference": A reference model or dataset established based on the initial contact resistance to characterize the electrode's health and contact characteristics. This reference is dynamically updated due to factors such as electrode wear and contamination. "Main welding current": A high-intensity current applied during the welding stage to locally melt the metal parts to be welded and form weld joints. "Main welding signal": Real-time electrical signals, such as voltage and current waveforms, collected by sensors during the application of the main welding current. These signals contain rich information about the welding process.

[0016] "Disturbance": Refers to abnormal fluctuations or deviations in the main welding signal caused by non-ideal factors (such as changes in electrode condition, material differences, etc.). "Identification and Decoupling": Refers to using algorithms and models to separate disturbances in the main welding signal from the normal signal and determine the specific source of the disturbance (e.g., electrode wear or material differences). "Welding Parameters": Includes welding current, energizing time, electrode pressure, etc. Precise control of these parameters is crucial to ensuring welding quality. "Differentiated Adjustment": Based on the results of identification and decoupling, targeted and refined adjustments are made to the welding parameters, rather than simple global adjustments.

[0017] The core of the metal welding method of this application lies in the precise perception, modeling and adaptive adjustment of the contact state between the electrode and the metal part to be welded during the welding process.

[0018] like Figure 1 The method shown is a machine tool welding method for welding metal materials. The method is implemented by the execution system of the welding machine tool and includes the following steps: S101. During the pre-pressing stage of electrode pressing of the metal parts to be welded on the welding machine tool, a probe current that does not cause metal melting is injected into the welding circuit, and the initial contact resistance of the welding circuit is measured. It should be noted that this step can be implemented in various ways. For example, a separate low-voltage constant current source can be used to apply a microampere-level probe current to the welding circuit. The intensity of this current is far below the threshold that would cause metal melting, ensuring that no thermal damage to the parts is caused during the measurement process. Simultaneously, by placing voltage sensors at or near the contact interface of the electrodes, the voltage drop generated when the probe current flows is measured, and the initial contact resistance is calculated using Ohm's law. Another approach is to utilize the low-power mode of the main welding power supply to apply a pulsed probe current for an extremely short time (e.g., a few milliseconds), simultaneously acquiring voltage and current data, and obtaining the initial contact resistance through data processing.

[0019] S102. Establish and track electrode state references based on initial contact resistance; It should be noted that electrode condition benchmarks can be established using various methods. One approach is to measure a series of initial contact resistance values ​​when the electrode is brand new or in a known good condition, and calculate their average or establish an initial resistance distribution range as a benchmark. As the number of welds increases, the initial contact resistance can be measured periodically or under specific conditions and compared with the initial benchmark. If the measured value deviates from the benchmark by more than a preset threshold, the electrode condition benchmark is updated to reflect the actual wear or contamination level of the electrode. For example, a moving average or exponentially weighted average method can be used to track the dynamic changes in the electrode condition benchmark.

[0020] S103. After the main welding current is applied to the welding machine tool, the disturbance caused by the change of electrode state in the main welding signal is identified and decoupled using the electrode state reference. It should be noted that during the application of the main welding current, main welding signals such as electrode voltage and welding current are acquired in real time. These signals contain rich process information, but may also be disturbed by changes in electrode state. To identify and decouple these disturbances, signal processing techniques can be employed. For example, the real-time acquired main welding signals can be compared with an "ideal" signal predicted based on an electrode state reference, and anomalies can be identified through differential or residual analysis. Another approach is to utilize a machine learning model, using the electrode state reference as one of the input features, to train the model to distinguish signal features caused by changes in electrode state from those caused by material differences or other factors. For example, a support vector machine (SVM) or neural network model can be trained, taking real-time signal features and the electrode state reference as input, and outputting the type and degree of disturbance.

[0021] S104. Based on the identification and decoupling results, make differentiated adjustments to the welding parameters of the welding machine tool.

[0022] It's important to note that once the disturbance caused by changes in electrode state is identified and decoupled, welding parameters can be adjusted specifically based on the nature and extent of the disturbance. For example, if electrode wear leads to an increased contact area and a decrease in current density, the amplitude of the main welding current can be appropriately increased or the energizing time extended to compensate for heat loss. If electrode surface contamination leads to increased contact resistance and localized overheating, the main welding current can be appropriately reduced or the energizing time shortened, and electrode re-grinding or replacement should be considered. This differentiated adjustment avoids a "one-size-fits-all" approach to parameter adjustments, making the welding process more precise and efficient. For instance, a parameter adjustment rule library can be established to find corresponding parameter adjustment strategies based on different types and degrees of disturbance.

