A method for matching parameters of high-speed welding of composite magnetic field assisted pulse GMAW

Abstract

The present application relates to the technical field of gas shielded welding with a consumable electrode, and particularly relates to a method for matching parameters of high-speed welding of composite magnetic field assisted pulse GMAW, which comprises: S100, performing pulse waveform optimization configuration: configuring pulse welding waveform parameters, setting welding current to sequentially experience base value stage, rising stage, peak value stage, stage of falling to intermediate current, intermediate current maintaining stage and stage of falling back to base value current in a pulse period, so as to obtain optimal welding pulse waveform meeting stable droplet transfer mode of one pulse one drop; the present application solves the problem of hump and undercut defects caused by severe backward flow of liquid metal and insufficient transverse heat diffusion in high-speed pulse GMAW welding by constructing a linear collaborative matching model of multi-step pulse waveform and alternating excitation parameters, and introducing a double closed-loop control mechanism based on arc-droplet space dynamics and surface temperature field form characteristics of the molten pool.

Description

Technical Field

[0001] This invention relates to the field of gas metal arc welding technology, specifically to a method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse. Background Technology

[0002] Gas metal arc welding (GMAW) plays a vital role in modern industrial manufacturing due to its high deposition rate and low cost. With ever-increasing demands for production efficiency, high-speed welding, reaching speeds of 1.0 m / min or even 2.0 m / min and above, has become a significant development trend. However, the high-speed moving heat source severely disrupts the fluid dynamics and thermodynamic balance of the molten pool system, easily inducing serious weld formation defects such as undercut and humps. To suppress these defects, introducing an external alternating composite magnetic field to deflect the arc and control the droplet behavior has become an effective method.

[0003] Existing external magnetic field-assisted GMAW technology is mainly designed for constant current welding processes. However, when applied to high-speed pulsed GMAW (GMAW-P) welding, which requires more precise heat input control, it reveals serious technical bottlenecks. The core problem lies in the fact that the arc plasma in pulsed GMAW exhibits extremely strong nonlinear dynamic characteristics during peak, base, and transient transition phases. Existing control methods have failed to establish a quantitative multi-physics coupling model between pulse waveform parameters (such as current amplitude and timing at each stage) and alternating magnetic field excitation parameters (such as excitation current and frequency). Under high-speed welding conditions, on the one hand, the liquid metal inside the molten pool is subjected to arc dynamic pressure and moving shear force, generating a huge backward flow momentum; on the other hand, heat is highly concentrated in the weld center, making it difficult to effectively conduct to the base material on both sides. Due to the lack of a scientific "magnetic-electric" fundamental matching algorithm, existing technologies can only rely on manual experience to statically set excitation parameters, failing to match the applied alternating Lorentz force with the rapidly changing mechanical state of the arc plasma at the microscopic scale. Furthermore, existing systems lack real-time closed-loop feedback mechanisms for microscopic dynamics (such as the spatial deflection angle between the arc and the molten droplet) and macroscopic thermodynamics (such as the temperature gradient and distribution morphology of the molten pool surface). When welding conditions are fine-tuned or the thermal accumulation state of the molten pool changes, the system cannot adaptively adjust the forward thrust of the magnetic field to suppress the backward surge of the liquid metal (causing a hump), nor can it dynamically adjust the lateral oscillation frequency to replenish the wetting heat of the liquid metal at the weld toe (causing undercut). As a result, the forming quality of pulsed GMAW under high-speed welding conditions is extremely unstable and cannot meet the requirements of practical industrial applications. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for matching parameters in high-speed GMAW welding with a composite magnetic field-assisted pulse, comprising the following steps: S100, perform pulse waveform optimization configuration: configure pulse welding waveform parameters, set the welding current to go through the base value stage, rising stage, peak stage, falling to the middle current stage, middle current holding stage and falling back to the base value current stage in one pulse cycle, so as to obtain the optimal welding pulse waveform that meets the stable droplet transfer mode of one pulse and one droplet. S200, Establish basic collaborative matching of excitation parameters: Based on the optimal welding pulse waveform, obtain the current welding speed and / or the effective value of welding current, and calculate and set the initial excitation current and initial excitation frequency of the composite magnetic field generator based on the preset linear matching relationship model; wherein, the composite magnetic field generator outputs two alternating excitation current waveforms with phase difference. S300 performs closed-loop control of arc and droplet dynamics: during the welding process, it acquires arc oscillation images and droplet transition images in real time, and calculates and extracts the arc oscillation speed and the arc dwell time at the weld edge; When the arc oscillation speed is detected to deviate from the preset target speed range, the excitation current is dynamically adjusted until the arc oscillation speed returns to the target speed range. When the residence time of the electric arc at the weld edge is detected to deviate from the preset target time range, the excitation frequency is dynamically adjusted until the residence time is restored to the target time range; S400 performs molten pool temperature field feedback optimization control: During the welding process, the temperature field distribution image of the upper surface of the molten pool is continuously acquired, the temperature distribution data along the weld centerline direction is extracted to calculate the temperature gradient, and the temperature distribution pattern along the direction perpendicular to the weld is extracted. When the temperature gradient is greater than a set threshold, it is determined that the backward flow velocity of the liquid metal is too high, and the excitation current is increased to enhance the arc tilt angle. When the temperature distribution pattern along the direction perpendicular to the weld is unimodal, it is determined that the heat has not diffused sufficiently to both sides of the weld. The excitation frequency is increased to enhance the heat input effect of the arc oscillation on the weld edge until the temperature distribution pattern transforms into a bimodal pattern.

[0005] Furthermore, in step S100, the pulse welding waveform parameters satisfy the following distribution setting in the time dimension: The duration of the rising phase is 0.4-0.8 ms; The duration of the peak phase is 0.7-1.2 ms; The duration of the drop to the intermediate current stage is 0.3-0.6 ms; The duration of the intermediate current holding phase is 1.2-1.8 ms; The duration of the current drop back to the base value phase is 0.3-0.6 ms; The duration of the base value phase is 1.5-2.1 ms.

