Intelligent control method and system for welding current of electric welding machine

Abstract

The invention discloses an intelligent control method and system for welding current of an electric welding machine, and relates to the technical field of welding. The method comprises the following steps that optical information of a to-be-welded area on the surface of a structural part in front of the advancing direction of a welding gun of the electric welding machine is obtained; based on the optical information, pollution information of the to-be-welded area is recognized, and the pollution information comprises at least one of the number, the category, the thickness, the distribution and the layered structure of pollutants; before the welding arc advancing along with the welding gun acts on the to-be-welded area, welding current parameters of the electric welding machine are adjusted according to the pollution information; in the welding process, electrical parameters used for reflecting the welding arc state are collected in real time, and the deviation between the electrical parameters and the adjusted welding current parameters is calculated; and welding current parameters of the electric welding machine are adjusted in real time according to the deviation so as to maintain welding arc stability. When pollutants exist on the surface of the to-be-welded workpiece, the welding current can be accurately managed, and the stability of the welding arc and the welding quality are improved.

Description

Technical Field

[0001] This application relates to the field of welding technology, and in particular to an intelligent control method and system for welding current in an electric welding machine. Background Technology

[0002] In modern manufacturing workshops, welding machines commonly employ an intelligent control system to manage the welding current. This system stores a large amount of expert data, pre-setting optimal welding current, voltage, and wire feed speed relationships for different metal materials, plate thicknesses, and joint types. During welding, the system collects arc voltage and current information in real time and compares it with internally set reference values. If a deviation is detected, the system quickly adjusts the inverter output, thereby changing the welding current and achieving a closed-loop control process. This method performs exceptionally well when handling standardized, clean workpieces, ensuring uniform weld formation, precise penetration depth and width, significantly improving production efficiency and product quality stability.

[0003] However, the situation becomes complicated when these welding machines are applied to non-standardized work environments. For example, in the segmented construction of large ships or on-site construction of bridge steel structures, the steel plates to be welded are often no longer in an ideal clean state due to long-term storage and transportation. The surface may be covered with a thin, uneven layer of rust, or contaminated with rust-preventive oil, grease, or other oils during transport. Although workers are usually required to grind and clean the welding area before starting welding, this pretreatment is often not thorough enough in some complex corners or due to tight deadlines, leaving some contaminants behind.

[0004] When an electric arc is applied to a rusty surface, the resistance of the rust layer causes arc instability, leading to violent voltage fluctuations. The system misjudges this as insufficient arc energy and blindly increases the current, resulting in excess energy after the arc burns through the rust layer, causing the base material to burn through and exacerbating spatter. When oil is present, the arc decomposes it, producing gas and altering the arc characteristics. The system similarly misadjusts the current, not only failing to stabilize the arc but also causing latent defects such as porosity and hydrogen-induced cracks in the weld.

[0005] Experienced welders can proactively adapt by using visual and auditory information. However, current intelligent control systems rely solely on electrical signal feedback and cannot identify the root cause of contaminants. Therefore, their "intelligence" is powerless in the face of such complex and non-ideal working conditions, and may even exacerbate the problem due to incorrect adjustments. Summary of the Invention

[0006] Therefore, this application proposes an intelligent control method and system for welding current of an electric welding machine, which aims to solve the technical problem that existing welding equipment is unable to accurately manage the welding current when there are contaminants on the surface of the workpiece to be welded, thus leading to unstable welding quality.

[0007] In a first aspect, this application provides an intelligent control method for welding current in an electric welding machine, applied in the welding process of structural components, the method comprising the following steps: Acquire optical information of the area to be welded on the surface of the structural component in front of the direction of travel of the welding torch of the electric welding machine; Based on the optical information, the contamination information of the area to be welded is identified, and the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants. Before the welding arc traveling with the welding torch acts on the area to be welded, the welding current parameters of the welding machine are adjusted according to the contamination information. During the welding process, electrical parameters reflecting the state of the welding arc are collected in real time, and the deviation between the collected electrical parameters and the adjusted welding current parameters is calculated. The welding current parameters of the welding machine are adjusted in real time according to the deviation to maintain the stability of the welding arc.

[0008] According to some embodiments of this application, the step of acquiring optical information of the area to be welded on the surface of the structural member in front of the welding torch traveling in the direction of travel of the welding machine includes: Multi-band image data of the area to be welded on the surface of the structural component in front of the welding torch of the electric welding machine is acquired by a multispectral imaging sensor integrated on the welding torch of the electric welding machine. Geometric and radiometric corrections are performed on the multi-band image data; Based on the corrected multi-band image data, optical information of the area to be welded on the surface of the structural component is obtained.

[0009] According to some embodiments of this application, the step of identifying contamination information of the area to be welded based on the optical information, wherein the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants, includes: Obtain a pre-stored feature library, which includes characteristic spectral bands corresponding to various pollutants; The reflectance ratios of the optical information in the characteristic spectral bands corresponding to various contaminants are analyzed to identify the quantity and type of contaminants in the area to be welded. The thickness and distribution of the contaminants are estimated by analyzing the absorption intensity of the optical information of the area to be welded where the contaminants are identified in the corresponding characteristic spectral bands.

[0010] According to some embodiments of this application, the pre-stored feature library further includes preset ratio thresholds corresponding to pollutants; The step of analyzing the reflectance ratios of the optical information in the characteristic spectral bands corresponding to various contaminants, and identifying the quantity and type of contaminants in the area to be welded, includes: Calculate the reflectance ratio of the optical information in the characteristic spectral bands corresponding to various pollutants; The reflectance ratios were compared with preset ratio thresholds for various pollutants to obtain the comparison results. Analyzing the comparison results, the quantity and type of contaminants in the area to be welded are obtained.

[0011] According to some embodiments of this application, the step of analyzing the absorption intensity of the optical information of the area to be welded, where contaminants are identified, in the corresponding characteristic spectral bands, and estimating the thickness and distribution of the contaminants includes: When multiple types of pollutants are identified, the optical information is subjected to spectral stripping analysis to obtain the layered structure of the pollutants; Calculate the absorption intensity of each pollutant layer in the layered structure in the corresponding characteristic spectral band; The thickness and distribution of pollutants in each layer are estimated based on the absorption intensity.

[0012] According to some embodiments of this application, the step of adjusting the welding current parameters of the welding machine based on the contamination information before the welding arc traveling with the welding torch acts on the area to be welded includes: Before the welding arc traveling with the welding torch acts on the area to be welded, the layered structure of the contamination information is identified, the layered structure including single-layer contaminants and multiple layers of contaminant superposition; When the layered structure is identified as a single-layer contaminant, the type and thickness of the contaminant are further identified, and the welding current parameters of the welding machine are adjusted by calling the adjustment mode corresponding to the type and thickness. When the layered structure is identified as a superposition of multiple contaminants, the welding current parameters of the welding machine are adjusted according to the identified layered structure, from the outermost layer to the innermost layer, and the adjustment mode corresponding to each layer of contaminants is called in turn.