[0023] The metal welding method of this application injects a probe current that does not cause metal melting into the welding circuit during the pre-pressing stage of electrode pressing against the metal parts to be welded, and measures the initial contact resistance of the welding circuit to obtain precise information on the initial contact state between the electrode and the metal parts to be welded. Based on this initial contact resistance, an electrode state reference is established and tracked, which can dynamically reflect changes in electrode condition such as wear and contamination. After the main welding current is applied, this electrode state reference is used to identify and decouple disturbances in the main welding signal caused by changes in electrode state. This means that the system can distinguish whether signal anomalies originate from problems with the electrode itself or from other factors such as batch differences in materials. Finally, based on the identification and decoupling results, welding parameters are adjusted differentially to achieve precise control of the welding process.

[0024] For example, on resistance spot welding production lines in automobile manufacturing, traditional methods often struggle to distinguish signal changes caused by electrode wear from those caused by minor batch variations in steel sheets. When electrode wear increases the contact area, the current density per unit area decreases, potentially leading to a smaller weld nugget size. Similarly, slightly lower conductivity of the steel sheet material can also produce similar signals. Traditional systems may fail to accurately diagnose this, resulting in incorrect parameter adjustments. For instance, they might misinterpret signal changes caused by electrode wear as material problems and increase the welding current accordingly. This not only fails to address the electrode wear issue but could also accelerate electrode deterioration or even cause spatter.

[0025] The method described in this application effectively solves the aforementioned problems. By measuring the initial contact resistance during the pre-pressure stage, the system can establish a "health record" for the electrodes. When the electrodes wear or become contaminated, the initial contact resistance will change predictably, thus updating the electrode condition reference. During the main welding process, if an anomaly occurs in the main welding signal, the system will first use the electrode condition reference to determine whether these anomalies are related to changes in the electrode condition. For example, if the electrode condition reference indicates severe electrode wear, then disturbances in the main welding signal related to a decrease in current density will be identified and decoupled as being caused by electrode wear. Once the source of the disturbance is determined, the system can make differentiated adjustments. For example, for electrode wear, the welding current can be appropriately increased or the energizing time extended to compensate for heat loss, while simultaneously prompting the operator to perform electrode regrinding or replacement; and if the signal anomaly is decoupled to material differences, the current waveform or pressure curve can be adjusted to adapt to the material characteristics. This precise identification and decoupling makes the adjustment of welding parameters more targeted, avoiding misjudgments and ineffective adjustments, and significantly improving welding quality and production efficiency.

[0026] As one embodiment of the present invention, after the main welding current is applied to the welding machine tool, the steps of identifying and decoupling the disturbance caused by the change in electrode state in the main welding signal using an electrode state reference include: Acquire transient electrical signals at the interface between the electrode and the metal part to be welded; It should be noted that the above step refers to acquiring instantaneous electrical parameters such as voltage, current, or resistance at the contact interface between the electrode and the metal part to be welded in real time through sensors or probes during the application of the main welding current. These transient electrical signals can reflect microscopic changes at the contact interface, such as micro-discharge, instantaneous adhesion, or rapid fluctuations in contact resistance.

[0027] Extracting dynamic features from transient electrical signals; It should be noted that the above step refers to processing and analyzing the acquired transient electrical signals to identify and quantify the dynamic information related to changes in electrode state. These dynamic characteristics may include signal amplitude, frequency, duration, waveform shape, or statistical properties, which can characterize the instantaneous instability and dynamic evolution of the contact state between the electrode and the metal part to be welded.

[0028] By combining electrode condition references and dynamic characteristics, the dynamic contact state between the electrode and the metal part to be welded is determined. It should be noted that the above step refers to comparing and comprehensively analyzing the real-time extracted dynamic features with a pre-established and tracked electrode state benchmark. The electrode state benchmark provides reference information about the electrode under ideal or stable conditions, while the dynamic features reveal instantaneous anomalies at the current contact interface. By combining the two, it is possible to accurately assess whether the contact between the electrode and the metal part to be welded is in a stable, unstable, adhered, or separated state.

[0029] Based on the judgment results, the disturbances caused by changes in electrode state in the main welding signal are identified and decoupled.

[0030] It should be noted that the above step refers to the ability to identify non-ideal components or noise in the main welding signal caused by changes in these contact states once the dynamic contact state between the electrode and the metal part to be welded is determined. For example, if unstable contact is identified, voltage or current disturbances caused by fluctuations in contact resistance can be identified and separated from the main welding signal, thereby obtaining a purer and more accurate welding process signal, providing a reliable basis for subsequent welding parameter adjustments.