[0006] Furthermore, the optimal welding pulse waveform is further limited in terms of current amplitude and its functional control as follows: The peak current corresponding to the peak phase is configured to be 480-520A to ensure that molten droplets are formed at the tip of the welding wire at the top of the peak phase. The intermediate current corresponding to the intermediate current holding stage is configured to be 110-130A, in order to maintain the arc energy after the peak current drops, so as to achieve the smooth shedding and transition of the molten droplet; The base current corresponding to the base phase is configured to be 60-80A to maintain the basic arc and prepare for the arc initiation of the next pulse cycle.

[0007] Furthermore, in step S200, the basic physical characteristics of the dual-path alternating excitation current waveform with a phase difference are set as follows: The phase difference between the two excitation current waveforms is fixed at 180°, the duty cycle is 50%, and the base current is 0A. Meanwhile, the excitation current of the composite magnetic field generator is adjustable in the range of 5-30A, and the excitation frequency is adjustable in the range of 15-40Hz.

[0008] Furthermore, based on a preset linear matching relationship model, the initial excitation current and initial excitation frequency of the composite magnetic field generator are calculated and set, and its internal dynamic matching logic specifically satisfies the following mathematical rules: When the obtained welding speed v ≥ 1.8 m / min, the initial excitation current Ie is set to follow a linear evolution relationship with the welding speed v: The slope parameter k1 is selected as 25-35A·min / m, and the intercept parameter b1 is selected as -25 to -15A. Simultaneously, the initial excitation frequency f is set to have a linear relationship with the welding speed v: k3 is selected as b3 is selected as -30 to -20 Hz; When the effective value of the welding current is obtained At that time, set the initial excitation current. There is a current coupling relationship between the welding current and the effective value I. The coupling coefficient k2 is selected as 0.1-0.3A / A, and the bias constant b2 is selected as -50 to -10A.

[0009] Furthermore, in step S300, the dynamic quantitative evaluation and control rules for the arc oscillation speed and dwell time are specifically as follows: The arc oscillation speed is obtained by determining the time interval between the maximum leftward deviation and the maximum rightward deviation of the arc. Calculate the arc oscillation radius during this time interval. The arc oscillation velocity can be calculated using this formula:

[0010] The preset target speed range is set to 2.0-3.0° / ms; The method for obtaining the dwell time of the electric arc at the edge of the weld is as follows: extract the number of continuously acquired image frames when the electric arc swings to the position of the maximum deflection angle, and multiply it by the time interval of a single frame; wherein, the preset target time range is set to 18-26ms.

[0011] Furthermore, step S300 also includes a coordinated control sub-step based on the deflection angle difference: By extracting features from the acquired arc oscillation images and droplet transition images, the arc deflection angle between the arc axis and the welding wire axis, and the droplet deflection angle between the droplet centerline and the welding wire axis are calculated and determined respectively. The target range for the arc deflection angle θa is set to 15-25°, and the target range for the droplet deflection angle θd is set to 5-15°. The difference between the two deflection angles is then calculated. Limited to the 4-8° co-encapsulation range, among which ; When the detected arc deflection angle θa or droplet deflection angle When the deviation deviates from the corresponding target range, a dual-variable linkage adjustment mechanism is triggered, that is, the excitation current and the excitation frequency are adjusted synchronously until the deflection angle and its difference return to the preset range.

[0012] Furthermore, in step S400, the control logic for extracting temperature distribution data and performing closed-loop optimization is based on the decoupling of multidimensional characteristics of the heat flow in the molten pool space, specifically including the following hierarchical control rules: The grayscale images of the upper surface of the molten pool are continuously acquired by an infrared thermal imager and converted into actual temperature images based on the blackbody radiation law. Longitudinal heat flow suppression rule: Extract temperature distribution data along the weld centerline (longitudinal direction) and calculate the temperature gradient. When the temperature gradient If the backward flow velocity of the liquid metal is too high, the excitation current is increased by a compensation step of 2-5A to enhance the arc tilt angle and thus suppress the backward flow. Lateral heat flow broadening rule: Extract the temperature distribution pattern along the direction perpendicular to the weld, i.e., the lateral direction; when the temperature distribution pattern is a single peak with the highest temperature in the center, it is determined that the heat has not diffused sufficiently to both sides of the weld. Increase the excitation frequency with a compensation step of 3-8Hz, and use enhanced arc oscillation to promote edge heat input until the lateral temperature distribution pattern evolves into a double peak that deviates from the center of the molten pool. Melt pool size coordinated control rule: real-time monitoring of melt pool length When the length of the molten pool is detected If it is determined that the liquid metal has severely accumulated backward at the tail of the molten pool, the excitation current is increased by 2-8A and the excitation frequency is increased by 5-10Hz to forcibly reduce the length of the molten pool.

[0013] Furthermore, step S400 also includes a built-in excitation parameter self-optimization program to cope with extreme heat loss conditions, specifically: During the temperature field feedback control process, the lowest temperature at the center of the molten pool and the highest temperature on both sides of the weld are monitored simultaneously. When the lowest temperature at the center of the molten pool is detected to be less than 1400°C and the highest temperature on both sides of the weld is less than 1500°C, the excitation parameter self-optimization program is triggered.

[0014] Furthermore, the excitation parameter self-optimization program performs a global two-dimensional scan search within the excitation current space of 10-30A and the excitation frequency space of 15-35Hz, while maintaining the constraints of fixed wire extension length, shielding gas flow rate, and shielding gas composition. It constructs an objective function with the goal of maximizing weld width and optimizing weld surface quality, and solves and updates the optimal combination of excitation current and optimal excitation frequency under the current process conditions.