[0013] According to some embodiments of this application, the step of further identifying the type and thickness of the contaminant when the layered structure is identified as a single-layer contaminant, and then invoking an adjustment mode corresponding to the type and thickness to adjust the welding current parameters of the welding machine includes: When the layered structure is identified as a single-layer contaminant, the type and thickness of the contaminant are further identified, including rust and oil stains; If the category is rust, the welding current parameters of the welding machine are adjusted using a high-energy pulse mode. The high-energy pulse mode includes: before the arc contacts the rust layer, increasing the peak value of the arc-starting current of the welding machine to a preset first current value and maintaining it for a duration corresponding to the identified thickness. If the category is oil stains, the welding current parameters of the welding machine are adjusted using a gentle vaporization mode. The gentle vaporization mode includes: applying a preset preheating pulse before the main welding pulse of the welding machine, and reducing the peak current of the main welding pulse to a preset second current value; the energy parameters of the preheating pulse are set according to the identified thickness, and the second current value is lower than the first current value.

[0014] According to some embodiments of this application, the step of adjusting the welding current parameters of the welding machine in real time according to the deviation to maintain a stable welding arc includes: When the deviation exceeds the preset adjustment threshold, the welding current parameter is adjusted compensatorily to maintain the stability of the welding arc. When the deviation is lower than or equal to the preset adjustment threshold, the current welding current parameters are maintained to keep the welding arc stable.

[0015] According to some embodiments of this application, the step of compensatingly adjusting the welding current parameters to maintain a stable welding arc when the deviation exceeds a preset adjustment threshold includes: When the deviation exceeds a preset adjustment threshold, a spectrum analysis is performed on the electrical parameters collected in real time. When a continuous characteristic ripple caused by abnormal internal material of the structural component is identified in the preset frequency band through the spectrum analysis, the first adjustment strategy is activated to compensate for the welding current parameter in order to maintain the stability of the welding arc. The first adjustment strategy includes: reducing the frequency of adjusting the welding current parameter and switching the current welding current mode to a constant current mode or a stable pulse mode with extended base time and reduced pulse peak current. When the continuous characteristic ripple is not identified within the preset frequency band through the spectrum analysis, the second adjustment strategy is activated to compensate for the welding current parameters in order to maintain the stability of the welding arc. The second adjustment strategy includes: according to the preset adjustment range, the welding current parameters are reverse compensated and corrected according to the deviation.

[0016] Secondly, this application also provides an intelligent control system for welding current of an electric welding machine, comprising: The information acquisition module is used to acquire optical information of the area to be welded on the surface of the structural component in front of the welding torch traveling in the direction of the welding machine; An information recognition module is used to identify contamination information of the area to be welded based on the optical information, wherein the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants. The parameter adjustment module is used to adjust the welding current parameters of the welding machine according to the contamination information before the welding arc traveling with the welding torch acts on the area to be welded. The deviation calculation module is used to collect electrical parameters reflecting the state of the welding arc in real time during the welding process, and to calculate the deviation between the collected electrical parameters and the adjusted welding current parameters. The parameter stabilization module is used to adjust the welding current parameters of the welding machine in real time according to the deviation, so as to maintain the stability of the welding arc.

[0017] The technical solution according to the embodiments of this application has at least the following beneficial effects: This application addresses the issue of contaminants on the surface of workpieces to be welded by acquiring and identifying contamination information before the welding arc begins, and adjusting welding current parameters accordingly to proactively prevent potential problems. Through this forward-looking adjustment, combined with real-time feedback correction during the welding process, this application accurately manages the welding current, improving the stability of the welding arc and the quality of the weld.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0020] Figure 1 This is a flowchart illustrating an intelligent control method for welding current in an electric welding machine, as provided in an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of the architecture of an intelligent control system for welding current of an electric welding machine, provided as an embodiment of this application. Detailed Implementation

[0022] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0023] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0024] Traditional welding equipment is typically equipped with intelligent control systems designed to precisely manage welding current by monitoring the electrical characteristics of the arc to ensure weld quality. However, in practical industrial applications, especially during the on-site construction or manufacturing of large structural components, the surface condition of the workpieces to be welded is often less than ideal, potentially containing various contaminants. This poses a significant challenge to existing intelligent control systems. Failure to address these issues will lead to unstable welding quality and even welding defects, severely impacting the performance and reliability of the structural components.

[0025] In this regard, such as Figure 1 As shown, this application discloses an intelligent control method for welding current in an electric welding machine, applied to the welding process of structural components. The method includes the following steps: S110, acquire optical information of the area to be welded on the surface of the structural component in front of the direction of travel of the welding torch of the electric welding machine; S120, based on the optical information, identify contamination information of the area to be welded, wherein the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants; S130, before the welding arc traveling with the welding torch acts on the area to be welded, the welding current parameters of the welding machine are adjusted according to the contamination information. S140: During the welding process, electrical parameters that reflect the state of the welding arc are collected in real time, and the deviation between the collected electrical parameters and the adjusted welding current parameters is calculated. S150, the welding current parameters of the welding machine are adjusted in real time according to the deviation to maintain the stability of the welding arc.

[0026] It should be noted that the "optical information" mentioned in this application refers to visual data about the area to be welded on the surface of a structural component acquired through optical sensors. This data can include visible light images, infrared images, ultraviolet images, etc., and is used to reflect the presence, type, and distribution of surface contaminants. "Contamination information" refers to specific data about contaminants extracted from the optical information, such as the quantity, type (e.g., rust, oil), thickness, distribution on the surface, and whether there are multi-layered structures. This information is crucial for subsequent adjustments to welding current parameters. "Welding current parameters" encompass various current settings during the welding process, such as arc initiation current, main welding current, pulse frequency, and duty cycle. Adjustments to these parameters directly affect the energy input and stability of the welding arc.

[0027] First, it is necessary to acquire optical information about the area to be welded on the surface of the structural component in front of the welding torch's travel direction. For example, one or more cameras can be installed near the welding torch to continuously capture images of the area to be welded as the torch travels. These cameras can be ordinary visible light cameras or sensors with specific spectral responses. The acquired image data is then transmitted to a processing unit for analysis. Alternatively, a laser scanner can be used to scan the area to be welded, and optical feature data of the surface can be obtained by analyzing laser reflection or scattering signals.

[0028] Based on the acquired optical information, the next step is to identify contamination in the area to be welded. For example, image processing algorithms can be used to analyze visible light images, using features such as color and texture to determine the presence of contaminants. When color patches or irregular textures appear in the image that differ from the clean metal surface, it can be preliminarily determined that contaminants are present. Furthermore, the quantity and distribution of contaminants can be estimated by analyzing the area of ​​these abnormal regions. As another approach, spectral analysis techniques can be used to identify the type of contaminant by analyzing the reflection or absorption characteristics of different wavelengths of light on the contaminant surface. For example, some contaminants have unique absorption peaks in specific infrared bands; detecting these absorption peaks can accurately identify the type of contaminant.

[0029] Before the welding arc traveling with the welding torch acts on the area to be welded, the welding current parameters of the welding machine need to be adjusted based on the identified contamination information. For example, if a thin layer of oil is detected in the area to be welded, the arc-starting energy of the welding current can be appropriately increased beforehand to help the oil be effectively removed or vaporized before the arc acts. Alternatively, if a thick layer of rust is detected, the pulse waveform of the welding current can be adjusted, for example, by increasing the duration of the peak pulse current, to ensure that the arc can penetrate the rust layer and form a stable molten pool.