[0031] This application's solution acquires transient electrical signals at the interface between the electrode and the metal part to be welded in real time after the main welding current is applied, and extracts dynamic features from these signals, thereby capturing subtle changes in the electrode contact state during the welding process. Because these transient electrical signals contain rich dynamic information, combined with a pre-established electrode state reference, the system can accurately determine the dynamic contact state between the electrode and the metal part to be welded. In this way, disturbances to the main welding signal caused by changes in electrode state (such as micro-discharge, transient adhesion, etc.) can be effectively identified and decoupled, preventing these disturbances from interfering with welding quality assessment and parameter control. Through the above technical solution, this application can more precisely monitor and analyze the interface state between the electrode and the metal part to be welded during the welding process, especially during the application of the main welding current. This enables the system to accurately identify and decouple disturbances to the main welding signal caused by changes in electrode state in real time, significantly improving the accuracy of welding process monitoring and the robustness of signal processing.

[0032] As one embodiment of the present invention, the step of extracting dynamic features from transient electrical signals includes: Multi-scale decomposition of transient electrical signals; It should be noted that the above step refers to decomposing the original transient electrical signal into multiple components within different scales or frequency ranges. This can be achieved using signal processing techniques such as wavelet transform, empirical mode decomposition (EMD), or variational mode decomposition (VMD). The aim is to effectively separate the signal features corresponding to different frequency components or different physical phenomena, laying the foundation for subsequent noise removal and feature extraction.

[0033] Separate the noise component from the signal component after multi-scale decomposition; It should be noted that the above step refers to distinguishing and eliminating noise components, mainly caused by random interference or non-target physical processes, from the components obtained by multi-scale decomposition, retaining the components carrying useful information (i.e., signal components). This can be achieved by setting thresholds, applying filters, or making judgments based on statistical characteristics. The purpose is to improve the signal-to-noise ratio of the signal and ensure the accuracy of subsequent analysis.

[0034] The separated signal components are reconstructed to obtain the denoised transient electrical signal; It should be noted that the above step refers to recombining the signal components after noise separation to form a purer transient electrical signal that better reflects the true dynamics of the contact interface between the electrode and the metal part to be welded. This denoised signal eliminates most of the interference, making the characteristics of key events such as micro-discharges or transient adhesion more prominent.

[0035] From the denoised transient electrical signal, the signal waveform features corresponding to micro-discharge or instantaneous adhesion are identified and quantified to obtain dynamic features.

[0036] It should be noted that micro-discharge refers to a tiny arc discharge phenomenon at the interface between the electrode and the metal part to be welded, caused by poor local contact or excessively high current density; transient adhesion refers to a brief, unexpected physical adhesion phenomenon that occurs between the electrode and the metal part to be welded during the welding process. These phenomena will exhibit specific waveform characteristics in transient electrical signals. For example, micro-discharge may manifest as sharp voltage or current pulses, while transient adhesion may cause sudden changes in resistance or voltage. By performing waveform analysis, pattern recognition, or feature extraction algorithms (such as peak detection, frequency analysis, duration measurement, etc.) on the denoised signal, these specific waveforms can be accurately identified, and their amplitude, frequency, duration, energy, and other parameters can be quantified, thereby forming dynamic characteristics characterizing the dynamic contact state between the electrode and the metal part to be welded.

[0037] The proposed solution employs refined multi-scale decomposition and denoising processing of transient electrical signals. This effectively extracts key signal components related to electrode state changes from the complex original signal and suppresses interference from irrelevant noise. This in-depth signal processing enables the accurate identification and quantification of signal waveform features reflecting the dynamic characteristics of the contact interface, such as micro-discharges or transient adhesion. These precisely extracted dynamic features more realistically and sensitively reflect the instantaneous contact instability between the electrode and the metal part to be welded, thus providing high-quality input data for subsequent assessment of the dynamic contact state between the electrode and the metal part. This significantly improves the accuracy of identifying and decoupling disturbances caused by electrode state changes in the main welding signal.

[0038] Through the above technical solution, this application overcomes the limitation of traditional methods in accurately extracting effective signal features in complex noise environments. By introducing multi-scale decomposition and denoising processing, the dynamic features extracted from transient electrical signals are ensured to have higher purity and reliability, thereby making the judgment of the dynamic contact state between the electrode and the metal part to be welded more accurate.