[0015] Beneficial effects This invention effectively solves the problems of hump and undercut defects caused by intense backward flow of liquid metal and insufficient lateral thermal diffusion in high-speed pulsed GMAW welding by constructing a linear cooperative matching model of multi-step pulse waveform and alternating excitation parameters, and introducing a dual closed-loop control mechanism based on arc-droplet spatial dynamics and the morphological characteristics of the molten pool surface temperature field. The invention achieves precise spatiotemporal coupling of the transient gradient of pulse energy and alternating Lorentz force, ensuring a high-speed and stable droplet transition. Simultaneously, through bidirectional dynamic monitoring and adaptive feedback of the longitudinal temperature gradient and lateral temperature distribution, it achieves precise constraint and intervention of the molten pool flow field and thermal field, significantly improving the self-optimizing ability, anti-interference ability, and final weld formation quality of the high-speed welding process. Attached Figure Description

[0016] Figure 1 This is the overall main flowchart of the parameter matching control method of the present invention; Figure 2 This is a diagram of the system collaborative control architecture of the present invention; Figure 3 This is a flowchart illustrating the basic collaborative matching dynamic calculation logic of the present invention. Figure 4 This is a flowchart of the dynamic closed-loop control multivariable linkage of the present invention; Figure 5 This is a flowchart of the temperature field hierarchical feedback and extreme condition self-optimization process of the present invention. Detailed Implementation

[0017] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but includes other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] The present invention will now be described in further detail with reference to the accompanying drawings: Example

[0020] like Figure 1-5 As shown, a method for matching parameters in high-speed GMAW welding with a composite magnetic field-assisted pulse includes the following steps: S100, perform pulse waveform optimization configuration: configure pulse welding waveform parameters, set the welding current to go through the base value stage, rising stage, peak stage, falling to the middle current stage, middle current holding stage and falling back to the base value current stage in one pulse cycle, so as to obtain the optimal welding pulse waveform that meets the stable droplet transfer mode of one pulse and one droplet. S200, Establish basic collaborative matching of excitation parameters: Based on the optimal welding pulse waveform, obtain the current welding speed and / or welding current effective value, and calculate and set the initial excitation current and initial excitation frequency of the composite magnetic field generator based on the preset linear matching relationship model; wherein, the composite magnetic field generator outputs two alternating excitation current waveforms with phase difference. S300 performs closed-loop control of arc and droplet dynamics: during the welding process, it acquires arc oscillation images and droplet transition images in real time, and calculates and extracts the arc oscillation speed and the arc dwell time at the weld edge; When the arc oscillation speed is detected to deviate from the preset target speed range, the excitation current is dynamically adjusted until the arc oscillation speed returns to the target speed range. When the residence time of the electric arc at the edge of the weld is detected to deviate from the preset target time range, the excitation frequency is dynamically adjusted until the residence time returns to the target time range; S400 performs molten pool temperature field feedback optimization control: During the welding process, the temperature field distribution image of the upper surface of the molten pool is continuously acquired, the temperature distribution data along the weld centerline direction is extracted to calculate the temperature gradient, and the temperature distribution pattern along the direction perpendicular to the weld is extracted. When the temperature gradient is greater than the set threshold, it is determined that the backward flow velocity of the liquid metal is too high, and the excitation current is increased to enhance the arc tilt angle. When the temperature distribution along the direction perpendicular to the weld is unimodal, it is determined that the heat has not diffused sufficiently to both sides of the weld. The excitation frequency is increased to enhance the heat input effect of the arc oscillation on the weld edge until the temperature distribution changes to bimodal.

[0021] Before executing steps S100 to S400, a physical hardware system compatible with this method needs to be constructed. Specifically, the physical execution end of the composite magnetic field generator includes dual electromagnetic coils symmetrically mounted on both sides of the GMAW welding torch. The excitation pole axes of the two sets of coils are perpendicular to each other and both intersect at the spatial geometric center of the arc combustion zone, thereby forming a composite Lorentz force field with continuously alternating vector directions in the welding area. Furthermore, to ensure the precise execution of steps S300 and S400, the main optical axis of the first high-speed CMOS camera is parallel to the welding forward direction and tilted downwards at a set angle, while the main optical axis of the second high-speed CMOS camera is perpendicular to the welding forward direction. The optical axis of the infrared thermal imager is tilted at an angle of 45°-60° towards the surface of the molten pool directly below the welding wire to avoid physical obstruction by the welding torch nozzle. Through the above rigid spatial topological constraints, the precise mapping between the acquired image coordinate system and the physical motion coordinate system of the welding machine tool is ensured, providing reliable underlying hardware support for subsequent dynamic closed-loop control.

[0022] Furthermore, the specific implementation process of step S100 is as follows: In the specific implementation, the detailed implementation and in-depth technical logic of step S100 (performing pulse waveform optimization configuration) are explained in detail. In the high-speed GMAW process environment assisted by a composite magnetic field, traditional pulse waveforms are difficult to adapt to the drastic changes in arc shape caused by the high-frequency magnetic field, easily leading to spatter or unstable transition. Therefore, in step S100, this invention constructs a six-stage pulse waveform structure with an intermediate current step. By precisely controlling the temporal coupling of the time dimension and current amplitude, refined hydrodynamic control of the wire tip melting, necking, and detachment processes is achieved, ensuring a stable one-pulse-one-droplet transfer mode.

[0023] When configuring pulse welding waveform parameters, the system precisely divides a complete pulse cycle into a base value phase, a rising phase, a peak phase, a drop to intermediate current phase, an intermediate current holding phase, and a drop back to the base value current phase. The electrical parameter settings for each phase strictly correspond to the evolution of the molten droplet's physical morphology.

[0024] First, in the base stage, the welding current is precisely configured within the range of 60-80A, for example, set to [specific value] in a particular implementation condition. Its duration The pulse duration is set to 1.5-2.1 ms (preferably 1.8 ms). The technical essence of this stage is to maintain the conductivity of the arc plasma channel with a low energy output, while preheating the tip of the welding wire and promoting the molten pool, which has undergone drastic thermophysical changes in the previous pulse cycle, to return to thermal equilibrium, thus preventing the welding wire from burning back due to excess heat.

[0025] This is followed by a rising phase, where the welding current rapidly increases from the base value. The duration of this phase is... The rise time is strictly controlled between 0.4-0.8 ms, preferably 0.6 ms. The extremely short rise time is designed to inject energy into the tip of the welding wire at an extremely high rate of current change, quickly generating a strong electromagnetic contraction force and avoiding droplet growth and eccentricity caused by slow heating.

[0026] Immediately following, the pulse waveform enters its peak phase, the core region where molten metal forms and begins to accumulate at the tip of the welding wire. The peak current corresponding to this peak phase is configured to 480-520A, with a duration of [duration missing]. The peak current is set to 500A and maintained for 0.9ms. Under such a high current density, the tip of the welding wire melts rapidly due to the combined effects of intense resistance heat and arc heat. Under the combined action of its own surface tension and the rapidly increasing electromagnetic contraction force, the liquid metal converges towards the center of the welding wire tip, forming a droplet to be transferred, and a significant necking physical characteristic is generated at the interface between the droplet and the solid welding wire.

