Resistance spot welding quality detection method and system based on ultrasonic monitoring

Abstract

The invention discloses a resistance spot welding quality detection method and system based on ultrasonic monitoring, and relates to the technical field of ultrasonic monitoring, and the method comprises the following steps: determining a welding defect area of a welding joint according to welding seam characteristics and a plurality of ultrasonic parameters; the multiple welding defect characteristics are determined based on the welding defect area, the welding defect grade is determined based on the multiple welding defect characteristics, the corresponding spatial position and the corresponding defect grade mapping relation, and the accuracy of the welding defect grade is guaranteed. Therefore, the corresponding welding data set is collected according to the traceability of the welding defect area, and the abnormal welding parameters are determined according to the detection of the welding data set; the abnormal welding grade is determined according to the abnormal welding parameters and the resistance spot welding head, the welding quality of resistance spot welding is determined according to the welding defect grade and the mapping relation between the welding abnormal grade and the welding quality, the welding quality is detected based on the dimensions such as the welding defect grade and the welding abnormal grade, and the accuracy of the welding quality of resistance spot welding is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic monitoring, and in particular to a method and system for detecting the welding quality of resistance spot welding based on ultrasonic monitoring. Background Art

[0002] With the development of technology, the resistance spot welding process is applied to the welding process of welding joints and workpieces to be welded. The welding joint and the workpiece to be welded are subjected to resistance spot welding, and corresponding weld seams are formed. In the prior art, the weld seams are located, and the weld seam area is introduced. The welding quality of resistance spot welding is determined by detecting the weld seam area. However, the single-dimensional detection of the weld seam area results in poor accuracy of the welding quality of resistance spot welding. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a method and system for detecting the welding quality of resistance spot welding based on ultrasonic monitoring.

[0004] An embodiment of the present invention provides a method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring, including: determining a plurality of ultrasonic parameters based on the ultrasonic monitoring of a welding joint by an ultrasonic probe, where the plurality of ultrasonic parameters include the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves; locating the weld seam of the welding joint, determining weld seam features based on the dynamic image of the weld seam of the welding joint, and determining the welding defect area of the welding joint according to the weld seam features and the plurality of ultrasonic parameters; determining a plurality of welding defect features based on the welding defect area, and determining the welding defect grade based on the mapping relationship between the plurality of welding defect features, corresponding spatial positions, and corresponding defect grades; collecting a corresponding welding data set according to the traceability of the welding defect area, and determining abnormal welding parameters according to the detection of the welding data set; determining the welding abnormality grade according to the abnormal welding parameters and the resistance spot welding head, and determining the welding quality of the resistance spot welding according to the mapping relationship between the welding defect grade, welding abnormality grade, and welding quality.

[0005] An embodiment of the present invention provides a system for detecting the welding quality of resistance spot welding based on ultrasonic monitoring. The system for detecting the welding quality of resistance spot welding based on ultrasonic monitoring is applied to the above-mentioned method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring. The system for detecting the welding quality of resistance spot welding based on ultrasonic monitoring includes: An ultrasonic parameter module, configured to determine a plurality of ultrasonic parameters based on the ultrasonic monitoring of a welding joint by an ultrasonic probe, where the plurality of ultrasonic parameters include the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves; A welding defect area module, configured to locate the weld seam of the welding joint, determine weld seam features based on the dynamic image of the weld seam of the welding joint, and determine the welding defect area of the welding joint according to the weld seam features and the plurality of ultrasonic parameters; The welding defect level module is used to determine multiple welding defect features based on the welding defect area, and determine the welding defect level based on the multiple welding defect features, the corresponding spatial positions, and the corresponding defect level mapping relationships; The abnormal welding parameter module is used to collect the corresponding welding data set according to the traceability of the welding defect area, and determine the abnormal welding parameters according to the detection of the welding data set; The welding quality module is used to determine the welding abnormality level according to the abnormal welding parameters and the resistance spot welding head, and determine the welding quality of the resistance spot welding according to the welding defect level, the welding abnormality level, and the welding quality mapping relationship.

[0006] Compared with the prior art, the beneficial effects of the present invention are as follows: In the embodiment of the present invention, by the method in the embodiment of the present invention, multiple ultrasonic parameters are determined based on the ultrasonic monitoring of the welding joint by the ultrasonic probe. The multiple ultrasonic parameters include the propagation speed, attenuation coefficient, and reflection coefficient of the ultrasonic wave; the weld seam of the welding joint is located, the weld seam features are determined based on the dynamic image of the weld seam of the welding joint, and the welding defect area of the welding joint is determined according to the weld seam features and the multiple ultrasonic parameters; multiple welding defect features are determined based on the welding defect area, and the welding defect level is determined based on the multiple welding defect features, the corresponding spatial positions, and the corresponding defect level mapping relationships, which accommodates the overall consideration of the multiple welding defect features, the corresponding spatial positions, and the corresponding defect level mapping relationships, and ensures the accuracy of the welding defect level.

[0007] Therefore, the corresponding welding data set is collected according to the traceability of the welding defect area, and the abnormal welding parameters are determined according to the detection of the welding data set; the welding abnormality level is determined according to the abnormal welding parameters and the resistance spot welding head, and the welding quality of the resistance spot welding is determined according to the welding defect level, the welding abnormality level, and the welding quality mapping relationship. The overall control of the welding defect level and the welding abnormality level is introduced, and the welding quality is detected based on dimensions such as the welding defect level and the welding abnormality level, improving the accuracy of the welding quality of the resistance spot welding. Description of the Drawings

[0008] Figure 1 is a schematic flow chart of a method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring in an embodiment of the present invention; Figure 2 is a schematic flow chart of step S11 in a method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring in an embodiment of the present invention; Figure 3 is a schematic flow chart of step S12 in a method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring in an embodiment of the present invention; Figure 4It is a schematic flow chart of step S13 in a resistance spot welding quality detection method based on ultrasonic monitoring in an embodiment of the present invention; Figure 5 It is a schematic flow chart of step S14 in a resistance spot welding quality detection method based on ultrasonic monitoring in an embodiment of the present invention; Figure 6 It is a schematic flow chart of step S15 in a resistance spot welding quality detection method based on ultrasonic monitoring in an embodiment of the present invention; Figure 7 It is a schematic diagram of the structural composition of a resistance spot welding quality detection system based on ultrasonic monitoring in an embodiment of the present invention. Detailed implementation manners

[0009] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention.

[0010] Please refer to Figures 1 to 7 , a resistance spot welding quality detection method based on ultrasonic monitoring, which is applied to a resistance spot welding quality detection scenario based on ultrasonic monitoring of an in-built induction cooker; a resistance spot welding quality detection method based on ultrasonic monitoring includes: Step S11: Determine a plurality of ultrasonic parameters based on the ultrasonic monitoring of the welding joint by an ultrasonic probe, and the plurality of ultrasonic parameters include the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves; Step S12: Locate the weld seam of the welding joint, determine the weld seam characteristics based on the dynamic image of the weld seam of the welding joint, and determine the welding defect area of the welding joint according to the weld seam characteristics and the plurality of ultrasonic parameters; Step S13: Determine a plurality of welding defect characteristics based on the welding defect area, and determine the welding defect grade based on the mapping relationship between the plurality of welding defect characteristics, the corresponding spatial positions, and the corresponding defect grades; Step S14: Collect the corresponding welding data set according to the traceability of the welding defect area, and determine the abnormal welding parameters according to the detection of the welding data set; Step S15: Determine the welding abnormality grade according to the abnormal welding parameters and the resistance spot welding head, and determine the welding quality of the resistance spot welding according to the mapping relationship between the welding defect grade, the welding abnormality grade, and the welding quality; Refer to Figure 2 , in step S11, determine a plurality of ultrasonic parameters based on the ultrasonic monitoring of the welding joint by an ultrasonic probe, and the plurality of ultrasonic parameters include the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves; In the specific implementation process of the present invention, the specific steps are as follows: S111: Perform resistance spot welding on the welded joint and weld the welded joint to the workpiece to be welded. At the same time, the ultrasonic monitor is arranged obliquely relative to the welded joint and performs ultrasonic monitoring on the welded joint; S112: When the ultrasonic probe performs ultrasonic monitoring on the welded joint, collect a plurality of ultrasonic signals, and determine an ultrasonic signal set according to the plurality of ultrasonic signals and the welding position of the welded joint and the workpiece to be welded; S113: Input the ultrasonic signal set into a preset ultrasonic parameter learning model. The ultrasonic parameter learning model classifies the ultrasonic signal set and forms a propagation combination, an attenuation combination, and a reflection combination of ultrasonic waves. The propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves are generated based on further identification of the propagation combination, attenuation combination, and reflection combination of ultrasonic waves.