[0039] As one embodiment of the present invention, the step of determining the dynamic contact state between the electrode and the metal part to be welded, by combining electrode state reference and dynamic characteristics, includes: If the instantaneous contact instability indicated by the dynamic feature reaches a preset threshold, the dynamic contact state between the electrode and the metal part to be welded is judged as an unstable contact state. If the instantaneous contact instability indicated by the dynamic characteristics does not reach the preset threshold, the dynamic contact state between the electrode and the metal part to be welded is determined based on the electrode state reference.

[0040] The transient contact instability indicated by dynamic characteristics refers to the degree of fluctuation or abnormality of the instantaneous state of the contact interface, quantified by processing transient electrical signals (e.g., through multi-scale decomposition, denoising, and identification of signal waveform characteristics corresponding to micro-discharges or transient adhesion). This instability can manifest as drastic changes in parameters such as the amplitude, frequency, or duration of the transient electrical signal. The preset threshold is a critical value pre-set based on empirical data, experimental results, or specific welding process requirements, used to define what degree of transient contact instability should be considered an abnormal state requiring immediate attention. When the transient contact instability indicated by dynamic characteristics reaches or exceeds the preset threshold, it indicates a significant transient abnormality at the contact interface between the electrode and the metal part to be welded, such as severe micro-discharge or adhesion tendency. In this case, regardless of the electrode condition baseline, the dynamic contact state between the electrode and the metal part to be welded is preferentially judged as an unstable contact state. This judgment mechanism ensures rapid response and priority handling of sudden, high-risk contact anomalies. Conversely, if the instantaneous contact instability indicated by the dynamic characteristics does not reach the preset threshold, i.e., the instantaneous contact state is relatively stable or the degree of abnormality is low, the judgment logic reverts to relying on the electrode state reference. The electrode state reference is established and tracked based on the initial contact resistance during the pre-pressure stage, reflecting the long-term or average contact characteristics of the electrode under relatively stable conditions. At this time, judging the dynamic contact state based on the electrode state reference can utilize its sensitivity to long-term changes such as electrode wear and oxidation to provide a more comprehensive assessment of the contact state.

[0041] This application's solution effectively addresses the limitations of traditional methods in handling transiently unstable contact problems by introducing a hierarchical judgment logic. Specifically, when the transient contact instability indicated by dynamic characteristics reaches a preset threshold, the system immediately identifies and prioritizes it as an unstable contact state. This allows for rapid and sensitive detection of high-risk events such as sudden micro-discharges or transient adhesion that may occur during welding. This priority judgment mechanism avoids response lag or misjudgment that may result from over-reliance on electrode state benchmarks, ensuring timely intervention in critical transient anomalies. Simultaneously, when the transient contact instability does not reach the preset threshold, the solution reverts to using electrode state benchmarks for judgment. Electrode state benchmarks reflect the gradual changes in electrode characteristics such as long-term wear and oxidation, providing a more macroscopic and stable perspective for contact state assessment. This combined approach enables the system to react quickly to transient anomalies while also considering the long-term health of the electrodes, thereby achieving a comprehensive and accurate judgment of the dynamic contact state between the electrode and the metal part to be welded.

[0042] In some preferred embodiments, a specific example is illustrated below. Assume that during the application of the main welding current, the system continuously acquires transient electrical signals at the interface between the electrode and the metal part to be welded. By processing these signals in real time, such as using wavelet analysis for multi-scale decomposition and extracting the characteristic waveforms of the micro-discharge, a transient contact instability index is quantified. If at some moment, this index (e.g., the instantaneous peak voltage or duration of the micro-discharge) suddenly spikes and reaches a preset threshold (e.g., exceeding 50 mV or lasting longer than 10 microseconds), even if the electrode state reference indicates that the electrode wear is still within an acceptable range, the system will immediately classify the current dynamic contact state as an "unstable contact state." Conversely, if at another moment, the transient contact instability index remains below the preset threshold, but the electrode state reference (e.g., by tracking the trend of the initial contact resistance) shows that the electrode contact resistance has slowly risen to the upper limit of its normal range, indicating that the electrode may have slight oxidation or wear. In this case, the system will classify the dynamic contact state as "deteriorated contact state" or "slightly unstable" based on the electrode state reference, rather than "unstable contact state." This tiered judgment mechanism ensures that appropriate and timely responses can be made to contact anomalies of different natures, thereby guiding subsequent adjustments to welding parameters. For example, for the former, it may be necessary to immediately reduce the welding current or shorten the welding time to avoid spatter, while for the latter, it may be necessary to fine-tune the welding pressure or energy to compensate for electrode wear.