[0030] During welding, it is necessary to collect electrical parameters in real time to reflect the state of the welding arc. For example, voltage and current sensors can be installed at the output of the welding machine to monitor the waveforms of the welding voltage and current in real time. These electrical parameters can directly reflect key information such as the length, stability, and energy input of the welding arc. As another approach, the arc state can also be monitored indirectly through acoustic or optical sensors, for example, by analyzing the intensity of sound or light radiation emitted by the arc to determine its stability.

[0031] Subsequently, the deviations between the real-time acquired electrical parameters and the adjusted welding current parameters are calculated. For example, the effective value of the real-time acquired welding current can be compared with a preset target current value to obtain the current deviation. Similarly, the real-time acquired arc voltage can be compared with a target voltage to obtain the voltage deviation. These deviation values ​​reflect the difference between the actual state and the desired state of the welding arc.

[0032] Finally, the welding current parameters of the welding machine are adjusted in real time based on the calculated deviations to maintain a stable welding arc. For example, when the welding current is detected to be lower than the target value, the output current of the welding machine can be appropriately increased to compensate for energy loss. When large fluctuations in the arc voltage are detected, the wire feed speed or the distance between the welding torch and the workpiece can be adjusted to stabilize the arc length. Through this closed-loop control method, the welding arc can be ensured to remain stable throughout the welding process, thereby obtaining a high-quality weld.

[0033] The intelligent control method for welding current of the electric welding machine disclosed in this application obtains and identifies the contamination information of the area to be welded before the welding arc acts on the area to be welded, and adjusts the welding current parameters of the electric welding machine accordingly, thereby effectively dealing with the impact of contaminants on the welding process.

[0034] This application addresses the issue of contaminants on the surface of workpieces to be welded by acquiring and identifying contamination information before the welding arc begins, and adjusting welding current parameters accordingly. This proactive adjustment, combined with real-time feedback correction during the welding process, enables accurate management of the welding current, improves the stability of the welding arc and welding quality, and effectively solves the problem of unstable welding quality caused by contaminants in existing technologies.

[0035] In a specific embodiment of this application, the step of acquiring optical information of the area to be welded on the surface of the structural member in front of the welding torch traveling in the direction of travel of the welding machine preferably includes: Multi-band image data of the area to be welded on the surface of the structural component in front of the welding torch of the electric welding machine is acquired by a multispectral imaging sensor integrated on the welding torch of the electric welding machine. Geometric and radiometric corrections are performed on the multi-band image data; Based on the corrected multi-band image data, optical information of the area to be welded on the surface of the structural component is obtained.

[0036] Multi-band image data refers to image information acquired across multiple discrete or continuous spectral bands, capable of capturing subtle differences in the interaction between different wavelengths of light and objects. A multispectral imaging sensor is a device capable of simultaneously imaging in multiple specific spectral bands. By analyzing the spectral responses of different bands, it can acquire richer and more detailed physical and chemical information about the surface of the area to be welded. This sensor is integrated into the welding torch of a welding machine to ensure that, during the welding process, optical information of the area to be welded on the surface of the structural component in front of the welding torch's travel direction can be acquired in real time and at close range, thus providing accurate raw data for subsequent contamination identification.

[0037] The multi-band image data undergoes geometric and radiometric correction. Geometric correction aims to eliminate geometric distortions caused by factors such as sensor orientation, viewing angle, and terrain undulations during image acquisition, ensuring an accurate correspondence between pixels in the image and their actual spatial locations. Radiometric correction eliminates the influence of factors such as uneven sensor response, atmospheric effects, and changes in lighting conditions on image radiometric values, ensuring that the image brightness accurately reflects the reflective or radiative characteristics of the ground surface, thereby improving the accuracy and reliability of subsequent pollution information identification. Therefore, based on the multi-band image data after geometric and radiometric correction, the optical information of the area to be welded on the surface of the structural component can be accurately obtained. This optical information may include, but is not limited to, spectral characteristics such as reflectivity, absorptivity, and emissivity of the area to be welded, as well as image features such as texture and color. These features are key evidence for identifying the type, quantity, thickness, distribution, and layered structure of pollutants.

[0038] This application's solution integrates a multispectral imaging sensor onto the welding torch, enabling real-time acquisition of multi-band image data of the area to be welded in front of the torch. This multispectral data acquisition method can capture the unique responses of contaminants in different spectral bands, thus providing rich spectral feature information for the precise identification of contaminants. Subsequently, by performing rigorous geometric and radiometric corrections on the acquired multi-band image data, errors caused by the acquisition environment and the equipment itself are eliminated, ensuring the spatial accuracy and radiometric authenticity of the optical information.

[0039] Through the above technical solution, this application can obtain more comprehensive, accurate, and reliable optical information about the area to be welded. Compared with methods that only acquire single-band or simple image information, multispectral imaging technology, combined with geometric and radiometric correction, significantly improves the quality and usability of optical information. This not only helps to more accurately identify the type, quantity, thickness, distribution, and layering structure of contaminants in the area to be welded, but also provides more solid data support for subsequent intelligent adjustment of welding current parameters based on contaminant information.

[0040] In some embodiments of this application, the step of identifying contamination information of the area to be welded based on the optical information, wherein the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants, preferably includes: Obtain a pre-stored feature library, which includes characteristic spectral bands corresponding to various pollutants; The reflectance ratios of the optical information in the characteristic spectral bands corresponding to various contaminants are analyzed to identify the quantity and type of contaminants in the area to be welded. The thickness and distribution of the contaminants are estimated by analyzing the absorption intensity of the optical information of the area to be welded where the contaminants are identified in the corresponding characteristic spectral bands.

[0041] The pre-stored feature library can be understood as a pre-established dataset containing the optical response characteristics of different types of pollutants (such as rust, oil, and oxide layers) in specific spectral bands. These characteristic spectral bands are determined through experimental or theoretical analysis and are wavelength ranges that can effectively distinguish different pollutants or reflect their specific properties. The establishment of this pre-stored feature library aims to provide a reference standard for subsequent pollutant identification.

[0042] Analyzing the reflectance ratios of optical information across characteristic spectral bands corresponding to various contaminants involves comparing the reflection intensities of optical information collected from the area to be welded across different characteristic spectral bands and calculating their ratios. These ratios are then matched with known contaminant characteristics in a pre-stored feature library to identify the quantity and specific category of contaminants present in the area to be welded. For example, different types of contaminants may have unique reflectance spectral characteristics in the visible, near-infrared, or mid-infrared bands. By analyzing these characteristic ratios, a preliminary classification and quantity determination of the contaminants can be achieved.

[0043] Analyzing the absorption intensity of the optical information of the area to be welded, where contaminants have been identified, in the corresponding characteristic spectral bands refers to a detailed analysis of the absorption characteristics of that type of contaminant in a specific characteristic spectral band after determining the type of contaminant. Contaminants of different thicknesses and distributions will absorb light of specific wavelengths to varying degrees. By measuring and analyzing these absorption intensities, the specific thickness of the contaminant and its distribution on the surface of the area to be welded can be estimated. For example, a higher absorption intensity may indicate a thicker or denser distribution of contaminants.

[0044] This application's solution achieves refined identification of surface contamination information on structural components by introducing a pre-stored feature library and combining it with the analysis of reflectivity ratios and absorption intensities of optical information in characteristic spectral bands. Therefore, this solution can comprehensively and accurately obtain contamination information of the area to be welded, laying the foundation for precise adjustment of subsequent welding current parameters.