[0011] In the embodiment of the present application, resistance spot welding is performed on the welded joint and the welded joint is welded to the workpiece to be welded. At the same time, the ultrasonic monitor is arranged obliquely relative to the welded joint and performs ultrasonic monitoring on the welded joint, realizing ultrasonic monitoring of the welded joint.

[0012] At this time, ensure that the welded joint and the workpiece to be welded have been cleaned and pretreated to remove impurities such as oil stains and oxides to ensure good welding quality; assemble the welded joint and the workpiece to be welded according to the design requirements to ensure that the position, angle, and gap of the joint meet the process requirements; set appropriate welding current, voltage, welding time, electrode pressure and other parameters according to the material and thickness of the welded joint; start the resistance spot welding equipment and weld the welded joint; during the welding process, the electrodes apply pressure and conduct electricity to generate resistance heat to melt and connect the joints together.

[0013] Optionally, assume that a steel plate component of an automobile body is to be welded; first, clean and pretreat the steel plate joint and the body part to be welded; then, assemble the joint and the body part according to the design requirements to ensure that the position and gap of the joint meet the process requirements; then, according to the material and thickness of the steel plate, set the welding current to 8000A, the voltage to 24V, the welding time to 0.2 seconds, and the electrode pressure to 4000N; finally, start the resistance spot welding equipment and weld the steel plate joint to form a firm connection.

[0014] After welding, check the appearance and internal quality of the welded joint to ensure that there are no defects such as cracks, slag inclusions, and lack of fusion; use measuring tools to check the size and position of the welded joint to ensure that they meet the design requirements; if defects or non - compliance with the design requirements are found in the welded joint, necessary corrections or rework are required.

[0015] Optionally, after welding is completed, methods such as visual inspection and magnetic particle inspection are used to check the appearance and internal quality of the steel plate joint to ensure there are no defects such as cracks and slag inclusions. At the same time, a caliper and measuring tools are used to check the dimensions and positions of the joint to ensure they meet the design requirements of the automobile body. If defects or non-conformities are found in the joint, necessary corrections or rework will be carried out to ensure the welding quality.

[0016] Select a suitable ultrasonic monitor to ensure that parameters such as its frequency, sensitivity, and resolution meet the detection requirements. Arrange the ultrasonic monitor obliquely relative to the welded joint to optimize the propagation path and reception effect of the ultrasonic wave. The inclination angle should be selected according to the shape, size of the welded joint, and the characteristics of the ultrasonic probe. During or after welding, start the ultrasonic monitor to continuously monitor the welded joint. During the monitoring process, record and analyze the ultrasonic signals to evaluate the internal quality and defect conditions of the welded joint.

[0017] Optionally, after welding the steel plate joint is completed, select an ultrasonic monitor with a frequency of 5 MHz for detection. Arrange the ultrasonic monitor obliquely at 45 degrees relative to the steel plate joint to ensure that the ultrasonic wave can fully penetrate the joint and receive the reflected signal. Then, start the ultrasonic monitor to continuously monitor the steel plate joint. During the monitoring process, record and analyze the ultrasonic signals, and it is found that there is a small porosity defect inside the joint. According to the analysis results, the defect is marked and recorded for subsequent processing and improvement.

[0018] Furthermore, when the ultrasonic probe monitors the welded joint, collect multiple ultrasonic signals, and determine an ultrasonic signal set based on the multiple ultrasonic signals and the welding area between the welded joint and the workpiece to be welded, taking into account the overall consideration of multiple ultrasonic signals and the welding area between the welded joint and the workpiece to be welded, ensuring the accuracy of the ultrasonic signal set.

[0019] At this time, select a suitable ultrasonic probe according to the material and thickness of the welded joint, and perform necessary calibration to ensure the accuracy of the measurement. Determine the monitoring position of the ultrasonic probe, which is above or on the side of the welded joint, and plan the monitoring path according to the shape and size of the joint. Start the ultrasonic monitoring equipment, and make the probe scan the welded joint along the predetermined path while recording the ultrasonic signals.

[0020] Optionally, assume that an ultrasonic monitoring is to be carried out on an aluminum alloy welded joint with a thickness of 10 mm. First, select an ultrasonic probe with a frequency of 2.5 MHz and calibrate it to ensure its accuracy. Then, place the probe above the welded joint and plan a straight scan path from one end of the joint to the other end. Finally, start the ultrasonic monitoring equipment, and make the probe scan the welded joint along the predetermined path and record the ultrasonic signals.

[0021] During the ultrasonic monitoring process, the probe will receive multiple ultrasonic signals from inside the welded joint. These signals contain information about the internal structure, defects, and material properties of the joint. The collected ultrasonic signals are stored in a computer or other storage devices for subsequent analysis and processing. Optionally, during the ultrasonic monitoring process, the probe receives multiple ultrasonic signals from inside the aluminum alloy welded joint, and these signals are recorded and stored in digital form on the hard disk of the computer. Each signal contains information about the internal structure, defects (such as pores, cracks, etc.), and the properties of the aluminum alloy material.

[0022] Select signals directly related to the welding area between the welded joint and the workpiece to be welded from the multiple collected ultrasonic signals. These signals are located in the central area of the welded joint and reflect the welding quality and internal defect conditions of the joint. Combine the selected signals into an ultrasonic signal set for subsequent analysis and processing.

[0023] Optionally, after the acquisition and storage of the ultrasonic signals are completed, the signal screening starts. Note that the ultrasonic signals located in the central area of the welded joint are directly related to the welding area and reflect the welding quality and internal defect conditions of the joint. Therefore, these signals are screened out and combined into an ultrasonic signal set, which contains all ultrasonic signals directly related to the welding area of the welded joint, providing basic data for subsequent analysis and processing.

[0024] Therefore, input the ultrasonic signal set into a preset ultrasonic parameter learning model. The ultrasonic parameter learning model classifies the ultrasonic signal set and forms combinations of ultrasonic wave propagation, attenuation, and reflection. Based on the further identification of the combinations of ultrasonic wave propagation, attenuation, and reflection, the propagation speed, attenuation coefficient, and reflection coefficient of the ultrasonic wave are generated. At this time, the preset ultrasonic parameter learning model is an algorithm model based on machine learning or deep learning. It has been trained to extract key parameters from ultrasonic signals. Use the ultrasonic signal set formed in step S112 as input data and transfer it to the preset ultrasonic parameter learning model. Before inputting into the model, it is necessary to preprocess the ultrasonic signal set, such as denoising, filtering, normalization, etc., to improve the recognition accuracy of the model.

[0025] Optionally, assume there is an ultrasonic parameter learning model based on a convolutional neural network (CNN). This model has been trained to accurately extract the propagation speed, attenuation coefficient, and reflection coefficient from ultrasonic signals. Use the ultrasonic signal set screened and combined in step S112 as input data and transfer it to this CNN model. Before inputting, the signal set is denoised and normalized to ensure that the model can accurately identify.

[0026] The ultrasonic parameter learning model first extracts features from the input ultrasonic signals, including the amplitude, frequency, phase, etc. of the signals; based on the extracted features, the model classifies the ultrasonic signals into different categories, which are related to the propagation, attenuation, and reflection characteristics of ultrasonic waves; according to the classification results, the model combines similar signals together to form the propagation combination, attenuation combination, and reflection combination of ultrasonic waves.

[0027] Optionally, the CNN model extracts features from the input set of ultrasonic signals, extracting key features such as the amplitude, frequency, and phase of the signals; then, based on these features, the model classifies the signals into three categories related to propagation characteristics, attenuation characteristics, and reflection characteristics; finally, the model combines the classified signals into a propagation combination, an attenuation combination, and a reflection combination respectively.

[0028] For each combination, the model further identifies and estimates the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves, which involves techniques such as statistical analysis, pattern matching, or regression analysis of the signals within the combination; the ultrasonic parameter learning model takes the estimated ultrasonic parameters as the output results, and these parameters can be used for subsequent weld quality assessment or defect detection.