[0043] As one embodiment of the present invention, the step of differentially adjusting the welding parameters of the welding machine tool based on the identification and decoupling results includes: The adjustment targets for welding parameters are determined based on the identification and decoupling results; It should be noted that determining the adjustment target for welding parameters refers to quantifying the deviation between the current welding state and the ideal welding state based on the identification and decoupling of disturbances caused by changes in electrode state in the main welding signal. Based on this deviation, the desired direction and magnitude of welding parameter adjustments are then set. For example, if electrode wear is detected as causing increased contact resistance, the adjustment target might be to increase the welding current or extend the welding time.

[0044] The adjustment target is broken down into multiple adjustment components; It's important to note that decomposing the adjustment target into multiple adjustment components means breaking down a large, overall adjustment task into a series of controllable, small-amplitude adjustments. This decomposition can be linear or non-linear, aiming to provide finer control granularity for real-time feedback and correction during the adjustment process. For example, if the adjustment target is to increase the welding current by 200A, it can be decomposed into 20 adjustment components, each increasing by 10A.

[0045] The adjustment components are applied sequentially, and the instantaneous electrical signal at the contact interface between the electrode and the metal part to be welded is monitored in real time when each adjustment component is applied. It should be noted that the above step refers to immediately acquiring and analyzing the instantaneous voltage, current, or resistance signals of the electrode-workpiece contact area after each application of an adjustment component. These instantaneous electrical signals can sensitively reflect microscopic changes at the contact interface, such as the presence of micro-discharges and the stability of the contact area.

[0046] The stability of the contact interface is evaluated based on the instantaneous electrical signal; It should be noted that this step refers to determining whether the physical and electrical state of the current contact interface between the electrode and the metal part to be welded is within the expected stable range by analyzing the characteristics of the monitored instantaneous electrical signal, such as its amplitude, frequency, waveform, or duration. For example, abnormal voltage spikes or current fluctuations may indicate contact instability.

[0047] If the stability of the contact interface is lower than the preset level, the application method of the subsequent adjustment components will be adjusted to differentiate the welding parameters.

[0048] It should be noted that the above step refers to the system's ability to immediately and dynamically adjust the remaining adjustment components when the evaluation results indicate poor contact interface stability. This adjustment can include changing the application step size of subsequent components (e.g., reducing the step size for finer adjustments), the application rate (e.g., slowing down the adjustment speed to give the system more time to respond), or the adjustment sequence, or even pausing the adjustment, to avoid introducing new welding defects due to unstable contact conditions. This adaptive adjustment mechanism ensures that the optimization process of welding parameters is always carried out under controlled and stable conditions.

[0049] This application's solution decomposes the overall adjustment target of welding parameters into multiple controllable adjustment components and introduces a real-time monitoring and stability assessment mechanism for the instantaneous electrical signal at the contact interface between the electrode and the metal part to be welded after each component is applied. This effectively solves the problem of insufficient response to transient contact instability that may exist in traditional differentiated adjustments. When signs of decreased stability appear at the contact interface, the system can immediately provide feedback and adjust the subsequent parameter application strategy, avoiding continued preset parameter adjustments in an unstable state, thereby preventing welding defects. This closed-loop, adaptive adjustment method allows the welding process to dynamically adapt to contact state fluctuations caused by various factors such as electrode wear and workpiece surface changes, ensuring the accuracy and effectiveness of welding parameter adjustments.

[0050] As one embodiment of the present invention, the step of adjusting the application method of the subsequent adjustment component includes: Obtain the stability of the contact interface; It should be noted that the above step refers to real-time monitoring of the instantaneous electrical signal at the interface between the electrode and the metal part to be welded during the welding parameter adjustment process, and evaluating the stability of the interface based on this signal. This stability assessment can be based on the waveform characteristics of the instantaneous electrical signal, such as amplitude, frequency, or duration, or by performing statistical analysis on these characteristics, such as calculating their mean, variance, or kurtosis, and comparing them with a preset stability range or statistical control value. The purpose is to provide a quantitative indicator to guide subsequent parameter adjustments.

[0051] Adjust the application step size or application rate of subsequent adjustment components based on the stability of the contact interface.

[0052] It should be noted that the above step can be understood as dynamically changing the application method of the welding parameter adjustment components that have not yet been applied based on the real-time obtained contact interface stability assessment results. For example, when the contact interface stability assessment results show low stability, the application step size of subsequent adjustment components can be reduced to make the adjustment more precise, or the application rate can be reduced to give the system more time to respond and stabilize. Conversely, when the contact interface stability is high, the application step size or application rate can be appropriately increased to speed up the adjustment process. The purpose is to achieve adaptive and precise control of welding parameters, ensuring the continuous stability of the contact interface throughout the welding process.