[0045] Through the above technical solution, this application can achieve comprehensive and accurate identification of surface contamination information of structural components, including key parameters such as the quantity, type, thickness, and distribution of contaminants. Compared with methods that only identify the presence of contamination or simple classification, this solution significantly improves the accuracy and reliability of contamination information identification by utilizing a pre-stored feature library and multispectral analysis technology.

[0046] In a further embodiment of this application, the pre-stored feature library also includes a preset ratio threshold corresponding to the pollutants.

[0047] The step of analyzing the reflectance ratios of the optical information in the characteristic spectral bands corresponding to various contaminants to identify the quantity and type of contaminants in the area to be welded preferably includes: Calculate the reflectance ratio of the optical information in the characteristic spectral bands corresponding to various pollutants; The reflectance ratios were compared with preset ratio thresholds for various pollutants to obtain the comparison results. Analyzing the comparison results, the quantity and type of contaminants in the area to be welded are obtained.

[0048] Specifically, in addition to containing characteristic spectral bands corresponding to various pollutants, the pre-stored feature library is also configured to store preset ratio thresholds corresponding to these pollutants. These preset ratio thresholds are determined in advance based on a large amount of experimental data and expert experience, and are used as quantitative standards to determine the existence or category of a specific pollutant. For example, if the reflectance ratio of a certain pollutant in a specific characteristic spectral band is higher than a certain preset threshold, then the pollutant can be determined to exist.

[0049] Calculating the reflectance ratios of the optical information in the characteristic spectral bands corresponding to various contaminants refers to extracting the reflectance data of the optical information acquired from the area to be welded, for each contaminant defined in a pre-stored feature library, and calculating the reflectance ratios between these bands. For example, the ratio of reflectance between two or more specific bands, or the ratio of reflectance between a certain band and a reference band, can be calculated to highlight the spectral characteristics of different contaminants.

[0050] The reflectance ratio is compared with preset threshold values ​​for various contaminants to obtain comparison results. This involves comparing the calculated reflectance ratio for each contaminant with the corresponding preset threshold value in a pre-stored feature library. This comparison can employ logical judgment methods such as greater than, less than, equal to, or within a certain range. Thus, a series of comparison results are obtained, directly indicating whether a certain contaminant exists in the area to be welded, or which contaminant its spectral characteristics most closely match.

[0051] Finally, by analyzing the comparison results, the quantity and type of contaminants in the area to be welded are obtained. Specifically, by comprehensively analyzing the comparison results of all contaminants, the types of contaminants present in the area to be welded, as well as the possible quantities of contaminants, can be determined. For example, if the reflectance ratios of multiple contaminants all meet their corresponding preset ratio threshold conditions, it indicates that the area to be welded may contain multiple contaminants. This analysis process can be implemented using decision trees, logical judgment rules, or machine learning models to improve the accuracy and robustness of identification.

[0052] Through the above technical solution, this application can significantly improve the accuracy and reliability of contaminant identification in the welding area. By introducing a preset ratio threshold, contaminant identification no longer relies on subjective judgment or fuzzy analysis, but rather on quantitative standards for decision-making, thus effectively avoiding misjudgments caused by similar spectral characteristics or environmental interference. Furthermore, by analyzing the comparison results, the quantity and type of contaminants can be determined more accurately, providing a more solid data foundation for subsequent fine-tuning of welding current parameters.

[0053] The following is a specific example to illustrate this.

[0054] Assume that the pre-stored feature library contains the characteristic spectral bands of two common pollutants, rust and oil, and their corresponding preset ratio thresholds. For example, for rust, there is a preset threshold T_rust for the ratio of reflectance in a specific infrared band A to that in the visible light band B (R_A / R_B); for oil, there is a preset threshold T_oil for the ratio of reflectance in the ultraviolet band C to that in the near-infrared band D (R_C / R_D).

[0055] After acquiring the optical information of the area to be welded, the system first calculates the reflectivity ratio of that area between band A and band B and compares it with T_rust. If the calculated ratio is higher than T_rust, it is preliminarily determined that rust is present. Simultaneously, the system also calculates the reflectivity ratio of that area between band C and band D and compares it with T_oil. If the calculated ratio is higher than T_oil, it is preliminarily determined that oil contamination is present.

[0056] By comprehensively analyzing these comparison results, the contamination status of the area to be welded can be determined. For example, if only the ratio of rust meets the threshold condition, it is identified as single-layer rust contamination; if only the ratio of oil meets the threshold condition, it is identified as single-layer oil contamination; if both meet the threshold condition, it may be identified as multi-layer contamination (e.g., oil covering rust, or both coexisting), in which case further spectral stripping analysis may be needed to determine the layered structure. This threshold-based judgment mechanism makes the contaminant identification process more automated and accurate, providing a clear basis for subsequent adjustment of welding current parameters.

[0057] In a further embodiment of this application, the step of analyzing the absorption intensity of the optical information of the area to be welded, where contaminants are identified, in the corresponding characteristic spectral band, and estimating the thickness and distribution of the contaminants preferably includes: When multiple types of pollutants are identified, the optical information is subjected to spectral stripping analysis to obtain the layered structure of the pollutants; Calculate the absorption intensity of each pollutant layer in the layered structure in the corresponding characteristic spectral band; The thickness and distribution of pollutants in each layer are estimated based on the absorption intensity.

[0058] "Spectral stripping analysis" can be understood as a spectral unmixing technique, aiming to decompose a mixed spectral signal into the spectra of its constituent pure components and their corresponding abundances (i.e., the proportion or contribution of each component in the mixture). Specifically, when multiple contaminants are superimposed in the area to be welded, forming a layered structure, the optical information acquired by a multispectral imaging sensor is a mixture of the spectral information of these contaminant layers and the substrate material. Spectral stripping analysis uses mathematical models and algorithms, such as linear or nonlinear mixing models, combined with the pure spectral features of known contaminants in a pre-stored feature library, to separate the independent spectral contribution of each contaminant layer from the mixed spectrum, thereby identifying the layered structure of the contaminants. For example, an endmember extraction algorithm can be used to identify the pure component spectra, and then an abundance inversion algorithm can be used to calculate the relative content of each component, thus inferring the layering of the contaminants.

[0059] After obtaining the layered structure of the pollutants, it is necessary to calculate the absorption intensity of each layer in its corresponding characteristic spectral band. This means that for each layer of pollutants obtained through spectral stripping analysis, its absorption characteristics in a specific spectral band will be analyzed individually. For example, if the first layer is identified as oil and the second layer as rust, the absorption intensity of the oil layer in its characteristic spectral band and the absorption intensity of the rust layer in its characteristic spectral band will be calculated separately. The purpose of this step is to obtain independent absorption information for each layer of pollutants, avoiding spectral interference between different pollutants.

[0060] Subsequently, the thickness and distribution of each layer of contaminants are estimated based on the absorption intensity. A quantitative relationship exists between absorption intensity and the concentration and thickness of the substance, which can be modeled, for example, using the Lambert-Beer Law. Using a pre-established calibration curve or model between absorption intensity and contaminant thickness, the specific thickness of each contaminant layer can be accurately estimated based on its calculated absorption intensity. Simultaneously, by combining the spatial distribution of optical information, the distribution of each layer of contaminants within the welding area can be further determined, such as whether it is uniformly distributed or locally concentrated.