[0029] Optionally, for the propagation combination, attenuation combination, and reflection combination, the CNN model respectively performs further identification and parameter estimation; by statistically analyzing the amplitude and phase changes of the signals within the combination, the model estimates the propagation speed of ultrasonic waves in the aluminum alloy welded joint; by using pattern matching techniques, the model identifies the attenuation characteristics of the signals and estimates the attenuation coefficient; similarly, by using regression analysis techniques, the model estimates the reflection coefficient; finally, the CNN model outputs the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves, and these parameters can be used for subsequent weld quality assessment or defect detection; through the detailed explanation of the above steps and the illustration of specific examples, each small step in step S113 and their execution methods in practical applications can be understood more clearly; in actual operation, the accuracy and meticulousness of these steps are crucial for ensuring the accuracy and reliability of ultrasonic parameter extraction.

[0030] In an embodiment of the present application, assuming there is an aluminum alloy welded joint, after ultrasonic monitoring and parameter extraction, the ultrasonic parameter learning model outputs the following ultrasonic parameters: propagation speed: 5300 m / s; attenuation coefficient: 0.02 dB / mm; reflection coefficient: 0.75, and these parameters reflect the internal structure and material properties of the welded joint and can be used to evaluate the quality of the weld and detect potential defects.

[0031] Reference Figure 3, in step S12, locate the weld of the welded joint, determine the weld features based on the dynamic image of the weld of the welded joint, and determine the welding defect area of the welded joint according to the weld features and multiple ultrasonic parameters; In the specific implementation process of the present invention, the specific steps are as follows: S121: Dynamically monitor the welding of the welded joint, and gradually form the weld of the welded joint with the resistance spot welding of the welded joint. Collect the dynamic image of the weld of the welded joint, and generate multiple sub-weld areas according to the dynamic image of the weld of the welded joint and the division of the surface color difference of the welded joint; S122: Synchronously identify multiple sub-weld areas, and determine the weld features according to the positions and corresponding regional shapes of the multiple sub-weld areas; S123: Collect the material of the welded joint and the distribution map of the welded joint, determine the first defect area according to the weld features and the material of the welded joint, determine the second defect area according to the weld features and multiple ultrasonic parameters, and determine the welding defect area of the welded joint based on the first defect area, the second defect area and the distribution map of the welded joint; In the embodiment of the present application, the welding of the welded joint is dynamically monitored, and the weld of the welded joint is gradually formed with the resistance spot welding of the welded joint. The dynamic image of the weld of the welded joint is collected, and multiple sub-weld areas are generated according to the dynamic image of the weld of the welded joint and the division of the surface color difference of the welded joint. Multiple sub-weld areas are introduced to further control the multiple sub-weld areas.

[0032] At this time, monitoring devices such as a high-definition camera or an infrared thermal imager are used to monitor the welded joint in real time and continuously; pay attention to key parameters such as the temperature change, material melting state, electrode pressure and welding current of the welded joint; record and store the monitored data in real time for subsequent analysis and processing.

[0033] Optionally, a high-definition camera is used to dynamically monitor the welded joint during the resistance spot welding process; the camera is installed above the welding machine to shoot the changes of the welded joint from a vertical perspective; at the same time, a temperature sensor and a current sensor are also connected to monitor the temperature change and welding current during the welding process in real time.

[0034] Apply pressure to the welded joint through the electrodes and energize to generate resistance heat, so that the welded joint is locally melted and a weld is formed; as the welding process progresses, the weld gradually starts to form from the contact surface of the welded joint and expands outward; the quality of the weld is affected by welding parameters (such as welding current, electrode pressure, welding time, etc.) and the material characteristics of the welded joint.

[0035] Optionally, during the resistance spot welding process, it was observed that the welded joint gradually came into contact and closely fit under the action of the electrode pressure; with the passage of the welding current, the local part of the joint began to melt and gradually formed a clear weld seam; the width and depth of the weld seam gradually increased as the welding process progressed until the preset welding parameter requirements were met.

[0036] Use a high-definition camera or other image acquisition devices to capture the dynamic changes of the weld seam in real time; set an appropriate image acquisition frequency according to the welding speed and the weld seam formation speed to ensure that each key stage of the weld seam formation is captured; ensure that the acquired images are clear and stable, and can accurately reflect the shape and color changes of the weld seam.

[0037] Optionally, the dynamic changes of the weld seam were captured in real time using a high-definition camera; the acquisition frequency of the camera was set to 10 frames per second to ensure that each key stage of the weld seam formation could be captured; at the same time, the focal length and exposure time of the camera were also adjusted to ensure that the acquired images were clear and stable.

[0038] By analyzing the dynamic images of the weld seam, observe the color changes on the surface of the weld seam; the color changes are caused by factors such as temperature differences, material melting states, or internal defects in the weld seam; according to the color differences on the surface of the weld seam, divide the weld seam into multiple sub-regions; each sub-region represents a different welding state or potential quality problem; record and store the divided sub-weld seam regions and their related information (such as position, size, color, etc.) for subsequent analysis and processing.

[0039] Optionally, when observing the dynamic images of the weld seam, it was found that there were obvious color differences on the surface of the weld seam; among them, a part of the weld seam had a lighter color, indicating that the temperature in this area was lower or the material was not fully melted; another part of the weld seam had a darker color, indicating that the temperature in this area was too high or there were internal defects; therefore, the weld seam was divided into two sub-regions: the light-colored region and the dark-colored region; at the same time, the position, size, color and other information of these two sub-regions were also recorded for further analysis and processing later.

[0040] Furthermore, synchronously identify multiple sub-weld seam regions, determine the weld seam characteristics according to the positions and corresponding regional morphologies of multiple sub-weld seam regions, taking into account the overall positions and corresponding regional morphologies of multiple sub-weld seam regions, and ensure the accuracy of the weld seam characteristics.

[0041] At this time, technologies such as image recognition, machine learning, or computer vision are used to synchronously identify multiple sub-weld regions. These technologies can automatically analyze features in the image, such as color, shape, texture, etc., so as to identify different sub-weld regions; ensure that the accuracy of the recognition technology is high enough to accurately distinguish different sub-weld regions and avoid misrecognition or missed recognition; since the welding process is continuous, it is necessary to ensure that the recognition of multiple sub-weld regions is carried out synchronously to capture the real-time changes during the welding process.

[0042] Optionally, assume there are two sub-weld regions, labeled as Region A and Region B respectively; use image recognition technology to synchronously identify these two regions through a trained model; the model can automatically analyze features such as color, shape, and texture in the image and accurately identify Region A and Region B; at the same time, since the recognition technology is carried out synchronously, it can capture the changes during the welding process in real time, such as the width, depth, and surface morphology of the weld.

[0043] Analyze the positions of multiple sub-weld regions, which includes determining the position of each sub-weld region in the overall weld, the relationship between adjacent regions, and their relative positions to the welding joint, etc.; then, analyze the morphology of each sub-weld region, which includes observing features such as the width, depth, shape, and surface texture of the weld to understand the quality and defects of the weld; comprehensively considering the results of position analysis and morphology analysis, determine the overall characteristics of the weld, and these characteristics include the continuity, uniformity, surface quality, and potential defects of the weld.

[0044] Optionally, after identifying Region A and Region B, further analyze their positions and morphologies; by observing the image, it is found that Region A is located at the starting part of the weld, its shape is relatively regular, and the width and depth are uniform; while Region B is located in the middle of the weld, its shape is slightly irregular, and there are some tiny cracks on the surface; based on this information, it can be determined that the overall characteristics of the weld are: the quality of the starting part is good, and there are potential defects in the middle; at the same time, since Region A and Region B are adjacent in position, it can also be inferred the quality changes that occur during the continuous welding process of the weld.

[0045] Therefore, collect the material of the welding joint and the distribution map of the welding joint, determine the first defect region according to the weld characteristics and the material of the welding joint, determine the second defect region according to the weld characteristics and multiple ultrasonic parameters, and determine the welding defect region of the welding joint based on the first defect region, the second defect region, and the distribution map of the welding joint, which comprehensively considers the first defect region, the second defect region, and the distribution map of the welding joint and ensures the accuracy of the welding defect region of the welding joint.

[0046] At this time, obtain the material information of the welded joint through material testing or by referring to relevant technical documents, including key parameters such as the type, composition, strength, and toughness of the material. This information is crucial for evaluating the strength and durability of the weld seam; draw a distribution map of the welded joints according to the actual layout and arrangement of the welded joints; the distribution map should clearly show information such as the position, quantity, shape, and size of each welded joint, which helps to accurately locate and analyze weld defects in the follow-up.

[0047] Optionally, assume that the welding quality of a car body is being inspected; first, obtain the material information of the steel used in the body through referring to relevant technical documents and conducting material tests, including key parameters such as its tensile strength, yield strength, and toughness; then, draw a detailed distribution map of the welded joints according to the actual layout of the body's welded joints; the distribution map clearly shows information such as the position, quantity, shape, and size of each welded joint, providing an important reference for subsequent quality inspections.