[0053] This application's solution introduces a real-time acquisition and feedback mechanism for the stability of the contact interface, enabling dynamic adjustment of subsequent adjustment components based on the actual welding conditions. Specifically, when the contact interface stability is assessed as below a preset level, this indicates potential uncertainties or defects in the current welding conditions, such as electrode wear or oxidation of the metal parts to be welded. In this case, by adjusting the application step size or rate of subsequent adjustment components—for example, reducing the step size or speed—the adjustment process of welding parameters can be made smoother and more precise, avoiding the introduction of new instabilities due to large-scale adjustments at once. This adaptive adjustment based on real-time feedback effectively suppresses or eliminates disturbances caused by changes in electrode state, thereby ensuring the purity of the main welding signal and the stability of the welding process. It is precisely this refined and responsive adjustment strategy that allows the welding process to better adapt to various complex and dynamic working conditions. Through the above technical solution, this application achieves refined and adaptive control of the welding parameter adjustment process.

[0054] As one embodiment of the present invention, the step of evaluating the stability of the contact interface based on the instantaneous electrical signal includes: To obtain the amplitude, frequency, or duration of waveform characteristics in an instantaneous electrical signal; Compare the amplitude, frequency, or duration to a preset stability range; Based on the comparison results, the stability of the contact interface is evaluated.

[0055] The instantaneous electrical signal is a real-time monitoring signal at the contact interface between the electrode and the metal part to be welded, obtained by sequentially applying adjustment components. The amplitude, frequency, and duration of the waveform characteristics are key parameters of this instantaneous electrical signal, reflecting the dynamic behavior and stability of the contact interface at the microscopic level. For example, amplitude indicates the magnitude of change in contact resistance, frequency reflects the rate of fluctuation in the contact state, and duration indicates the duration of unstable events. The preset stability range is determined based on extensive experimental data, theoretical models, or empirical knowledge, and is used to define the allowable variation range of the instantaneous electrical signal waveform characteristics under normal and stable contact conditions. By comparing the real-time acquired waveform characteristics with this preset stability range, the current stability of the contact interface can be quantitatively determined.

[0056] This application's solution achieves precise evaluation of contact interface stability by acquiring the amplitude, frequency, or duration of waveform characteristics in a transient electrical signal and comparing them with a preset stability range. When minute changes occur at the contact interface between the electrode and the metal part to be welded, such as micro-discharge, transient adhesion, or fluctuations in contact resistance, these events are immediately reflected in the waveform characteristics of the transient electrical signal. By real-time monitoring and quantification of these characteristics (such as amplitude, frequency, and duration) and comparing them with a preset stability range, abnormal states of the contact interface can be identified promptly and accurately. This quantitative evaluation method based on specific waveform characteristics makes the judgment of contact interface stability no longer a vague qualitative description, but has a clear quantitative standard, thus providing a reliable basis for subsequent welding parameter adjustments. Through the above technical solution, a more refined and quantitative evaluation of the stability of the contact interface between the electrode and the metal part to be welded can be achieved.

[0057] As one embodiment of the present invention, the step of evaluating the stability of the contact interface includes: To obtain the amplitude, frequency, or duration of waveform characteristics in an instantaneous electrical signal; It should be noted that the amplitude of waveform characteristics in instantaneous electrical signals can refer to the peak, trough, or peak-to-peak value of the signal, reflecting the intensity or energy level of the electrical signal. Frequency can refer to the repetition rate of a specific oscillation mode in the signal, such as the occurrence frequency of micro-discharge events, which may indicate the dynamic behavior of the contact interface. Duration can refer to the duration of a specific event (such as a micro-discharge pulse), reflecting the persistence of the unstable state. Once these waveform characteristics are acquired, they can provide basic data for subsequent stability assessments.

[0058] Perform statistical analysis on amplitude, frequency, or duration; It should be noted that the above step can be understood as quantifying the distribution characteristics of these features within a certain time window or across multiple sampling points. For example, calculating the mean can reflect the average level of the feature; calculating the variance can reflect the degree of fluctuation or dispersion of the feature, with a larger variance usually indicating a more unstable contact interface; calculating the kurtosis can reflect the sharpness or tail thickness of the feature distribution, which helps identify outliers or sudden events. Through statistical analysis, the dynamic characteristics of the contact interface can be captured more comprehensively and accurately, reducing the impact of instantaneous noise or random fluctuations on the evaluation results.