[0061] This application's solution effectively solves the problem of accurately estimating the thickness and distribution of each contaminant layer when multiple contaminants overlap to form a layered structure. Spectral stripping analysis decomposes mixed spectral signals into individual component spectra, revealing the layered structure of the contaminants. This ability to identify the layered structure allows for the separate calculation of the absorption intensity of each contaminant layer in its corresponding characteristic spectral band. This layered, independent absorption intensity calculation method avoids spectral aliasing effects between different contaminant layers, ensuring accurate capture of the absorption characteristics of each layer. Ultimately, based on this accurate absorption intensity data, combined with a pre-defined physical model or calibration relationship, the thickness and spatial distribution of each contaminant layer can be precisely estimated, providing more refined and reliable contamination information for subsequent welding current parameter adjustments.

[0062] The following is a specific example to illustrate this.

[0063] Suppose that there is a layer of oil covering the rust layer on the surface of the structural component to be welded in front of the welding torch of the electric welding machine, forming a typical multi-layered contaminant superposition structure.

[0064] First, optical information of the area is acquired using a multispectral imaging sensor. Then, spectral stripping analysis is performed on the acquired optical information. This analysis utilizes the pure spectral features of oil and rust from a pre-stored feature library to decompose the mixed spectral signal into independent spectral contributions from oil and rust, thereby identifying the layered structure of the "oil layer" and "rust layer."

[0065] Next, for the identified oil layer, its absorption intensity in the characteristic spectral band of oil (e.g., a specific infrared band) is calculated; at the same time, for the identified rust layer, its absorption intensity in the characteristic spectral band of rust (e.g., a specific visible or near-infrared band) is calculated.

[0066] Finally, based on the absorption intensity of the oil and rust layers respectively, and combined with the pre-established relationship model between absorption intensity and thickness, the specific thicknesses of the oil layer (e.g., 0.05 mm) and the rust layer (e.g., 0.1 mm) are estimated, and their distribution within the welding area is determined. In this way, even with complex multi-layered contaminants, accurate information on each contaminant layer can be obtained, providing refined data support for subsequent adjustments to welding current parameters.

[0067] In some embodiments of this application, the step of adjusting the welding current parameters of the welding machine based on the contamination information before the welding arc traveling with the welding torch acts on the area to be welded preferably includes: Before the welding arc traveling with the welding torch acts on the area to be welded, the layered structure of the contamination information is identified, the layered structure including single-layer contaminants and multiple layers of contaminant superposition; When the layered structure is identified as a single-layer contaminant, the type and thickness of the contaminant are further identified, and the welding current parameters of the welding machine are adjusted by calling the adjustment mode corresponding to the type and thickness. When the layered structure is identified as a superposition of multiple contaminants, the welding current parameters of the welding machine are adjusted according to the identified layered structure, from the outermost layer to the innermost layer, and the adjustment mode corresponding to each layer of contaminants is called in turn.

[0068] Specifically, identifying the layered structure of contamination information involves in-depth analysis of acquired optical information to determine whether the contaminants on the surface of the area to be welded exist as a single layer or are composed of multiple layers of contaminants. A single layer of contaminants means that the surface of the area to be welded is covered by only one type of contaminant, such as a single layer of rust or oil. Multiple layers of contaminants refer to the presence of two or more different types of contaminants on the surface of the area to be welded, forming a layered structure, such as oil covering rust. Identifying the layered structure is a crucial step in ensuring the effectiveness of subsequent current adjustment strategies.

[0069] When a layered contaminant structure is identified as a single layer, the system further identifies the specific type of contaminant (e.g., rust, oil, oxide layer, etc.) and its thickness. Based on this detailed information, the system will retrieve the adjustment mode from a preset adjustment mode library that best matches the specific type and thickness to adjust the welding current parameters of the welding machine. For example, different energy inputs or pulse durations may be required for the same type of contaminant with different thicknesses.

[0070] When a layered structure is identified as multiple layers of superimposed contaminants, the solution in this application adjusts the welding current parameters of the welding machine sequentially, from the outermost layer to the innermost layer, using the adjustment mode corresponding to each layer of contaminants. This means that the system first applies the corresponding current adjustment mode to the outermost layer of contaminants. After that layer of contaminants is effectively treated, the corresponding adjustment mode is applied to the next layer of contaminants, until all contaminant layers are treated. This layer-by-layer treatment method ensures that each layer of contaminants is effectively and specifically removed, avoiding the impact of improper treatment in a single step on the treatment effect of subsequent layers.

[0071] This application's solution effectively overcomes the limitations of traditional methods when dealing with complex contaminants by precisely identifying the layered structure of contamination information before the welding arc acts on the area to be welded, and then adopting differentiated current adjustment strategies based on this layered structure. This layered and orderly processing method avoids the possibility of insufficient removal of internal contaminants or unnecessary damage to the substrate due to a single high-intensity treatment. It also effectively prevents external contaminants from hindering the arc's effect on internal contaminants, thus ensuring that the welding arc remains stable throughout the welding process and effectively removes contaminants at all levels.

[0072] Through the above technical solution, this application can significantly improve the welding current control accuracy and adaptability of welding machines in complex and polluted environments. Compared with solutions that adjust based solely on overall pollution information, this application, by identifying the layered structure of contaminants and adopting different, layer-by-layer adjustment strategies for single or multiple layers of contaminants, can more effectively remove various contaminants from the area to be welded, especially complex contaminants with multiple layers. This not only avoids welding defects such as porosity and slag inclusions caused by incomplete contaminant treatment, but also ensures that the welding arc remains stable throughout the welding process, thereby significantly improving welding quality and efficiency, reducing rework rates, and extending the service life of structural components.

[0073] In a specific embodiment of this application, the step of further identifying the type and thickness of the contaminant when the layered structure is identified as a single-layer contaminant, and then calling the adjustment mode corresponding to the type and thickness to adjust the welding current parameters of the welding machine preferably includes: When the layered structure is identified as a single-layer contaminant, the type and thickness of the contaminant are further identified, including rust and oil stains; If the category is rust, the welding current parameters of the welding machine are adjusted using a high-energy pulse mode. The high-energy pulse mode includes: before the arc contacts the rust layer, increasing the peak value of the arc-starting current of the welding machine to a preset first current value and maintaining it for a duration corresponding to the identified thickness. If the category is oil stains, the welding current parameters of the welding machine are adjusted using a gentle vaporization mode. The gentle vaporization mode includes: applying a preset preheating pulse before the main welding pulse of the welding machine, and reducing the peak current of the main welding pulse to a preset second current value; the energy parameters of the preheating pulse are set according to the identified thickness, and the second current value is lower than the first current value.

[0074] The contaminants include rust and oil, both common contaminants on structural components, but with significantly different physicochemical properties. Rust, mainly composed of iron oxide, is usually solid and has a certain thickness. Its presence hinders arc penetration, easily leading to welding defects such as porosity, slag inclusions, or lack of fusion. Oil, primarily organic matter, is volatile and decomposes at high temperatures, producing gases. If these gases are not expelled promptly, they can also cause porosity and spatter. Therefore, employing different welding current adjustment strategies for different types of contaminants is crucial.