[0048] Through comprehensive analysis of information such as the dynamic image, morphology, and position of the weld seam, determine the main characteristics of the weld seam, such as the width, depth, shape, and surface quality of the weld seam; match the weld seam characteristics with the material information of the welded joint to evaluate whether the strength and durability of the weld seam meet the expectations; if the weld seam characteristics do not match the material information, there are potential defects; for the first defect area, based on the matching results of the weld seam characteristics and the material information, determine the first defect area, where the weld seam quality fails to meet the standards due to material mismatch, improper welding parameters, or other problems during the welding process.

[0049] Optionally, after obtaining the weld seam characteristics and the material information of the welded joint, conduct a matching analysis; it is found that at some welded joints, the width and depth of the weld seam are uneven, and there are obvious cracks and pores on the surface; at the same time, the material of the steel used for these welded joints does not match the expectations, and its tensile strength and toughness are lower than the requirements; based on this information, determine the first defect area, which is mainly concentrated in some key structural parts of the body, such as the frame connection and the door hinge installation.

[0050] Use ultrasonic testing equipment to conduct non-destructive testing on the weld seam to obtain multiple ultrasonic parameters, such as the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves. These parameters can reflect the internal microstructure and defects of the weld seam; analyze the collected ultrasonic parameters to evaluate whether there are defects inside the weld seam, such as cracks, inclusions, pores, etc.; at the same time, make a comprehensive judgment in combination with the weld seam characteristics to improve the accuracy of defect detection; for the second defect area, based on the analysis results of the ultrasonic parameters and the comprehensive judgment of the weld seam characteristics, determine the second defect area, where the quality fails to meet the standards due to internal microstructure problems or abnormal ultrasonic parameters of the weld seam.

[0051] Optionally, after determining the first defect area, a non-destructive test was performed on the weld using an ultrasonic testing device; by collecting and analyzing multiple ultrasonic parameters, it was found that at some welded joints, the propagation speed of ultrasonic waves decreased significantly, the attenuation coefficient increased, and the reflection coefficient was abnormal; combined with the comprehensive judgment of the weld characteristics, the second defect area was determined, and these areas were mainly concentrated in some hidden parts of the vehicle body, such as under the floor and inside the trunk. These defects were caused by improper operation or material problems during the welding process.

[0052] Overlay and compare the distribution maps of the first defect area, the second defect area, and the welded joints, and comprehensively judge the overall quality status of the welded joints; based on the results of the comprehensive judgment, determine the welded defect areas of the welded joints. These areas have unqualified weld quality due to various factors and need to be repaired or replaced; for the determined welded defect areas, formulate corresponding repair plans or replacement plans to ensure the quality and safety of the welded joints.

[0053] Optionally, after comprehensively judging the distribution maps of the first defect area, the second defect area, and the welded joints, the welded defect areas of the welded joints were determined. These areas were mainly concentrated in some key structural parts and hidden parts of the vehicle body, such as the frame connection, the door hinge installation, under the floor, and inside the trunk; for these defect areas, detailed repair plans were formulated, including measures such as grinding, repair welding, and replacing welded joints, to ensure the welding quality and safety of the vehicle body; through the detailed explanation of the above steps and the description of specific examples, each small step in step S123 and their implementation methods in practical applications can be understood more clearly. The accuracy and meticulousness of these steps are crucial for ensuring the quality and safety of the welded joints.

[0054] In an embodiment of the present application, record the width, depth, shape, surface quality, etc. of the weld; compare the weld characteristics with the material information of the welded joints to identify the defect areas; introduce a first defect area matching table, and the first defect area matching table is shown in Table 1: Table 1 First Defect Area Matching Table

[0055] Record the propagation speed, attenuation coefficient, reflection coefficient, etc. of ultrasonic waves; combine the weld characteristics and ultrasonic parameters to identify the internal defects of the weld; introduce a second defect area matching table, and the second defect area matching table is shown in Table 2: Table 2 Second Defect Area Matching Table

[0056] Determine the welding defect area of the welding joint based on the distribution maps of the first defect area, the second defect area, and the welding joint. At this time, overlay the information of the first defect area and the second defect area with the distribution map of the welding joint to determine the final welding defect area; introduce a final defect area matching table, and the final defect area matching table is shown in Table 3: Table 3 Final Defect Area Matching Table

[0057] Reference Figure 4 , in step S13, determine multiple welding defect features based on the welding defect area, and determine the welding defect grade based on the mapping relationships of the multiple welding defect features, the corresponding spatial positions, and the corresponding defect grades; In the specific implementation process of the present invention, the specific steps are as follows: S131: Mark the location of the welding defect area, and determine multiple welding defect features according to the location of the welding defect area and the shape of the defect area. The multiple welding defect features include pores, cracks, weld beads, or arc pits; S132: Determine the spatial positions of the multiple welding defect features based on the multiple welding defect features and the distribution map of the welding joint, generate a corresponding welding transition area according to two adjacent welding defect features, and mark the area location of the welding transition area; S133: Match the multiple welding defect features and the area locations of the welding transition areas with the corresponding defect grade mapping relationships, and generate corresponding welding defect grades, which are used to present the quality grades of the resistance spot welding in the dimension of welding defects.

[0058] In the embodiment of the present application, mark the location of the welding defect area, and determine multiple welding defect features according to the location of the welding defect area and the shape of the defect area. The multiple welding defect features include pores, cracks, weld beads, or arc pits, which takes into account both the location of the welding defect area and the shape of the defect area, ensuring the accuracy of the multiple welding defect features.

[0059] At this time, accurately record the specific location of the welding defect on the workpiece or the welding joint for subsequent analysis and processing; physical marks (such as scribing, dotting) or digital marks (such as marking in CAD drawings, digital models) can be used to mark the defect location; physical marks are suitable for on-site operations, while digital marks are more suitable for subsequent data analysis and report preparation; carefully observe the shape of the welding defect, including size, shape, color, surface condition, etc.; at the same time, consider the location of the defect on the workpiece, such as whether it is close to the weld edge, whether it is located in the stress concentration area, etc.; according to the results of observation and analysis, combined with the common types of welding defects (such as pores, cracks, weld beads, arc pits, etc.), determine multiple welding defect features.

[0060] Optionally, assume that on a resistance spot welding production line in an automobile manufacturing plant, a quality inspector is inspecting a batch of welded joints; he uses ultrasonic testing equipment to scan the welded joints and observes two obvious abnormal areas on the display screen; the quality inspector first draws a thin line on the first abnormal area and marks it with the number "D1". This area is located in the central part of the weld, near a stress concentration point; then, he also draws a thin line on the second abnormal area and marks it with the number "D2". This area is located at the edge of the weld, at the junction with the base metal.

[0061] For the "D1" area, the quality inspector notices that the ultrasonic signal has obvious reflections in this area and the reflected signal is relatively strong; he further observes that the surface of this area is slightly convex and irregular in shape; combining these characteristics, he judges that the "D1" area is a weld bead defect; for the "D2" area, the quality inspector finds that the ultrasonic signal suddenly interrupts in this area, forming an obvious "void" effect; he carefully observes the surface of this area and finds that there are some small cracks extending from the edge of the weld to the base metal; therefore, he judges that the "D2" area is a crack defect.

[0062] Furthermore, based on multiple welding defect characteristics and the distribution map of the welded joint, determine the spatial positions of multiple welding defect characteristics, generate corresponding welding transition regions according to two adjacent welding defect characteristics, and mark the regional positions of the welding transition regions, which takes into account the overall consideration of multiple welding defect characteristics and the distribution map of the welded joint, and ensures the accuracy of the spatial positions of multiple welding defect characteristics.

[0063] At this time, clarify the specific positions of each welding defect characteristic in three-dimensional space and their relationships with the overall layout of the welded joint; use the three-dimensional model or distribution map of the welded joint as a reference, and combine the previously determined welding defect characteristics (such as pores, cracks, weld beads, arc pits, etc.) to accurately mark the spatial positions of each defect on the model or drawing, which involves locating the coordinates of the defect in three-dimensional space or locating the defect on a two-dimensional drawing through a scale and reference lines.