[0059] Compare the statistical analysis results with the preset statistical control values; It should be noted that the above step refers to comparing the calculated statistical measures such as mean, variance, or kurtosis with pre-set thresholds or control ranges. These preset statistical control values ​​are usually determined based on a large amount of experimental data or empirical knowledge, representing the normal fluctuation range of waveform characteristics under stable contact interfaces. The purpose is to provide an objective judgment standard to distinguish between normal and abnormal fluctuations, thereby accurately identifying the stability state of the contact interface.

[0060] Based on the comparison results, the stability of the contact interface is evaluated.

[0061] This application's solution overcomes the limitations of relying solely on instantaneous values ​​for evaluation by statistically analyzing the waveform characteristics of instantaneous electrical signals. Because minute, transient contact instabilities may exist at the contact interface during welding, these instabilities may manifest as subtle fluctuations in the amplitude, frequency, or duration of the electrical signal. If only simple instantaneous comparisons are used, these subtle fluctuations may be ignored or misjudged as noise. By introducing statistical quantities such as mean, variance, or kurtosis, the overall distribution and variability of these waveform characteristics can be quantified, thereby more sensitively capturing potential unstable trends or abnormal behaviors at the contact interface. For example, an increase in variance may indicate drastic fluctuations in contact resistance, while anomalies in kurtosis may reveal frequent occurrences of micro-discharge or transient adhesion events. This statistical method makes the assessment of contact interface stability more comprehensive and robust, and can more accurately reflect the actual contact state.

[0062] The above technical solution significantly improves the accuracy and robustness of the stability assessment of the contact interface between the electrode and the metal part to be welded. By employing statistical analysis of waveform characteristics, the interference of transient noise can be effectively filtered out, and subtle disturbances caused by changes in electrode state can be identified more accurately. This allows for differentiated adjustments to welding parameters based on more reliable contact state assessment results, thereby optimizing the control precision of the welding process, reducing welding defects, and ultimately improving welding quality and production efficiency.

[0063] As one embodiment of the present invention, the step of determining the adjustment target of welding parameters based on the identification and decoupling results includes: Based on the identification and decoupling results, calculate the deviation between the identification and decoupling results and the ideal state; Based on the deviation, determine the adjustment target.

[0064] Specifically, the adjustment target is the direction and magnitude of adjustments to welding parameters (e.g., welding current, welding voltage, pressure, welding time, etc.) based on the calculated deviations, in order to bring the welding process back to or closer to the ideal state. For example, if the deviation indicates that the electrode contact resistance is too high, the adjustment target might be set to increase the pressure or slightly increase the welding current; if the deviation indicates the presence of micro-discharge or momentary adhesion, the adjustment target might be set to adjust the current waveform or reduce the current rise rate. The determination of the adjustment target can be based on preset control strategies, lookup tables, fuzzy logic, or machine learning models to ensure the accuracy and effectiveness of the adjustment.

[0065] The proposed solution first quantifies the deviation between the identification and decoupling results and the ideal state, thus providing a precise basis for subsequent welding parameter adjustments. It is precisely this clear definition of the difference between the actual and ideal states that ensures parameter adjustments are no longer blind or empirical, but based on real-time data and quantitative analysis. This mechanism ensures that the determined adjustment targets directly and effectively address disturbances caused by changes in electrode state, thereby achieving refined control of the welding process.

[0066] like Figure 2 The machine tool welding system shown includes: The detection module 201 is used to inject a detection current into the welding circuit without causing the metal to melt during the pre-pressing stage of the electrode pressing of the metal parts to be welded on the welding machine tool, and to measure the initial contact resistance of the welding circuit. The reference establishment and tracking module 202 is used to establish and track the electrode state reference based on the initial contact resistance; The signal recognition and decoupling module 203 is used to identify and decouple the disturbance caused by the change of electrode state in the main welding signal after the main welding current is applied to the welding machine tool, using the electrode state reference. The parameter adjustment module 204 is used to make differentiated adjustments to the welding parameters of the welding machine tool based on the identification and decoupling results.

[0067] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A machine tool welding method for welding metal materials, characterized in that, The method is implemented by the execution system of the welding machine tool, and the method includes the following steps: During the pre-pressing stage of electrode pressing of the metal parts to be welded on the welding machine tool, a probe current that does not cause metal melting is injected into the welding circuit, and the initial contact resistance of the welding circuit is measured. Based on the initial contact resistance, establish and track the electrode state reference; After the main welding current is applied to the welding machine tool, the disturbance caused by the change in electrode state in the main welding signal is identified and decoupled using the electrode state reference. Based on the identification and decoupling results, the welding parameters of the welding machine tool are adjusted in a differentiated manner.