[0075] When the identified contaminant is rust, the welding current parameters of the welding machine are adjusted using a high-energy pulse mode. This mode aims to effectively remove or decompose the rust layer through a burst of high energy input. Specifically, the high-energy pulse mode is implemented by increasing the peak value of the welding machine's arc-starting current to a preset first current value before the welding arc contacts the rust layer. This high current peak provides strong heat and arc force, rapidly vaporizing or melting the rust layer, thus creating a relatively clean surface for the subsequent main welding process. Simultaneously, the duration of this high current peak is set according to the identified rust layer thickness to ensure thorough rust removal and avoid affecting weld quality due to incomplete removal.

[0076] When the identified contaminant is oil, a gentle vaporization adjustment mode is used to adjust the welding current parameters of the welding machine. This mode aims to remove oil gently, avoiding welding instability caused by rapid oil vaporization. Specifically, the gentle vaporization mode is implemented by applying a preset preheating pulse before the main welding pulse of the welding machine. The energy parameters of this preheating pulse are set according to the identified oil thickness, and its function is to slowly heat the oil, causing it to gradually vaporize and evaporate, thus avoiding spatter or porosity caused by sudden decomposition of oil and the generation of large amounts of gas during the main welding process. Furthermore, the peak current of the main welding pulse is reduced to a preset second current value, which is lower than the first current value used for rust removal. Reducing the peak current of the main welding pulse helps maintain the stability of the welding process and further reduces the impact of oil vaporization on weld quality.

[0077] This application's solution, through in-depth analysis of the characteristics of different types of contaminants, designs targeted welding current adjustment modes, effectively overcoming the limitations of traditional welding methods in handling various surface contaminants. For rust, the high-energy pulse mode provides sufficient energy to rapidly remove the oxide layer before arc contact, ensuring stable arc action on the base material and avoiding problems such as insufficient penetration or slag inclusions caused by rust layer obstruction. For oil contaminants, the gentle vaporization mode achieves gradual vaporization of the oil contaminant through preheating pulses, effectively controlling the rate and amount of gas release, avoiding defects such as spatter, porosity, or hydrogen embrittlement caused by rapid oil decomposition, while reducing the peak current of the main welding pulse further ensures the stability of the welding process. This refined adjustment strategy allows the welding arc to better adapt to the actual contamination conditions of the area to be welded, thereby significantly improving the stability and quality of the weld.

[0078] The following is a specific example to illustrate this.

[0079] Suppose that during the welding process of a structural component, a single layer of contaminants is detected in the area to be welded in front of the direction of the welding torch.

[0080] Specifically, if the information recognition module identifies the single-layer contaminant as a rust layer approximately 0.8 mm thick, the parameter adjustment module will activate the high-energy pulse mode. In this mode, before the welding arc contacts the rust layer, the peak value of the welding machine's arc-starting current is increased to a preset first current value (e.g., set to 320 amps) and maintained for a duration set according to the 0.8 mm thickness (e.g., set to 60 milliseconds). This high-energy pulse can quickly break down and remove the rust layer, providing a clean surface for the subsequent main welding process.

[0081] On the other hand, if the information recognition module identifies the single-layer contaminant as an oil layer approximately 0.2 mm thick, the parameter adjustment module will activate a gentle vaporization adjustment mode. In this mode, a preset preheating pulse is applied before the main welding pulse of the welding machine, with its energy parameters set according to the 0.2 mm oil thickness (e.g., applying a preheating pulse with a peak current of 60 amps and a duration of 120 milliseconds). Subsequently, the peak current of the main welding pulse is reduced to a preset second current value (e.g., set to 200 amps), which is lower than the first current value used for rust removal. In this way, the oil is gently vaporized, effectively avoiding spatter and porosity caused by rapid decomposition of the oil, ensuring the smoothness of the welding process and the quality of the weld.

[0082] In a further embodiment of this application, the step of adjusting the welding current parameters of the welding machine in real time according to the deviation to maintain the stability of the welding arc preferably includes: When the deviation exceeds the preset adjustment threshold, the welding current parameter is adjusted compensatorily to maintain the stability of the welding arc. When the deviation is lower than or equal to the preset adjustment threshold, the current welding current parameters are maintained to keep the welding arc stable.

[0083] Specifically, the deviation refers to the difference between the real-time collected electrical parameters reflecting the welding arc state and the adjusted welding current parameters. This deviation quantifies the degree of deviation between the actual and desired state of the welding arc. The preset adjustment threshold is a key parameter, designed to distinguish between normal fluctuations and significant deviations requiring intervention. For example, this threshold can be empirically set or determined through experimental optimization based on factors such as welding materials, welding processes, equipment precision, and desired welding quality. When the deviation exceeds this threshold, it indicates that the welding arc state has significantly deviated from the target, requiring proactive compensatory adjustments. Conversely, when the deviation is below or equal to this threshold, the current welding arc state is considered to be within an acceptable stable range, requiring no additional adjustments, thus avoiding excessive intervention in minor fluctuations.

[0084] The proposed solution employs a tiered approach to address deviations in the welding arc state by introducing preset adjustment thresholds. When the deviation is small, i.e., below or equal to the preset threshold, the system determines that the current arc state is within an acceptable stable range and therefore maintains the current welding current parameters, avoiding unnecessary and frequent adjustments to normal fluctuations. This effectively reduces the burden on the control system and mitigates the risk of system oscillations or instability that may be introduced due to over-adjustment. Conversely, when the deviation is large, i.e., above the preset threshold, the system identifies a significant deviation in the arc state and immediately initiates a compensatory adjustment mechanism to correct the welding current parameters, quickly bringing the arc back to a stable state. This threshold-based, differentiated adjustment strategy enables the control system to respond more intelligently and efficiently to changes in the arc state, ensuring precise intervention when necessary while maintaining stability when not needed, thereby optimizing the overall welding process control.

[0085] In a further embodiment of this application, the step of compensatingly adjusting the welding current parameters of the welding machine to maintain a stable welding arc when the deviation exceeds a preset adjustment threshold preferably includes: When the deviation exceeds a preset adjustment threshold, a spectrum analysis is performed on the electrical parameters collected in real time. When a continuous characteristic ripple caused by abnormal internal material of the structural component is identified in the preset frequency band through the spectrum analysis, the first adjustment strategy is activated to compensate for the welding current parameter in order to maintain the stability of the welding arc. The first adjustment strategy includes: reducing the frequency of adjusting the welding current parameter and switching the current welding current mode to a constant current mode or a stable pulse mode with extended base time and reduced pulse peak current. When the continuous characteristic ripple is not identified within the preset frequency band through the spectrum analysis, the second adjustment strategy is activated to compensate for the welding current parameters in order to maintain the stability of the welding arc. The second adjustment strategy includes: according to the preset adjustment range, the welding current parameters are reverse compensated and corrected according to the deviation.

[0086] Specifically, when the deviation between the real-time acquired electrical parameters and the adjusted welding current parameters exceeds a preset adjustment threshold, spectral analysis is performed on the real-time acquired electrical parameters to more accurately diagnose the cause of arc instability. Spectral analysis refers to converting a time-domain signal into a frequency-domain signal using mathematical methods such as Fourier transform, thereby revealing the various frequency components and their intensities contained in the signal. Its purpose is to identify frequency characteristics that may be caused by specific physical phenomena.