[0064] Identify the potential influence areas between adjacent defects, which exhibit special properties or risks due to the interaction of defects; analyze the spatial positions and morphologies of two adjacent welding defect characteristics to determine the transition region between them. This region is a continuous region with a gradually changing morphology and is also an acute or obtuse angle region formed by the intersection of the edges of the two defects; depending on the specific situation, the transition region can be a geometric shape (such as a rectangle, circle, ellipse, etc.) or a more complex region defined by curves or irregular shapes.

[0065] Clearly mark the position of the welding transition area on the 3D model, 2D drawing or actual workpiece for subsequent analysis, repair or monitoring; use the same or similar methods as those for marking the characteristics of welding defects (such as physical marking, digital marking, etc.) to mark the position of the welding transition area on the 3D model, 2D drawing or actual workpiece, which can be achieved by means such as scribing, painting, marking text or numbers.

[0066] Specifically, assume that during the steel structure welding process of a bridge construction project, the quality inspector has discovered several welding defects through non-destructive testing techniques and has determined the spatial positions and characteristics of these defects according to the previous steps; now, he needs to perform step S132 to determine the welding transition area; the quality inspector uses the 3D model of the bridge steel structure as a reference and accurately marks the spatial position of each defect on the model according to the previously determined welding defect characteristics (such as pores, cracks, etc.); for example, he has discovered a pore (marked as D1) located at the center of the weld seam and a crack (marked as D2) located at the edge of the weld seam.

[0067] Analyzing the spatial positions and shapes of D1 and D2, the quality inspector finds that they are adjacent and not far apart; considering that both the pore and the crack have an adverse impact on the strength of the weld seam, he decides to generate a welding transition area to cover these two defects and the potential affected areas around them. This transition area is defined as an elliptical area, with its major axis along the connection line between D1 and D2, and the minor axis determined according to the size and shape of the defects; on the 3D model, the quality inspector uses different colors or line styles to mark the position of the welding transition area; for example, he selects a prominent red color to fill the elliptical transition area and marks a text description (such as "Transition Area - D1&D2") on the area edge. In this way, subsequent analysts or repairers can clearly see the position and scope of the transition area. Through the above steps, the quality inspector has successfully determined the position of the welding transition area and made corresponding marks, and this information is very important for subsequent defect analysis, repair plan formulation and quality monitoring.

[0068] Therefore, match the relationships between multiple welding defect characteristics, the regional positions of the welding transition areas and the corresponding defect grades, and generate the corresponding welding defect grades. These welding defect grades are used to present the quality grades of resistance spot welding in the dimension of welding defects, taking into account the overall consideration of multiple welding defect characteristics, the corresponding spatial positions and the corresponding defect grade mapping relationships, ensuring the accuracy of the welding defect grades.

[0069] At this time, in order to quantitatively evaluate welding defects, a mapping relationship between defect characteristics, transition regions, and defect grades needs to be established, and this relationship is determined based on industry standards, enterprise specifications, or empirical data; according to factors such as the severity of welding defects, their impact on structural performance, and the difficulty of repair, different defect characteristics and transition regions are classified into different grades, and these grades can be represented by numbers (such as 1, 2, 3, etc.) or letters (such as A, B, C, etc.), and each grade corresponds to the nature of specific defect characteristics and transition regions.

[0070] Match the actually detected welding defect characteristics and transition regions with the established mapping relationship of defect grades to determine the grades they belong to; at this time, check each welding defect characteristic and transition region one by one, and compare them with the grade standards in the mapping relationship according to factors such as their morphology, size, position, and relationship with other defects to find the most suitable grade. Based on the above matching results, assign a specific defect grade to each welding defect characteristic and transition region; at this time, record the matched grade in the corresponding report or database for subsequent analysis and processing, and this grade can be used as a quality grade index for resistance spot welding in the dimension of welding defects, for evaluating welding quality, formulating repair plans, or carrying out quality improvement.

[0071] Specifically, assume that in a resistance spot welding project, the quality inspector has completed the detection and marking of welding defects and determined the location of the welding transition region; now, he needs to generate welding defect grades according to step S133; the quality inspector refers to industry standards and establishes a mapping relationship of defect grades; for example, he classifies defects such as pores and cracks into three grades according to their size, quantity, and position: minor (grade 1), medium (grade 2), and severe (grade 3); for the welding transition region, he also classifies it into three grades according to factors such as the area, shape of the transition region, and its relationship with adjacent defects.

[0072] The quality inspector inspected the detected welding defect characteristics and transition regions one by one; for example, he found that a porosity located at the center of the weld seam was small and few in number, and according to the mapping relationship, he determined it to be of a minor grade (grade 1); another crack located at the edge of the weld seam was large and had a long extension, and he determined it to be of a severe grade (grade 3); for the welding transition region, he found a transition region connecting two medium-grade defects, with a moderate area and regular shape, and he determined it to be of a medium grade (grade 2). The quality inspector recorded the grade of each welding defect characteristic and transition region in the report; for example, he recorded information such as the porosity being grade 1, the crack being grade 3, and the transition region being grade 2, which will be used for subsequent welding quality assessment, repair plan formulation, and quality improvement work; through the above steps, the quality inspector successfully generated the welding defect grades, providing an important basis for the quality management and control of the resistance spot welding project.

[0073] In an embodiment of the present application, a defect grade matching table is introduced, and the defect grade matching table is shown in Table 4: Table 4 Defect Grade Matching Table

[0074] At this time, assume that in a resistance spot welding quality inspection, the following welding defects and transition regions are found: a small and few-porosity (matching grade 1); a short crack that does not penetrate the weld seam (matching grade 2); a large weld bead that does not affect the appearance and strength of the weld seam (matching grade 1, although large in size, but according to the standard, as long as it does not affect the strength and appearance, it is still determined to be minor); a welding transition region with a small area, regular shape, and connecting minor defects (matching grade 1); according to the defect grade matching table, a corresponding welding defect grade report can be generated: porosity: grade 1 (minor); crack: grade 2 (medium); weld bead: grade 1 (minor); welding transition region: grade 1 (minor).

[0075] Reference Figure 5 , in step S14, according to the traceability of the welding defect area, the corresponding welding data set is collected, and abnormal welding parameters are determined based on the detection of the welding data set; In the specific implementation process of the present invention, the specific steps are as follows: S141: Collect the welding defect area, mark the corresponding traceability nodes for the welding defect area, and achieve the positioning traceability of the welding defect area according to the trigger of the traceability nodes. Based on the positioning traceability of the welding defect area, collect the corresponding welding data set; S142: Determine multiple groups of welding data combinations based on the traversal of the welding data set. The multiple groups of welding data combinations are welding data combinations of different dimensions, covering the welding current dimension, welding voltage dimension, welding angle dimension, and welding speed dimension; S143: Determine corresponding abnormal welding parameters according to the abnormal recognition of multiple groups of welding data combinations. The abnormal welding parameters include abnormal welding current, abnormal welding voltage, abnormal welding angle, or abnormal welding speed.

[0076] In the embodiments of the present application, collect the welding defect area, mark the corresponding traceability nodes for the welding defect area, realize the positioning traceability of the welding defect area according to the triggering of the traceability nodes, and collect the corresponding welding data set based on the positioning traceability of the welding defect area; At this time, collect the welding defect area: mark the corresponding traceability nodes for the welding defect area; establish the association between the defect area and relevant parameters or events during the welding process to trace the cause of the defect; at this time, during the welding process, set multiple traceability nodes, and these nodes can be time stamps, change points of welding parameters, state changes of welding equipment, etc.; when a defect is detected, according to the position and time of the defect, trace back to the nearest traceability node and mark it as the node related to the defect.

[0077] Realize the positioning traceability of the welding defect area according to the triggering of the traceability nodes. By triggering the traceability nodes, accurately trace the specific position and time when the defect occurs, as well as the related welding parameters or events; at this time, use the recorded data during the welding process (such as welding parameter records, equipment state records, etc.), combined with the marked traceability nodes, to perform positioning traceability, which involves a detailed analysis of the welding data to find out the specific factors leading to the defect.

[0078] Collect the corresponding welding data set based on the positioning traceability of the welding defect area. Collect the welding data related to the defect for subsequent analysis and improvement; at this time, according to the results of the positioning traceability, extract the welding data related to the occurrence of the defect, including parameters such as welding current, voltage, speed, temperature, and other relevant information during the welding process (such as welding materials, welding processes, etc.). These data will be integrated into a data set for subsequent analysis and evaluation.