[0027] After the necking occurs, the waveform enters a downward phase to the intermediate current stage, lasting for [duration missing]. The time is set to 0.3-0.6ms, preferably 0.4ms. Unlike traditional pulsed GMAW which drops abruptly to the base value, this invention introduces an intermediate current holding stage.

[0028] The intermediate current corresponding to this holding phase is configured to be 110-130A (e.g., 120A), and the duration is... It is 1.2-1.8ms (e.g., 1.5ms).

[0029] Specifically, if the current is directly reduced to the baseline value, the sudden drop in arc pressure will cause large-volume droplets to oscillate randomly or even undergo explosive transitions under the influence of gravity and residual electromagnetic forces. However, in the intermediate current range of 110-130A, the system can provide moderate axial arc pressure and plasma gas shear force. This moderate energy maintenance not only ensures the stability of the droplet when it detaches from the welding wire, but also provides a stable carrier of charged particle flow for subsequent spatial coupling with the composite magnetic field. This allows the droplet to maintain a controlled trajectory after detaching into the arc space, achieving smooth detachment and transition.

[0030] Finally, after the droplet transfer is complete, the waveform enters a descent phase and returns to the base current stage. The welding current is set to an extremely short time of 0.3-0.6ms (preferably 0.4ms) to accurately drop back to the base current, signaling the end of the current pulse cycle and preparing for the next arc-starting cycle.

[0031] Through the close combination of timing and amplitude in each stage, within a total pulse period of approximately 6.0 ms (effective welding current approximately 260 A, arc voltage approximately 22.3 V), the energy input gradient of the pulse waveform and the surface tension, gravity, and electromagnetic contraction force of the molten metal of the welding wire achieve a dynamic mechanical balance.

[0032] Furthermore, the specific implementation process of step S200 is as follows: In the specific implementation, the detailed implementation and underlying mathematical logic of step S200 (establishing basic coordination matching of excitation parameters) are explained in detail. After obtaining the optimal welding pulse waveform configured in step S100, the system enters the basic physical matching stage of magneto-electric coordination. Due to the drastic changes in the surface tension of the weld pool and the hydrodynamic characteristics of the liquid metal under high-speed movement conditions, longitudinal energy adjustment of the pulse waveform alone is insufficient to suppress the backward flow of the liquid metal and the loss of edge heat. Therefore, it is necessary to introduce dual external alternating magnetic fields to construct a multi-dimensional Lorentz force field to forcibly drive the arc plasma and liquid metal laterally and forward.

[0033] Firstly, in terms of the underlying structure of the physical waveform, the composite magnetic field generator is configured to output dual-path alternating excitation current waveforms with a phase difference. Specifically, the phase difference between the two excitation coils is strictly fixed at 180°, the duty cycle is set to 50%, and the base current is forced to zero at 0A. This waveform architecture ensures that in any given microscopic time segment, only one excitation coil is in an excited state, thereby generating a periodically polarity-reversing magnetic field vector within the welding area. The 50% duty cycle combined with the 0A base current creates alternating zero intersection points of the magnetic field without dead zones or overlaps. This effectively avoids the magnetic saturation and heat accumulation phenomenon of the iron core from an electromagnetic perspective, while ensuring that the transverse electromagnetic thrust and parallel electromagnetic thrust applied to the arc can generate high-frequency and clear directional switching, thus enabling the arc to exhibit stable transverse oscillation and forward tilting motion on a macroscopic scale. Based on this hardware, the excitation current adjustment range of the composite magnetic field generator is limited to 5-30A, and the excitation frequency adjustment range is limited to 15-40Hz, which constitute the basic parameter boundaries for subsequent dynamic control.

[0034] Furthermore, to address the technical shortcomings of traditional trial-and-error methods in quickly finding optimal solutions under different welding conditions, this invention incorporates a linear matching relationship model based on fluid dynamics compensation in step S200. The system acquires the current macroscopic process parameters (including welding speed) in real time. With the effective value of welding current And substitute it into the dynamic matching logic to calculate the initial excitation parameters.

[0035] When the system detects the welding speed During ultra-high-speed operation, the molten metal inside the molten pool is subjected to shear forces in the direction of motion, causing a sharp increase in its backward flow momentum. To counteract this fluid momentum, it is necessary to redistribute the heat mass by increasing the arc's forward tilt angle and oscillation frequency. Therefore, the system sets the initial excitation current... With welding speed The relationship between them follows a linear evolutionary equation: Among them, the slope parameter The value was selected as 25-35 A·min / m, which physically represents the additional electromagnetic thrust compensation coefficient required to overcome the surge in backward liquid metal momentum per unit welding speed increment; intercept parameter The range of -25 to -15A represents the basic energy threshold for activating this linear compensation mechanism.

[0036] Synchronously, the initial excitation frequency With welding speed The relationship between them follows the formula: In the formula, the parameters Configured to 25-40 Hz·min / m, its technical essence lies in the fact that as the heat source moving speed increases, the effective heating time obtained per unit length of weld edge decreases inversely. Therefore, it is necessary to... To linearly increase the arc oscillation frequency, ensuring the heat flow can frequently scour the weld edge and prevent undercut caused by incomplete fusion of the base metal; intercept parameter Then set it to -30 to -20Hz.

[0037] On the other hand, when the system obtains the effective value of the welding current At this time, the charged particle density and axial velocity within the arc plasma channel increase sharply, leading to a significant enhancement in the arc's stiffness (i.e., its ability to resist deflection by an external magnetic field). Furthermore, its inherent self-induced magnetic field also repels and interferes with the applied composite magnetic field. To address this high-current anti-deflection effect, the system sets the initial excitation current... With the effective value of welding current They follow a current coupling relationship: Among them, the current coupling coefficient The selected value is 0.1-0.3 A / A. This coefficient plays a role in stiffness compensation in the electromagnetic coupling model. That is, for every 1 A increase in the main welding current, the excitation current needs to be increased by 0.1-0.3 A simultaneously to ensure that the transverse Lorentz force applied to the rigid arc can overcome the dynamic pressure constraint of the plasma jet and maintain the required deflection angle; bias constant. Selected as -50 to -10A.