2. The machine tool welding method for welding metal materials according to claim 1, characterized in that, The step of identifying and decoupling disturbances caused by changes in electrode state in the main welding signal after applying the main welding current to the welding machine tool, using the electrode state reference, includes: Acquire transient electrical signals at the interface between the electrode and the metal part to be welded; Extract dynamic features from the transient electrical signal; By combining the electrode state reference and the dynamic characteristics, the dynamic contact state between the electrode and the metal part to be welded is determined. Based on the judgment results, the disturbances caused by changes in electrode state in the main welding signal are identified and decoupled.

3. The machine tool welding method for welding metal materials according to claim 2, characterized in that, The step of extracting dynamic features from the transient electrical signal includes: The transient electrical signal is decomposed into multiple scales. Separate the noise component from the signal component after the multi-scale decomposition; The separated signal components are reconstructed to obtain the denoised transient electrical signal; From the denoised transient electrical signal, the signal waveform features corresponding to micro-discharge or instantaneous adhesion are identified and quantified to obtain the dynamic features.

4. The machine tool welding method for welding metal materials according to claim 2, characterized in that, The step of determining the dynamic contact state between the electrode and the metal part to be welded, by combining the electrode state reference and the dynamic characteristics, includes: If the instantaneous contact instability indicated by the dynamic feature reaches a preset threshold, the dynamic contact state between the electrode and the metal part to be welded is judged as an unstable contact state. If the instantaneous contact instability indicated by the dynamic feature does not reach the preset threshold, the dynamic contact state between the electrode and the metal part to be welded is determined according to the electrode state benchmark.

5. The machine tool welding method for welding metal materials according to claim 1, characterized in that, The step of differentially adjusting the welding parameters of the welding machine tool based on the identification and decoupling results includes: The target for adjusting the welding parameters is determined based on the identification and decoupling results. The adjustment target is decomposed into multiple adjustment components; The adjustment components are applied sequentially, and the instantaneous electrical signal at the contact interface between the electrode and the metal part to be welded is monitored in real time when each adjustment component is applied. The stability of the contact interface is evaluated based on the instantaneous electrical signal. If the stability of the contact interface is lower than the preset level, the application method of the subsequent adjustment components will be adjusted to differentiate the welding parameters.

6. The machine tool welding method for welding metal materials according to claim 5, characterized in that, The steps for adjusting the application method of subsequent adjustment components include: Obtain the stability of the contact interface; Based on the stability of the contact interface, adjust the application step size or application rate of the subsequent adjustment component.

7. A machine tool welding method for welding metal materials according to claim 5, characterized in that, The step of evaluating the stability of the contact interface based on the instantaneous electrical signal includes: Obtain the amplitude, frequency, or duration of the waveform characteristics in the instantaneous electrical signal; The amplitude, frequency, or duration is compared with a preset stability range; The stability of the contact interface is evaluated based on the comparison results.

8. A machine tool welding method for welding metal materials according to claim 5, characterized in that, The steps for evaluating the stability of the contact interface include: Obtain the amplitude, frequency, or duration of the waveform characteristics in the instantaneous electrical signal; Perform statistical analysis on the amplitude, frequency, or duration; The statistical analysis results are compared with preset statistical control values; The stability of the contact interface is evaluated based on the comparison results.

9. A machine tool welding method for welding metal materials according to claim 5, characterized in that, The step of determining the adjustment target of welding parameters based on the identification and decoupling results includes: Based on the identification and decoupling results, calculate the deviation between the identification and decoupling results and the ideal state; The adjustment target is determined based on the deviation.

10. A machine tool welding system for welding metal materials, used to perform a machine tool welding method for welding metal materials as described in any one of claims 1-9, characterized in that, The system includes: The detection module is used to inject a detection current into the welding circuit without causing the metal to melt during the pre-pressing stage of the electrode pressing of the welding machine tool onto the metal parts to be welded, and to measure the initial contact resistance of the welding circuit. The reference establishment and tracking module is used to establish and track the electrode state reference based on the initial contact resistance; The signal recognition and decoupling module is used to identify and decouple the disturbances caused by changes in electrode state in the main welding signal after the main welding current is applied to the welding machine tool, using the electrode state reference. The parameter adjustment module is used to make differentiated adjustments to the welding parameters of the welding machine tool based on the identification and decoupling results.

Images