[0079] Specifically, assume that in a welding workshop of an automobile manufacturing plant, a quality inspector uses ultrasonic testing equipment to detect the welding joints of a batch of automobile frames; during the detection process, he finds a relatively large porosity defect in a weld located at the rear of the frame; the quality inspector records the specific position and shape of the porosity defect and marks it on the weld drawing; the quality inspector traces back to the recorded data during the welding process and finds that the porosity defect appears during a specific time period during the welding process; he further analyzes and finds that the welding current fluctuates abnormally during this time period; therefore, he marks the change point of the welding current during this time period as the traceability node related to the porosity defect.

[0080] The quality inspector used the recorded data during the welding process and combined with the marked traceability nodes to conduct location-based traceability. He determined the specific location and time of the porosity defect occurrence, as well as the related welding parameters (i.e., abnormal fluctuations in welding current). The quality inspector extracted the welding data related to the occurrence of the porosity defect, including parameters such as welding current, voltage, speed, and other relevant information during the welding process (such as welding materials, welding processes, etc.). He integrated these data into a data set and prepared it for subsequent analysis and improvement work. Through the above steps, the quality inspector successfully achieved location-based traceability of the welding defect area and collected the relevant welding data set, which will be used for subsequent analysis and improvement work to improve welding quality and production efficiency.

[0081] Furthermore, based on the traversal of the welding data set, multiple groups of welding data combinations were determined. The multiple groups of welding data combinations were welding data combinations of different dimensions, covering the welding current dimension, welding voltage dimension, welding angle dimension, and welding speed dimension. The welding current dimension, welding voltage dimension, welding angle dimension, and welding speed dimension were introduced to achieve multi-dimensional control.

[0082] At this time, comprehensively review and analyze the collected welding data set to ensure that no key information is missed. Use data processing software or scripts to traverse the welding data set. The traversal process includes reading, parsing, and preliminary classification of each data record to ensure the accuracy of subsequent analysis.

[0083] Combine the welding data according to different dimensions to facilitate in-depth analysis of how various factors in the welding process affect welding quality. At this time, according to the characteristics of the welding process and the factors to be analyzed, divide the welding data into different dimension combinations. These dimensions include welding current, welding voltage, welding angle, and welding speed, etc. Each dimension combination contains all the data records under that dimension, facilitating subsequent comparison and analysis.

[0084] Ensure the comprehensiveness of the analysis, covering all key factors in the welding process. At this time, in actual operation, it is necessary to ensure that each dimension is fully considered and that there is logical coherence and comparability between the data combinations. For example, the combination of welding current and welding voltage can reflect the heat input situation during the welding process, while the combination of welding angle and welding speed can reflect the geometric characteristics and process stability of the welded joint.

[0085] Specifically, assume that in a welding workshop of a steel structure manufacturing plant, a set of welding data for a batch of welding joints has been collected, including parameters such as welding current, welding voltage, welding angle, and welding speed; use data processing software to traverse the welding data set; he first reads all the data records and conducts preliminary classification and sorting; during the traversal process, he notices that there are outliers or missing values in some data records, so corresponding cleaning and correction are carried out.

[0086] According to the characteristics of the welding process and the factors to be analyzed, the following four groups of welding data combinations are determined: the combination of welding current and welding voltage; the combination of welding angle and welding speed; the combination of welding current and welding speed; the combination of welding voltage and welding angle; each group of data combinations contains all the data records in the corresponding dimension and has been sorted in chronological order or welding order for subsequent comparison and analysis.

[0087] Start analyzing these welding data combinations; for example, when analyzing the combination of welding current and welding voltage, he finds that as the welding current increases, the welding voltage also shows a corresponding increasing trend, indicating that the heat input during the welding process is stable and meets the expected process requirements; while when analyzing the combination of welding angle and welding speed, he finds that there are welding defects in some welding joints when the welding angle is large and the welding speed is fast, suggesting that the welding process parameters need to be adjusted to improve the welding quality and stability; through the above steps, multiple groups of welding data combinations are successfully determined and preliminary analysis is carried out, and these analysis results provide an important basis for subsequent welding process optimization and quality improvement.

[0088] Therefore, determine the corresponding abnormal welding parameters according to the abnormal identification of multiple groups of welding data combinations, and the abnormal welding parameters include abnormal welding current, abnormal welding voltage, abnormal welding angle, or abnormal welding speed.

[0089] At this time, in multiple groups of welding data combinations, identify abnormal data that does not conform to the normal welding process, and these data indicate problems or defects in the welding process; at this time, use methods such as statistical analysis, machine learning algorithms, or expert systems to perform abnormal detection on multiple groups of welding data combinations, and these methods can identify abnormal patterns or values that deviate from the normal range in the data, thus indicating potential welding problems.

[0090] According to the results of abnormal identification, determine the specific parameters that cause welding abnormalities, and these parameters are welding current, welding voltage, welding angle, or welding speed, etc.; at this time, by comparing the differences between abnormal data and normal data, combined with the knowledge and experience of the welding process, determine the specific parameters that cause welding abnormalities, which requires in-depth analysis and interpretation of the welding data, as well as in-depth understanding of the welding process.

[0091] Classify and record the identified abnormal welding parameters for subsequent analysis and improvement. At this time, according to the nature and impact degree of the abnormal welding parameters, they are divided into different categories (such as minor abnormalities, serious abnormalities, etc.) and recorded, which helps to optimize and improve the welding process targeted subsequently.

[0092] Specifically, assume that in a welding workshop of a shipyard, multiple groups of welding data combinations have been determined according to the steps of S142 and are ready for anomaly identification; use statistical analysis methods to detect anomalies in multiple groups of welding data combinations. He noticed that in a data combination of welding current and welding voltage in a certain group, several data points deviated significantly from the normal range, showing that the welding current was abnormally high and the welding voltage was abnormally low. At the same time, similar abnormal data points were also found in a data combination of welding angle and welding speed in another group.

[0093] By comparing the differences between the abnormal data and the normal data, combined with the knowledge and experience of welding technology, the specific parameters causing welding anomalies were determined. He found that the situation where the welding current was abnormally high and the welding voltage was abnormally low was due to unstable power supply of the welding equipment or improper setting of welding parameters. The anomalies in the welding angle and welding speed data were due to insufficient technical level of the operator or mismatch of welding process parameters.

[0094] The identified abnormal welding parameters were classified and recorded. He classified the anomalies of welding current and welding voltage as equipment or parameter setting problems, and the anomalies of welding angle and welding speed as operation technology problems. At the same time, he also recorded the specific values and occurrence time points of these abnormal data points for subsequent targeted optimization and improvement of the welding process. Through the above steps, abnormal data in multiple groups of welding data combinations were successfully identified, and the corresponding abnormal welding parameters were determined. These results provide an important basis for subsequent improvement of the welding process and quality improvement.

[0095] In an embodiment of the present application, assume that in a welding workshop of a heavy machinery factory, multiple groups of welding data have been collected and are ready to identify abnormal welding parameters based on these data. First, a matching table is formulated to match the welding data with the preset normal range to identify abnormal data. The welding parameter matching table is shown in Table 5: Table 5 Welding Parameter Matching Table

[0096] Compare the collected welding data with the welding parameter matching table, and the following abnormal data are found: the welding current in a set of data is 220A, which exceeds the normal range (150 - 200A), so it is identified as an abnormal welding current; the welding angle in another set of data is 38°, which is lower than the normal range (45° ± 5°), so it is identified as an abnormal welding angle; through the welding parameter matching table, the abnormal welding parameters are quickly and accurately identified.

[0097] Reference Figure 6 , in step S15, determine the welding abnormality level according to the abnormal welding parameters and the resistance spot welding head, and determine the welding quality of the resistance spot welding according to the mapping relationship between the welding defect level, the welding abnormality level and the welding quality; In the specific implementation process of the present invention, the specific steps are as follows: S151: Collect the resistance spot welding head, and determine the welding part of the resistance spot welding head according to the detection of the resistance spot welding head. Determine the welding abnormality level according to the welding part of the resistance spot welding head, multiple abnormal welding parameters and the welding abnormality mapping relationship; S152: Align the welding defect level and the welding abnormality level in the position dimension, and mark the welding defect level and the welding abnormality level in the same welding defect area. Determine multiple sub-welding qualities based on the mapping relationship between the welding defect level, the welding abnormality level and the welding quality; S153: Determine the weight parameters of multiple sub-welding qualities based on the area of the welding defect area and the weight mapping relationship; Determine the welding quality of the resistance spot welding according to the weight parameters of multiple sub-welding qualities and the weighted processing of multiple sub-welding qualities.