[0038] It should be further explained that the above linear evolutionary relationship... and current coupling relationship The slope parameters and bias constants in the model are derived through extensive orthogonal experiments and least squares data fitting, based on specific base materials (such as carbon steel or aluminum alloys) and specific shielding gas hydrodynamic properties. In practical engineering applications, the preferred range of these coefficients is suitable for high-speed welding under common operating conditions where the welding wire diameter is 1.0-1.2 mm and the shielding gas is an argon-rich mixture (such as 80%Ar+20%CO2 or 90%Ar+10%CO2). By limiting the above physical boundary conditions, the strong robustness and high prediction accuracy of the linear matching model in industrial environments are ensured.

[0039] Furthermore, the specific implementation process of step S300 is as follows: After the initial steady-state operating point is established in step S200, due to the microscopic fluctuations of the base material surface state and the nonlinear evolution of the liquid metal flow field during high-speed welding, the system must introduce a high-frequency dynamic visual feedback mechanism to deeply decouple and close the loop between the macroscopic magnetic field electrical parameters and the microscopic arc and droplet dynamics.

[0040] Specifically, the system employs two synchronously triggered high-speed CMOS cameras, coupled with a narrowband interference filter with a center wavelength of 808nm, to completely filter out broadband plasma strong light interference. The optical axis of the first camera is parallel to the welding direction, specifically designed to capture the lateral oscillation characteristics of the arc plasma under the influence of the Lorentz force. The optical axis of the second camera is perpendicular to the welding direction, specifically designed to record the arc's forward tilting behavior and the spatial deflection trajectory of the molten droplet parallel to the welding direction. The acquisition frequency of both cameras is set to no less than 3000Hz, and the exposure time is compressed to the order of 8μs to ensure that the microscopic thermal fluid morphology boundaries at extremely high speeds can be captured.

[0041] After acquiring a continuous high-frequency image sequence, the system incorporates a dedicated microscopic vision processing flow to ensure accurate extraction of the morphological features of the electric arc and molten droplets. First, a 3×3 median filter is applied to the original grayscale image sequence to eliminate high-frequency pulse noise caused by metal spatter and arc flash. Then, Otsu's method (maximum inter-class variance method) is used to dynamically adaptively threshold the filtered image, accurately separating the extremely bright electric arc plasma region, the relatively bright liquid droplet region, and the dark background. Finally, for the binarized connected components of the electric arc and molten droplets, the geometric centroid pixel coordinates are calculated based on the grayscale centroid method, and the two-dimensional pixel coordinates are mapped to absolute physical coordinates using the camera's extrinsic calibration matrix, providing a reliable data source for subsequent quantization calculations of dynamic parameters.

[0042] After acquiring a continuous high-frequency image sequence, the system first performs a process based on the arc oscillation velocity. Single-dimensional dynamic control of edge dwell time. For the arc oscillation velocity, a visual processing algorithm extracts the geometric centroid of the arc plasma luminescent core region in real time, marks the absolute timestamps of its arrival at the maximum left and right deviation positions on the time axis, and then calculates the time interval between the two. Simultaneously, image pixel calibration is used to quantify the absolute oscillation arc of the electric arc within that time interval. Through differential equations The transient arc oscillation velocity is calculated. The system compares the calculated value with the set target velocity range of 2.0-3.0° / ms.

[0043] The technical significance of this range lies in the fact that an oscillation speed below 2.0° / ms will cause excessive heat input concentration at the weld center, failing to effectively flush the edges; while a speed above 3.0° / ms will induce dynamic fracture of the plasma channel or arc instability. When detected... When the target range is deviated from, the system calls a single-variable optimization algorithm to dynamically increase or decrease the excitation current at high frequency with a compensation step size of 1-2A. The magnitude of the transverse Lorentz force is adjusted by directly changing the current amplitude until... It was forcibly pulled back to the target range.

[0044] Simultaneously, regarding the energy residence time of the electric arc at the weld edge, the system accurately calculates the actual residence time of the arc at both sides of the weld edge by statistically analyzing the number of consecutive image frames at which the arc's geometric centroid reaches the position of maximum deflection angle, and multiplying this number by the absolute time interval of a single frame (e.g., 1 / 3000s). The target time range for this parameter is strictly limited to 18-26ms. Within this time window, the arc can transfer sufficient enthalpy to the base metal edge, causing the metal in that area to reach its melting point. Simultaneously, it provides the necessary time margin for the liquid metal inside the molten pool to spread and fully fill the weld toe area under the drive of lateral surface tension, thus eliminating the hydrodynamic boundary conditions caused by undercut defects from a physical perspective. If this residence time is insufficient, the system directly adjusts the excitation frequency of the composite magnetic field in control steps of 2-5Hz. The duration of the electric arc at the edge is influenced by changing the period of magnetic field polarity switching.

[0045] Furthermore, step S300 breaks through the traditional single-parameter control dimension and constructs a multi-variable collaborative linkage mechanism based on the spatial envelopment configuration of the arc and molten droplet. The system extracts the arc core energy axis and the molten droplet gravity centerline from the second camera image, and calculates the angle between the two relative to the geometric axis of the welding wire in the initial undeflected state, i.e., the arc deflection angle. With droplet deflection angle The system will The target range is set at 15-25°. The target range is set at 5-15°, and a composite physical quantity characterizing the spatial coupling depth between the two—the deflection angle difference—is forcibly introduced. (in ), which strictly limits it to the 4-8° co-encapsulation range.

[0046] Angle difference The limitation is the core patent feature of this invention for achieving high-speed and stable transition in a single pulse and droplet. From the perspective of plasma aerodynamics, since the mass of the arc plasma gas is much lighter than that of the liquid metal droplet, under the same applied magnetic field, the response speed and deflection amplitude of the arc will inevitably be greater than those of the droplet. Controlling the angle within the 4-8° range means that the deflected arc can form an asymmetric thermodynamic and aerodynamic flow field on its periphery, stably enveloping the molten droplet at an off-center position within the high-temperature core region of the arc. This moderate spatial envelopment configuration generates a plasma flow shear force that tilts towards the front of the molten pool. This shear force not only precisely accelerates and pushes the molten droplet into the area to be welded, but also uses the droplet's own momentum to suppress the backward surge of the molten metal in the pool.