[0087] Continuous characteristic ripple can be understood as a frequency fluctuation pattern with a certain duration and stable amplitude that appears within a specific preset frequency band in the spectrum of electrical parameters. This ripple is usually associated with anomalies in the internal materials of a structural component, such as internal defects, uneven impurity distribution, or compositional changes, resulting in a specific physical response under the action of a welding arc. When such ripple is identified, it indicates that arc instability may originate from internal material problems.

[0088] The first adjustment strategy is designed to address arc instability caused by internal material abnormalities in the structural components. This strategy aims to reduce the frequency of adjustments to welding current parameters, thereby minimizing potential additional disturbances caused by frequent adjustments, and switching to a constant current mode or a stable pulse mode with extended base time and reduced peak pulse current. The constant current mode provides stable energy input, helping to form a more uniform molten pool in areas of material abnormality; while the stable pulse mode, by extending the base time and reducing the peak pulse current, allows for longer cooling and solidification time in the molten pool, while avoiding excessively high instantaneous energy input that exacerbates instability in areas of material abnormality. This, to some extent, mitigates the impact of material abnormalities on the welding process and improves welding quality.

[0089] When spectral analysis reveals no persistent characteristic ripples caused by internal material anomalies within the preset frequency band, it indicates that arc instability may be caused by other non-material internal factors, such as external airflow disturbances, minor changes in welding torch posture, or power supply fluctuations. In this case, a second adjustment strategy is activated: the welding current parameters are reverse-compensated according to the deviation within a preset adjustment range. Reverse compensation means that if the electrical parameters are too high, the welding current parameters are reduced; if they are too low, the welding current parameters are increased to offset the deviation and restore arc stability. The preset adjustment range ensures the stability and safety of the adjustment, preventing over-adjustment from causing new instability.

[0090] This application's solution effectively distinguishes different causes of welding arc instability by introducing a step of spectral analysis of real-time acquired electrical parameters. When persistent characteristic ripples caused by internal material anomalies in the structural component are identified, the system no longer blindly performs general compensation but instead activates a specially designed first adjustment strategy. This strategy aims to reduce further disturbance to the molten pool already affected by material anomalies and provide a more stable energy input by reducing the adjustment frequency and adopting a smoother current mode, thereby maintaining arc stability under complex material conditions. Conversely, when such material anomalies are not identified, a second adjustment strategy is activated, using conventional reverse compensation correction to address other types of deviations that can usually be resolved by directly adjusting the current.

[0091] The following is a specific example to illustrate this.

[0092] Suppose that during welding, real-time acquired electrical parameters show periodic fluctuations in the welding arc voltage, and the amplitude of these fluctuations exceeds a preset adjustment threshold. In this case, the system first performs a spectral analysis on the real-time acquired arc voltage signal. If the spectral analysis shows a continuous and stable characteristic peak in a specific low-frequency band (e.g., 50-100 Hz), and this characteristic peak matches a known arc response pattern caused by inclusions or porosity within the structural component, the system determines that the current arc instability is caused by an abnormality in the internal material of the structural component. In this situation, the system will activate a first adjustment strategy: first, reduce the frequency of adjusting the welding current parameters, for example, from 10 times per second to 2 times per second, to reduce disturbance to the molten pool; second, switch the current pulse welding mode to a constant current mode, or switch to a stable pulse mode with a 20% increase in base time and a 15% reduction in peak pulse current. In this way, even with anomalies within the material, a more stable energy input can be provided, preventing the arc from fluctuating violently due to material inhomogeneity, thereby maintaining the stability of the welding arc.

[0093] Conversely, if the spectral analysis results do not identify any persistent characteristic ripples caused by abnormal internal materials of the structural component within the preset frequency band, the system will determine that the arc instability may be caused by external factors, such as slight welding torch vibration or fluctuations in shielding gas flow. In this case, the system will activate a second adjustment strategy: based on the magnitude of the arc voltage deviation, within a preset adjustment range (e.g., ±5A), the welding current parameter will be adjusted in reverse compensation. For example, if the arc voltage is too high, the welding current will be appropriately reduced; if the arc voltage is too low, the welding current will be appropriately increased until the arc voltage returns to the target range, thereby quickly and effectively correcting the deviation and maintaining the stability of the welding arc.

[0094] like Figure 2 As shown in the figure, a specific embodiment of this application also discloses an intelligent control system 200 for welding current of an electric welding machine, comprising: Information acquisition module 210 is used to acquire optical information of the area to be welded on the surface of the structural component in front of the welding torch traveling in the direction of the welding machine; Information recognition module 220 is used to identify contamination information of the area to be welded based on the optical information, wherein the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants; The parameter adjustment module 230 is used to adjust the welding current parameters of the welding machine according to the contamination information before the welding arc traveling with the welding torch acts on the area to be welded. The deviation calculation module 240 is used to collect electrical parameters reflecting the state of the welding arc in real time during the welding process, and to calculate the deviation between the collected electrical parameters and the adjusted welding current parameters. The parameter stabilization module 250 is used to adjust the welding current parameters of the welding machine in real time according to the deviation, so as to maintain the stability of the welding arc.

[0095] It should be noted that the information acquisition module 210 may consist of one or more optical sensors, such as a visible light camera, an infrared sensor, or a multispectral imaging sensor. These sensors are configured near the welding torch of the welding machine to capture images or spectral data of the area to be welded in real time or near real time. In a preferred embodiment, the information acquisition module 210 may integrate an image acquisition unit and a data transmission unit. The image acquisition unit is responsible for capturing optical information, and the data transmission unit is responsible for sending the captured information to the subsequent processing module.

[0096] The information recognition module 220 may consist of one or more processors, memory, and image processing and spectral analysis algorithms running on the processors. For example, the processor may execute specific algorithms to analyze the optical information transmitted from the information acquisition module to detect and quantify various characteristics of surface contaminants. As a specific implementation, the information recognition module 220 may include a pattern recognition unit that analyzes the optical information using a pre-trained model to identify the specific properties of the contaminants.

[0097] The parameter adjustment module 230 can be a controller that receives contamination information from the information identification module and generates corresponding welding current adjustment commands based on a preset control strategy or algorithm. These commands are then sent to the power control unit of the welding machine to change parameters such as the waveform, amplitude, and frequency of the welding current. For example, the parameter adjustment module 230 can dynamically select and apply different welding current modes, such as a high-energy pulse mode or a gentle gasification mode, based on the identified contaminant type and thickness.

[0098] The deviation calculation module 240 can consist of a current sensor, a voltage sensor, and a data processing unit. The sensors monitor the current and voltage in the welding circuit in real time and convert analog signals into digital signals. The data processing unit receives these digital signals and compares them with the target welding current parameters set by the parameter adjustment module 230, thereby calculating the deviation between the actual arc state and the target state.

[0099] The parameter stabilization module 250 can be a closed-loop controller that receives deviation values ​​from the deviation calculation module and generates compensatory adjustment commands based on a preset control algorithm (such as a PID control algorithm). These commands are sent back to the power control unit of the welding machine to correct the welding current parameters in real time, thereby maintaining the arc state within the desired stable range. For example, when a deviation of the current or voltage from the target value is detected, the parameter stabilization module 250 will immediately issue a command to increase or decrease the current output to quickly restore the stability of the arc.