[0098] In the embodiment of the present application, the resistance spot welding head is collected, and the welding part of the resistance spot welding head is determined according to the detection of the resistance spot welding head. The welding abnormality level is determined according to the welding part of the resistance spot welding head, multiple abnormal welding parameters and the welding abnormality mapping relationship, which takes into account the welding part of the resistance spot welding head, multiple abnormal welding parameters and the welding abnormality mapping relationship as a whole, and ensures the accuracy of the welding abnormality level.

[0099] At this time, the resistance spot welding head is collected. After the detailed information of the resistance spot welding head is collected, the next step is to determine the welding part of the welding head, which involves analyzing the structure of the welding head and identifying the area directly related to the welding process; the welding part is the area where the welding head contacts the workpiece and transmits the welding current and pressure; the process of determining the welding part needs to refer to the design drawings of the welding head, the manufacturer's instructions or industry standards.

[0100] Determine the welding anomaly level based on the welding part of the resistance spot welding head, multiple abnormal welding parameters, and the welding anomaly mapping relationship. At this time, combine the collected information of the resistance spot welding head with the multiple abnormal welding parameters (such as abnormal welding current, voltage, angle, speed, etc.) determined in the previous steps; then, use the predefined welding anomaly mapping relationship to evaluate the impact degree of the combination of this information on the welding quality; the welding anomaly mapping relationship is a lookup table or algorithm constructed based on historical data, expert experience, or machine learning models, which can map a specific welding head state and welding parameter combination to the corresponding welding anomaly level; the welding anomaly level is a classification label used to indicate the severity of potential problems or defects existing in the welding process; the levels include categories such as "normal", "slight anomaly", "moderate anomaly", and "severe anomaly", depending on the definition of the mapping relationship.

[0101] Specifically, assume that on the resistance spot welding production line of an automobile manufacturing plant, a dedicated welding head detection device is used to collect detailed information of a resistance spot welding head; through analysis, the welding part of the welding head is determined, and it is noted that there are slight wear marks on the surface of this part; next, review the abnormal welding parameters determined in the previous steps and find that the welding current used by this welding head in the recent several welding processes is slightly higher than the normal range; the welding equipment on this production line has recently undergone a maintenance, but the maintenance record shows that the current control problem has not been completely solved.

[0102] With this information, use the predefined welding anomaly mapping relationship to evaluate the impact of the combination of the welding head state and welding parameters on the welding quality; according to the mapping relationship, slight wear of the welding head and a welding current slightly higher than the normal range are determined to be "moderate anomaly", which means that there are potential quality problems in the welding process, such as a decrease in the strength of the welded joint or an expansion of the heat affected zone in the welding area; based on this evaluation result, it can be decided to take further actions, such as adjusting the welding parameters, replacing the welding head, or conducting additional quality inspections, to ensure that the welded parts produced meet the quality standards.

[0103] Furthermore, align the welding defect level and the welding anomaly level in the position dimension, and mark the welding defect level and the welding anomaly level in the same welding defect area. Determine multiple sub-welding qualities based on the welding defect level, the welding anomaly level, and the welding quality mapping relationship, which accommodates the overall consideration of the welding defect level, the welding anomaly level, and the welding quality mapping relationship, and ensures the accuracy of multiple sub-welding qualities.

[0104] At this time, align the welding defect level (determined by non-destructive testing or other quality inspection methods) and the welding anomaly level (determined as in S151) in the position dimension. This means determining whether the welding defects and welding anomalies occur in the same welding area or adjacent areas, and their relative positional relationship; the alignment process requires the use of precise positioning technologies, such as coordinate measurement systems, image processing software, or laser scanning technologies, to ensure that the positions of the defects and anomalies can be accurately matched; in addition, the geometric shape and size of the welded parts need to be considered to ensure the accuracy of the alignment.

[0105] Mark the welding defect level and the welding anomaly level in the same welding defect area. At this time, once the welding defect level and the welding anomaly level are aligned in the position dimension, the next step is to mark these two levels in the same welding defect area, which involves adding annotations or marks in the drawing, model, or inspection report of the welded part to indicate the position, size, and level of the defects and anomalies; the marking process requires the use of standardized symbols, colors, or codes to represent different levels for subsequent analysis and quality control; in addition, the accuracy and readability of the marks need to be ensured to avoid confusion or misunderstanding.

[0106] Determine multiple sub-welding qualities based on the welding defect level, the welding anomaly level, and the welding quality mapping relationship. At this time, a predefined welding quality mapping relationship will be used to evaluate the impact of the welding defect level and the welding anomaly level on the welding quality; the welding quality mapping relationship is a lookup table or algorithm constructed based on experimental data, industry standards, or expert judgment, which can map specific defect and anomaly levels to corresponding sub-welding quality indicators; the sub-welding quality indicators include multiple aspects such as the strength, toughness, sealing performance, and corrosion resistance of the welded joint. These indicators are used to evaluate the overall quality and reliability of the welded part and are affected by welding defects and anomalies; based on the welding quality mapping relationship, the values or levels of multiple sub-welding quality indicators can be determined for each welding defect area, and these values or levels will be used for subsequent quality control and improvement decisions.

[0107] Specifically, the ultrasonic testing method was used to perform non-destructive testing on the welded joints of a section of steel beam. The test results showed that there was a secondary welding defect (such as porosity or slag inclusion) in a certain area of the welded joint. At the same time, the welding anomaly level determined in the previous steps was reviewed, and it was found that there were moderate abnormal welding current fluctuations during the welding process in this area. To evaluate the impact of these two problems on the welding quality, a predefined welding quality mapping relationship was used. According to the mapping relationship, the secondary welding defect and the moderate abnormal welding current fluctuations were mapped to two sub-welding quality indicators, namely the strength and toughness of the welded joint. Based on the mapping results, the strength of the welded joint in this area was determined to be "qualified but on the low side", and the toughness was "qualified". This means that although the welded joint meets the basic requirements in terms of strength and toughness, there are potential quality risks, and additional measures need to be taken to improve the welding quality, such as local repair or enhanced monitoring. Through this step, the welding quality status can be understood more accurately, and strong support is provided for subsequent quality control and improvement decisions.

[0108] Therefore, based on the area of the welding defect area and the weight mapping relationship, the weight parameters of multiple sub-welding qualities are determined. According to the weight parameters of multiple sub-welding qualities and the weighted processing of multiple sub-welding qualities for the welding quality of resistance spot welding, the overall control of the welding defect level and the welding anomaly level is introduced, and the welding quality is detected based on dimensions such as the welding defect level and the welding anomaly level, improving the accuracy of the welding quality of resistance spot welding.

[0109] At this time, to determine the weight parameters of multiple sub-welding qualities based on the area of the welding defect area and the weight mapping relationship, first, the area of the welding defect area needs to be measured or estimated, which can be achieved by using measuring tools (such as calipers, microscopes, or image analysis software) to ensure the accuracy of the area. The size of the defect area reflects the severity of the defect and the magnitude of the potential impact on the welding quality. Next, the predefined weight mapping relationship is used to determine the weight parameters of each sub-welding quality indicator. The weight mapping relationship is a look-up table or algorithm constructed based on historical data, industry standards, or expert judgment, which can map a specific defect area to the weights of the corresponding sub-welding quality indicators. The weight parameter represents the importance or contribution degree of each sub-welding quality indicator in the overall welding quality assessment. A larger weight means that this indicator has a greater impact on the welding quality, so more attention should be given during the assessment process.

[0110] The welding quality of resistance spot welding is determined based on the weight parameters of multiple sub-welding qualities and the weighted processing of multiple sub-welding qualities. The weight parameters determined in the previous step will be used to weight each sub-welding quality index; the weighted processing involves multiplying the value (or grade) of each sub-welding quality index by its corresponding weight parameter to obtain the weighted value; the weighted value reflects the relative contribution of each sub-welding quality index to the overall welding quality; then, these weighted values can be added together or subjected to other forms of comprehensive processing to obtain the overall welding quality score or grade of the resistance spot welding; the overall welding quality score or grade is used to evaluate whether the welded part meets specific quality requirements or standards; if the score is lower than the preset qualified threshold, further measures need to be taken to improve the welding quality, such as repair, re-welding, or strengthening quality control.

[0111] Specifically, the quality assessment of the resistance spot welding of an aircraft structural part was carried out; in step S152, two main sub-welding quality indexes were determined: the strength and toughness of the welded joint, and their values and grades in a certain defect area were given respectively; now, in step S153, first, the area of the defect area was measured and found to be relatively small, but still needed attention; then, the predefined weight mapping relationship was used to determine the weight parameters of the strength and toughness indexes; according to the mapping relationship, since the defect area is small and located in a non-critical part, the weights of strength and toughness were set to 0.6 and 0.4 respectively.