[0047] like A deflection greater than 8° indicates excessive arc deflection, causing the plasma flow field to detach from the droplet surface, resulting in the droplet losing its protection and aerodynamic acceleration, making it highly susceptible to splashing; if If the angle is less than 4°, the covering effect is too weak to provide sufficient forward drive component. When the system determines any angle or difference... Due to the time constraint, since angle deflection involves a strongly coupled nonlinear system involving magnetic force, plasma force, and fluid dynamics, adjusting a single parameter would cause system oscillation. Therefore, the controller is set to trigger a two-variable linkage mechanism to synchronously adjust the excitation current strictly according to a preset matrix. Amplitude and excitation frequency The frequency of the wave forcefully adjusts the plasma spatial shape until... It returns to the steady-state thermodynamic envelope range of 4-8°.

[0048] Furthermore, the aforementioned bivariate linkage adjustment mechanism is specifically implemented based on an embedded two-dimensional fuzzy decoupling control rule table. The system uses the deviation E between the current deflection angle difference Δθ and the target center value (e.g., 6°), and the rate of change EC of this deviation as two-dimensional input quantities. When it is determined that Δθ exceeds the 4-8° range: If E is greater than 0 and EC is greater than 0 (indicating excessive arc deflection and a worsening trend), the system looks up the table and outputs a negative excitation current increment and a positive excitation frequency increment to weaken the lateral thrust and accelerate heat dissipation.

[0049] If E is less than 0 and EC is less than 0 (indicating insufficient arc deflection and loss of coating), the system looks up the table and outputs a positive excitation current increment and a negative excitation frequency increment to forcibly enhance the forward shear force.

[0050] The decoupling lookup table logic based on the physical deviation state described above effectively overcomes the nonlinear interference caused by strong coupling between the magnetic field, electric arc and fluid, and prevents system oscillation during the control process.

[0051] Furthermore, the specific implementation process of step S400 is as follows: After completing the arc and droplet spatial dynamics control in step S300, although the motion trajectory of the fluid medium is constrained from the external input end, the microscopic enthalpy distribution of the liquid metal inside the molten pool still faces the risk of runaway due to the nonlinear change in the thermal conductivity of the base material and the potential Marangoni convection effect during high-speed welding. Therefore, in step S400, this invention constructs a multidimensional heat flow characteristic decoupling and hierarchical closed-loop feedback mechanism at the top level of the control architecture, directly anchoring the control core to the physical origin that determines the final weld formation quality—the temperature field distribution on the surface of the molten pool.

[0052] First, in the stage of extracting and reconstructing thermodynamic boundary information, the system is equipped with a high-resolution infrared thermal imager to continuously acquire grayscale radiation images of the upper surface of the molten pool. In an optimal engineering implementation environment, the image resolution of the infrared thermal imager is set to 288×382 pixels, the acquisition frame rate is no less than 80Hz, and the spectral response band is strictly limited to between 920nm and 1100nm to avoid the core radiation frequency band of the plasma arc. After the system hardware acquires the grayscale images, it calls the built-in image processor to accurately map the two-dimensional grayscale matrix pixel by pixel into an absolute temperature field matrix based on Planck's blackbody radiation law and the dynamically calibrated emissivity function of liquid steel material, thereby recovering the actual transient temperature topology of the molten pool surface.

[0053] The dynamic calibration process of the emissivity function of the liquid steel material is as follows: Before entering the high-speed welding operation, the system executes a calibration subroutine, using a high-precision thermocouple with known absolute temperature and an infrared thermal imager to synchronously collect temperature data of the central region of the reference molten pool; based on the collected infrared spectral radiative exitance and the measured true temperature of the thermocouple, the initial value of the actual emissivity of the base material under a specific protective gas is derived in reverse; in the subsequent dynamic welding process, the system performs linear compensation correction on the emissivity according to the real-time effective value of the welding current and a preset temperature-emissivity polynomial fitting curve, thereby effectively eliminating the interference of surface tension and oxide film fluctuations of the high-temperature liquid metal on the accuracy of infrared temperature measurement.

[0054] Based on this high-precision temperature field matrix, the system first performs the calculation of the longitudinal heat flow suppression rule. The controller extracts discrete data points of the longitudinal temperature distribution along the weld centerline and performs first-order differential operations to obtain the longitudinal temperature gradient. The physical nature of this longitudinal temperature gradient characterizes the rate of heat accumulation along the reverse direction of welding and the intensity of the backward flow of liquid metal. When the system determines the calculated... When the temperature exceeds the set safety threshold of 15°C / mm, it indicates that a large amount of high-temperature liquid metal is rapidly surging and accumulating towards the tail of the molten pool under the influence of arc dynamic pressure and its own inertia. This is a direct thermodynamic precursor to the high-speed welding hump defect. At this point, the system triggers closed-loop intervention, dynamically increasing the excitation current of the composite magnetic field in compensation steps of 2-5A. Since the applied transverse Lorentz force is positively correlated with the amplitude of the excitation current, an increase in current will immediately force the arc plasma to generate a larger forward tilt angle. This forward tilt configuration directs the momentum vector of the plasma gas flow more towards the welding advance direction, forcibly blocking the backward flow of liquid metal with powerful aerodynamic shear force until... It falls back to within the safety boundary of 15°C / mm.

[0055] Simultaneously, the system performs parallel calculations of the transverse heat flow broadening rule. The controller extracts transverse temperature profile data along a geometric direction perpendicular to the weld centerline. In conventional or insufficiently optimized magnetron welding, this transverse temperature distribution is typically unimodal, meaning the highest temperature extremes are concentrated in the geometric center of the molten pool, while the temperature at the edges decays rapidly. This thermodynamic characteristic implies excessive concentration of arc energy and insufficient heat input to diffuse into the base metal on both sides, inevitably leading to inadequate metallurgical wetting of the molten metal at the weld toe, resulting in undercut defects. When the pattern recognition algorithm determines that the current temperature profile is unimodal, the system immediately increases the excitation frequency with a compensation step size of 3-8 Hz. The high-frequency alternating polarity switching forces the arc to oscillate faster in the lateral direction, increasing the energy deposition density at the arc's dwell points on both sides, thus forcibly altering the lateral heat conduction path. After several closed-loop iterations, this lateral temperature distribution undergoes a phase-change topological reconstruction, evolving from a unimodal distribution to a bimodal distribution deviating from the center of the molten pool. The appearance of the bimodal characteristic indicates that the weld edge region has received peak heat input, the base material is fully melted, and this ensures the lateral spreading and defect-free filling of the liquid metal driven by surface tension.