[0100] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0101] The foregoing has provided a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined in this application.

Claims

1. A method for intelligent control of welding current of an electric welding machine, applied to a welding process of a structural member, characterized in that, The method includes the following steps: Acquire optical information of the area to be welded on the surface of the structural component in front of the direction of travel of the welding torch of the electric welding machine; Based on the optical information, the contamination information of the area to be welded is identified, and the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants. Before the welding arc traveling with the welding torch acts on the area to be welded, the welding current parameters of the welding machine are adjusted according to the contamination information. During the welding process, electrical parameters reflecting the state of the welding arc are collected in real time, and the deviation between the collected electrical parameters and the adjusted welding current parameters is calculated. The welding current parameters of the welding machine are adjusted in real time according to the deviation to maintain the stability of the welding arc.

2. The intelligent control method for welding current of an electric welding machine according to claim 1, characterized in that, The step of acquiring optical information of the area to be welded on the surface of the structural component in front of the welding torch traveling in the direction of travel of the welding machine includes: Multi-band image data of the area to be welded on the surface of the structural component in front of the welding torch of the electric welding machine is acquired by a multispectral imaging sensor integrated on the welding torch of the electric welding machine. Geometric and radiometric corrections are performed on the multi-band image data; Based on the corrected multi-band image data, optical information of the area to be welded on the surface of the structural component is obtained.

3. The intelligent control method for welding current of an electric welding machine according to claim 1, characterized in that, The step of identifying contamination information of the area to be welded based on the optical information, wherein the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants: Obtain a pre-stored feature library, which includes characteristic spectral bands corresponding to various pollutants; The reflectance ratios of the optical information in the characteristic spectral bands corresponding to various contaminants are analyzed to identify the quantity and type of contaminants in the area to be welded. The thickness and distribution of the contaminants are estimated by analyzing the absorption intensity of the optical information of the area to be welded where contaminants are identified in the corresponding characteristic spectral bands.

4. The intelligent control method for welding current of an electric welding machine according to claim 3, characterized in that, The pre-stored feature library also includes preset ratio thresholds for pollutants; The step of analyzing the reflectance ratios of the optical information in the characteristic spectral bands corresponding to various contaminants, and identifying the quantity and type of contaminants in the area to be welded, includes: Calculate the reflectance ratio of the optical information in the characteristic spectral bands corresponding to various pollutants; The reflectance ratios were compared with preset ratio thresholds for various pollutants to obtain the comparison results. Analyzing the comparison results, the quantity and type of contaminants in the area to be welded are obtained.

5. The intelligent control method for welding current of an electric welding machine according to claim 3, characterized in that, The steps of analyzing and identifying the absorption intensity of the optical information of the area to be welded containing contaminants in the corresponding characteristic spectral bands, and estimating the thickness and distribution of the contaminants, include: When multiple types of pollutants are identified, the optical information is subjected to spectral stripping analysis to obtain the layered structure of the pollutants; Calculate the absorption intensity of each pollutant layer in the layered structure in the corresponding characteristic spectral band; The thickness and distribution of pollutants in each layer are estimated based on the absorption intensity.

6. The intelligent control method for welding current of an electric welding machine according to claim 1, characterized in that, The step of adjusting the welding current parameters of the welding machine based on the contamination information before the welding arc traveling with the welding torch acts on the area to be welded includes: Before the welding arc traveling with the welding torch acts on the area to be welded, the layered structure of the contamination information is identified, the layered structure including single-layer contaminants and multiple layers of contaminant superposition; When the layered structure is identified as a single-layer contaminant, the type and thickness of the contaminant are further identified, and the welding current parameters of the welding machine are adjusted by calling the adjustment mode corresponding to the type and thickness. When the layered structure is identified as a superposition of multiple contaminants, the welding current parameters of the welding machine are adjusted according to the identified layered structure, from the outermost layer to the innermost layer, and the adjustment mode corresponding to each layer of contaminants is called in turn.

7. The intelligent control method for welding current of an electric welding machine according to claim 6, characterized in that, The step of further identifying the type and thickness of the contaminant when the layered structure is identified as a single-layer contaminant, and then adjusting the welding current parameters of the welding machine by calling the adjustment mode corresponding to the type and thickness, includes: When the layered structure is identified as a single-layer contaminant, the type and thickness of the contaminant are further identified, including rust and oil stains; If the category is rust, the welding current parameters of the welding machine are adjusted using a high-energy pulse mode. The high-energy pulse mode includes: before the arc contacts the rust layer, increasing the peak value of the arc-starting current of the welding machine to a preset first current value and maintaining it for a duration corresponding to the identified thickness. If the category is oil stains, the welding current parameters of the welding machine are adjusted using a gentle vaporization mode. The gentle vaporization mode includes: applying a preset preheating pulse before the main welding pulse of the welding machine, and reducing the peak current of the main welding pulse to a preset second current value; the energy parameters of the preheating pulse are set according to the identified thickness, and the second current value is lower than the first current value.

8. The intelligent control method for welding current of an electric welding machine according to claim 1, characterized in that, The step of adjusting the welding current parameters of the welding machine in real time according to the deviation to maintain a stable welding arc includes: When the deviation exceeds the preset adjustment threshold, the welding current parameter is adjusted compensatorily to maintain the stability of the welding arc. When the deviation is lower than or equal to the preset adjustment threshold, the current welding current parameters are maintained to keep the welding arc stable.

9. The intelligent control method for welding current of an electric welding machine according to claim 8, characterized in that, The step of adjusting the welding current parameters to maintain a stable welding arc when the deviation exceeds a preset adjustment threshold includes: When the deviation exceeds a preset adjustment threshold, a spectrum analysis is performed on the electrical parameters collected in real time. When a continuous characteristic ripple caused by abnormal internal material of the structural component is identified in the preset frequency band through the spectrum analysis, the first adjustment strategy is activated to compensate for the welding current parameter in order to maintain the stability of the welding arc. The first adjustment strategy includes: reducing the frequency of adjusting the welding current parameter and switching the current welding current mode to a constant current mode or a stable pulse mode with extended base time and reduced pulse peak current. When the continuous characteristic ripple is not identified within the preset frequency band through the spectrum analysis, the second adjustment strategy is activated to compensate for the welding current parameters in order to maintain the stability of the welding arc. The second adjustment strategy includes: according to the preset adjustment range, the welding current parameters are reverse compensated and corrected according to the deviation.

10. An intelligent control system for welding current in an electric welding machine, characterized in that, include: The information acquisition module is used to acquire optical information of the area to be welded on the surface of the structural component in front of the welding torch traveling in the direction of the welding machine; An information recognition module is used to identify contamination information of the area to be welded based on the optical information, wherein the contamination information includes at least one of the following: quantity, type, thickness, distribution, and layering structure of contaminants. The parameter adjustment module is used to adjust the welding current parameters of the welding machine according to the contamination information before the welding arc traveling with the welding torch acts on the area to be welded. The deviation calculation module is used to collect electrical parameters reflecting the state of the welding arc in real time during the welding process, and to calculate the deviation between the collected electrical parameters and the adjusted welding current parameters. The parameter stabilization module is used to adjust the welding current parameters of the welding machine in real time according to the deviation, so as to maintain the stability of the welding arc.

Images