[0056] Furthermore, as a supplement to the aforementioned longitudinal and transverse heat flow control constraints, the system monitors and extracts in real time the geometric length enclosed by the solid-liquid phase transition boundary in the temperature field matrix, i.e., the molten pool length. When detected At a length of 30mm, the system determines that there is a severe mismatch between the total heat input and the heat dissipation rate. An excessively long molten pool will cause the liquid metal behind it to lose arc protection, leading to severe high-temperature oxidation and backward gravity-induced sagging. In response, the system activates a multi-variable strong intervention mechanism, simultaneously increasing the excitation current in steps of 2-8A. And increasing the excitation frequency in steps of 5-10Hz By using electromagnetic force to compress the three-dimensional space composite extrusion, the backward extension of the heat flow is forcibly cut off, and the size of the molten pool is compressed to a safe threshold range.

[0057] Finally, to address potential energy coupling failures under complex and harsh conditions (such as severe oxide films on the base material surface or localized thermal bridge breaks), step S400 incorporates a self-optimization program for excitation parameters under extreme heat loss conditions. Throughout the entire welding sequence, the system monitors the absolute minimum temperature at the geometric center of the weld pool and the absolute maximum temperature at the solid-liquid boundaries on both sides of the weld in parallel. When the system detects that the minimum temperature at the center has fallen below 1400°C and the maximum temperatures on both sides have fallen below 1500°C, indicating a critical metallurgical melting point, it means that the current linear matching control logic has deviated from its effective operating range. At this point, the system immediately locks the macroscopic master process parameters (maintaining a wire extension length of 12-15 mm, a shielding gas flow rate of 18-22 L / min, and a shielding gas composition of 90% Ar + 10% CO2 constant), and based on the Lagrange multiplier method or genetic algorithm, initiates a global two-dimensional scan and optimization within the constrained 10-30 A excitation current space and 15-35 Hz excitation frequency space. The optimization algorithm takes the extraction of transient weld width as the first objective function and the weld surface smoothness characterized by minimizing the temperature gradient as the second objective function. It performs high-frequency virtual evaluation and experimental verification, thereby solving and updating the optimal excitation current and frequency combination under the current deviation from the conventional physical model in a very short time. This enables the welding system to achieve adaptive recovery and high-order intelligent control under extreme boundary conditions.

[0058] Comparative examples and verification: To further verify the actual technical effect of the composite magnetic field assisted pulse GMAW high-speed welding parameter matching method provided by the present invention, a comparative test was conducted under the same high-speed welding conditions (welding speed of 2.0m / min, plate thickness of 3mm for low carbon steel, and average welding current of 260A).

[0059] Comparative example (conventional pulsed GMAW high-speed welding): A traditional single-step pulse waveform was used without applying an external alternating magnetic field. Experimental results showed that a severe "hump" phenomenon appeared on the weld surface, with an average height difference of more than 2.5 mm, and continuous "undercut" defects appeared on both sides of the weld, with a maximum undercut depth of 0.8 mm, which could not meet the forming requirements.

[0060] This invention embodiment (integrated collaborative control): Utilizing the multi-step pulse waveform of this invention, and initiating basic collaborative and dual closed-loop control steps S200 to S400. Through real-time adjustment via visual sensing and temperature field feedback, experimental results show: the arc deflection angle difference... The value is stably constrained to a preset value. Within the enclosed area, there are no visible undercut defects on both sides of the weld, and the surface transitions smoothly; the backward flow of liquid metal is completely suppressed, and no hump is generated; the penetration depth increases by about 15%, and the weld width increases by about 20%, which fully verifies the significant superiority of this method in suppressing high-speed welding forming defects and improving the heat and mass transfer dimension.

[0061] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for matching parameters in high-speed GMAW welding using a composite magnetic field-assisted pulse, characterized in that... Includes the following steps: S100, perform pulse waveform optimization configuration: configure pulse welding waveform parameters, set the welding current to go through the base value stage, rising stage, peak stage, falling to the middle current stage, middle current holding stage and falling back to the base value current stage in one pulse cycle, so as to obtain the optimal welding pulse waveform that meets the stable droplet transfer mode of one pulse and one droplet. S200, Establish basic collaborative matching of excitation parameters: Based on the optimal welding pulse waveform, obtain the current welding speed and / or the effective value of welding current, and calculate and set the initial excitation current and initial excitation frequency of the composite magnetic field generator based on the preset linear matching relationship model; wherein, the composite magnetic field generator outputs two alternating excitation current waveforms with phase difference. S300 performs closed-loop control of arc and droplet dynamics: during the welding process, it acquires arc oscillation images and droplet transition images in real time, and calculates and extracts the arc oscillation speed and the arc dwell time at the weld edge; When the arc oscillation speed is detected to deviate from the preset target speed range, the excitation current is dynamically adjusted until the arc oscillation speed returns to the target speed range. When the residence time of the electric arc at the weld edge is detected to deviate from the preset target time range, the excitation frequency is dynamically adjusted until the residence time is restored to the target time range; S400 performs molten pool temperature field feedback optimization control: During the welding process, the temperature field distribution image of the upper surface of the molten pool is continuously acquired, the temperature distribution data along the weld centerline direction is extracted to calculate the temperature gradient, and the temperature distribution pattern along the direction perpendicular to the weld is extracted. When the temperature gradient is greater than a set threshold, it is determined that the backward flow velocity of the liquid metal is too high, and the excitation current is increased to enhance the arc tilt angle. When the temperature distribution pattern along the direction perpendicular to the weld is unimodal, it is determined that the heat has not diffused sufficiently to both sides of the weld. The excitation frequency is increased to enhance the heat input effect of the arc oscillation on the weld edge until the temperature distribution pattern transforms into a bimodal pattern.

2. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 1, characterized in that, In step S100, the pulse welding waveform parameters satisfy the following distribution settings in the time dimension: The duration of the rising phase is 0.4-0.8 ms; The duration of the peak phase is 0.7-1.2 ms; The duration of the drop to the intermediate current stage is 0.3-0.6 ms; The duration of the intermediate current holding phase is 1.2-1.8 ms; The duration of the current drop back to the base value phase is 0.3-0.6 ms; The duration of the base value phase is 1.5-2.1 ms.

3. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 2, characterized in that, The optimal welding pulse waveform is further limited in terms of current amplitude and its functional control as follows: The peak current corresponding to the peak phase is configured to be 480-520A to ensure that molten droplets are formed at the tip of the welding wire at the top of the peak phase. The intermediate current corresponding to the intermediate current holding stage is configured to be 110-130A, in order to maintain the arc energy after the peak current drops, so as to achieve the smooth shedding and transition of the molten droplet; The base current corresponding to the base phase is configured to be 60-80A to maintain the basic arc and prepare for the arc initiation of the next pulse cycle.

4. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 3, characterized in that, In step S200, the basic physical characteristics of the dual-path alternating excitation current waveform with a phase difference are set as follows: The phase difference between the two excitation current waveforms is fixed at 180°, the duty cycle is 50%, and the base current is 0A. Meanwhile, the excitation current of the composite magnetic field generator is adjustable in the range of 5-30A, and the excitation frequency is adjustable in the range of 15-40Hz.

5. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 4, characterized in that, The initial excitation current and initial excitation frequency of the composite magnetic field generator are calculated and set based on a preset linear matching relationship model. The internal dynamic matching logic specifically satisfies the following mathematical rules: When the welding speed is obtained At that time, the initial excitation current Ie and the welding speed are set. The relationship between them follows a linear evolutionary relationship: Among them, the slope parameter Selected as 25-35 A·min / m, intercept parameter Selected as -25 to -15A; Simultaneously, the initial excitation frequency f is set to have a linear relationship with the welding speed v: k3 is selected as b3 is selected as -30 to -20 Hz; When the effective value of the welding current is obtained At that time, set the initial excitation current. There is a current coupling relationship between the welding current and the effective value I. The coupling coefficient k2 is selected as 0.1-0.3A / A, and the bias constant b2 is selected as -50 to -10A.

6. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 5, characterized in that, In step S300, the dynamic quantitative evaluation and control rules for the arc oscillation speed and dwell time are as follows: The arc oscillation speed is obtained by determining the time interval between the maximum leftward deviation and the maximum rightward deviation of the arc. Calculate the arc oscillation radius during this time interval. The arc oscillation velocity can be calculated using this formula: The preset target speed range is set to 2.0-3.0° / ms; The method for obtaining the dwell time of the electric arc at the edge of the weld is as follows: extract the number of continuously acquired image frames when the electric arc swings to the position of the maximum deflection angle, and multiply it by the time interval of a single frame; wherein, the preset target time range is set to 18-26ms.

7. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 6, characterized in that, Step S300 further includes a coordinated control sub-step based on the deflection angle difference: By extracting features from the acquired arc oscillation images and molten droplet transfer images, the arc deflection angle between the arc axis and the welding wire axis is calculated and determined. And the droplet deflection angle between the droplet centerline and the welding wire axis. ; Set the arc deflection angle The target range is 15-25°, and the target range of the droplet deflection angle θd is 5-15°. The difference between the two deflection angles is calculated. Limited to the 4-8° co-encapsulation range, among which ; When the detected arc deflection angle Or droplet deflection angle When the deviation deviates from the corresponding target range, a dual-variable linkage adjustment mechanism is triggered, that is, the excitation current and the excitation frequency are adjusted synchronously until the deflection angle and its difference return to the preset range.

8. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 7, characterized in that, In step S400, the control logic for extracting temperature distribution data and performing closed-loop optimization is based on the decoupling of multidimensional characteristics of heat flow in the molten pool space, and specifically includes the following hierarchical control rules: The grayscale images of the upper surface of the molten pool are continuously acquired by an infrared thermal imager and converted into actual temperature images based on the blackbody radiation law. Longitudinal heat flow suppression rule: Extract temperature distribution data along the weld centerline (longitudinal direction) and calculate the temperature gradient. When the temperature gradient If the backward flow velocity of the liquid metal is too high, the excitation current is increased by a compensation step of 2-5A to enhance the arc tilt angle and thus suppress the backward flow. Lateral heat flow broadening rule: Extract the temperature distribution pattern along the direction perpendicular to the weld, i.e., the lateral direction; when the temperature distribution pattern is a single peak with the highest temperature in the center, it is determined that the heat has not diffused sufficiently to both sides of the weld. Increase the excitation frequency with a compensation step of 3-8Hz, and use enhanced arc oscillation to promote edge heat input until the lateral temperature distribution pattern evolves into a double peak that deviates from the center of the molten pool. Melt pool size coordinated control rule: real-time monitoring of melt pool length When the length of the molten pool is detected If it is determined that the liquid metal has severely accumulated backward at the tail of the molten pool, the excitation current is increased by 2-8A and the excitation frequency is increased by 5-10Hz to forcibly reduce the length of the molten pool.

9. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 8, characterized in that, The step S400 also includes a built-in excitation parameter self-optimization program to cope with extreme heat loss conditions, specifically: During the temperature field feedback control process, the lowest temperature at the center of the molten pool and the highest temperature on both sides of the weld are monitored simultaneously. When the lowest temperature at the center of the molten pool is detected to be less than 1400°C and the highest temperature on both sides of the weld is less than 1500°C, the excitation parameter self-optimization program is triggered.

10. The method for matching parameters in high-speed GMAW welding with a composite magnetic field assisted pulse according to claim 9, characterized in that, The excitation parameter self-optimization program performs a global two-dimensional scan search within the excitation current space of 10-30A and the excitation frequency space of 15-35Hz, while maintaining the constraints of fixed wire extension length, shielding gas flow rate and shielding gas composition. It constructs an objective function with the goal of maximizing weld width and optimizing weld surface quality, and solves and updates the optimal combination of excitation current and optimal excitation frequency under the current process conditions.

Images