[0112] Then, the values of strength and toughness (assumed to be 85 and 90 respectively, expressed as percentages) were multiplied by their corresponding weight parameters to obtain the weighted values: the strength was 51 and the toughness was 36; finally, the weighted strength and toughness values were added together to obtain an overall welding quality score of 87; according to the quality requirements of the project, the overall welding quality score needs to reach 90 or above to be considered qualified; therefore, it was determined that the welding quality of this resistance spot welding was unqualified, and a decision was made to take further measures to improve the welding quality.

[0113] Please refer to Figure 7 , Figure 7 which is a schematic diagram of the structural composition of a resistance spot welding quality detection system based on ultrasonic monitoring in an embodiment of the present invention; the resistance spot welding quality detection system based on ultrasonic monitoring includes: An ultrasonic parameter module 21, configured to determine a plurality of ultrasonic parameters based on ultrasonic monitoring of a welded joint by an ultrasonic probe, and the plurality of ultrasonic parameters include the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves; A welding defect area module 22, configured to locate the weld seam of the welded joint, determine the weld seam characteristics based on the dynamic image of the weld seam of the welded joint, and determine the welding defect area of the welded joint according to the weld seam characteristics and the plurality of ultrasonic parameters; The welding defect level module 23 is used to determine a plurality of welding defect features based on the welding defect area, and determine the welding defect level based on the plurality of welding defect features, the corresponding spatial positions, and the corresponding defect level mapping relationship; The abnormal welding parameter module 24 is used to collect the corresponding welding data set according to the traceability of the welding defect area, and determine the abnormal welding parameters according to the detection of the welding data set; The welding quality module 25 is used to determine the welding abnormality level according to the abnormal welding parameters and the resistance spot welding head, and determine the welding quality of the resistance spot welding according to the welding defect level, the welding abnormality level, and the welding quality mapping relationship.

[0114] For any combination of the technical features of the above embodiments, for the sake of brevity of description, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

Claims

1. A method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring, characterized in that, Including: Determining a plurality of ultrasonic parameters based on ultrasonic monitoring of a welded joint by an ultrasonic probe, the plurality of ultrasonic parameters including the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves; Locating the weld seam of the welded joint, determining weld seam features based on the dynamic image of the weld seam of the welded joint, and determining the welding defect area of the welded joint according to the weld seam features and the plurality of ultrasonic parameters; Determining a plurality of welding defect features based on the welding defect area, and determining the welding defect grade based on the mapping relationship between the plurality of welding defect features, corresponding spatial positions, and corresponding defect grades; Collecting a corresponding welding data set according to the traceability of the welding defect area, and determining abnormal welding parameters according to the detection of the welding data set; Determining the welding abnormality grade according to the abnormal welding parameters and the resistance spot welding head, and determining the welding quality of the resistance spot welding according to the mapping relationship between the welding defect grade, welding abnormality grade, and welding quality.

2. The method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 1, characterized in that, The determining of the plurality of ultrasonic parameters based on ultrasonic monitoring of the welded joint by the ultrasonic probe, the plurality of ultrasonic parameters including the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves, includes: Performing resistance spot welding on the welded joint and welding the welded joint to the workpiece to be welded. Meanwhile, the ultrasonic monitor is arranged obliquely relative to the welded joint and performs ultrasonic monitoring on the welded joint; When the ultrasonic probe performs ultrasonic monitoring on the welded joint, collecting a plurality of ultrasonic signals, and determining an ultrasonic signal set according to the plurality of ultrasonic signals and the welding location of the welded joint and the workpiece to be welded; Inputting the ultrasonic signal set into a preset ultrasonic parameter learning model, the ultrasonic parameter learning model classifies the ultrasonic signal set and forms propagation combinations, attenuation combinations, and reflection combinations of ultrasonic waves. The propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves are generated according to further identification of the propagation combinations, attenuation combinations, and reflection combinations of ultrasonic waves.

3. A method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 1, characterized in that, The locating of the weld seam of the welded joint, determining weld seam features based on the dynamic image of the weld seam of the welded joint, and determining the welding defect area of the welded joint according to the weld seam features and the plurality of ultrasonic parameters, includes: Dynamically monitoring the welding of the welded joint, gradually forming the weld seam of the welded joint with the resistance spot welding of the welded joint, collecting the dynamic image of the weld seam of the welded joint, and generating a plurality of sub-weld seam areas according to the dynamic image of the weld seam of the welded joint and the division of the surface color difference of the welded joint; Synchronously identifying the plurality of sub-weld seam areas, and determining weld seam features according to the positions and corresponding area shapes of the plurality of sub-weld seam areas; Collecting the material of the welded joint and the distribution map of the welded joint, determining the first defect area according to the weld seam features and the material of the welded joint, determining the second defect area according to the weld seam features and the plurality of ultrasonic parameters, and determining the welding defect area of the welded joint based on the first defect area, the second defect area, and the distribution map of the welded joint.

4. A method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 1, characterized in that, The determining of the plurality of welding defect features based on the welding defect area, and determining the welding defect grade based on the mapping relationship between the plurality of welding defect features, corresponding spatial positions, and corresponding defect grades, includes: Mark the location of the welded defect area, and determine multiple welded defect features according to the location of the welded defect area and the morphology of the defect area. The multiple welded defect features include pores, cracks, weld beads or craters.

5. A method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 4, wherein, Based on the welded defect area, determine multiple welded defect features. Based on the mapping relationship between the multiple welded defect features, the corresponding spatial positions and the corresponding defect grades, determine the welded defect grades. It further includes: Based on the multiple welded defect features and the distribution map of the welded joint, determine the spatial positions of the multiple welded defect features. Generate the corresponding welded transition area according to two adjacent welded defect features, and mark the area position of the welded transition area. Match the multiple welded defect features and the area positions of the welded transition areas with the corresponding defect grade mapping relationship, and generate the corresponding welded defect grades. The welded defect grades are used to present the quality grades of the resistance spot welding in terms of welded defect dimensions.

6. The method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 1, characterized in that, Collect the corresponding welding data set according to the traceability of the welded defect area, and determine the abnormal welding parameters according to the detection of the welding data set. It includes: Collect the welded defect area, mark the corresponding traceability nodes for the welded defect area, realize the location-based traceability of the welded defect area according to the trigger of the traceability nodes, and collect the corresponding welding data set based on the location-based traceability of the welded defect area.

7. A method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 6, characterized in that Collect the corresponding welding data set according to the traceability of the welded defect area, and determine the abnormal welding parameters according to the detection of the welding data set. It further includes: Based on the traversal of the welding data set, determine multiple groups of welding data combinations. The multiple groups of welding data combinations are welding data combinations in different dimensions, covering the welding current dimension, the welding voltage dimension, the welding angle dimension and the welding speed dimension. Determine the corresponding abnormal welding parameters according to the abnormal identification of the multiple groups of welding data combinations. The abnormal welding parameters include abnormal welding current, abnormal welding voltage, abnormal welding angle or abnormal welding speed.

8. A method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 1, characterized in that Determine the welding abnormality grade according to the abnormal welding parameters and the resistance spot welding head, and determine the welding quality of the resistance spot welding according to the mapping relationship between the welded defect grade, the welding abnormality grade and the welding quality. It includes: Collect the resistance spot welding head, and determine the welding part of the resistance spot welding head according to the detection of the resistance spot welding head. Determine the welding abnormality grade according to the welding part of the resistance spot welding head, the multiple abnormal welding parameters and the welding abnormality mapping relationship.

9. A method for detecting the welding quality of resistance spot welding based on ultrasonic monitoring according to claim 8, characterized in that, Determine the welding abnormality grade according to the abnormal welding parameters and the resistance spot welding head, and determine the welding quality of the resistance spot welding according to the mapping relationship between the welded defect grade, the welding abnormality grade and the welding quality. It further includes: Align the welded defect grade and the welding abnormality grade in the position dimension, and mark the welded defect grade and the welding abnormality grade in the same welded defect area. Determine multiple sub-welding qualities based on the mapping relationship between the welded defect grade, the welding abnormality grade and the welding quality. Determine the weight parameters of the multiple sub-welding qualities based on the area of the welded defect area and the weight mapping relationship; determine the welding quality of the resistance spot welding according to the weight parameters of the multiple sub-welding qualities and the weighted processing of the multiple sub-welding qualities.

10. A resistance spot welding quality detection system based on ultrasonic monitoring, characterized in that, The resistance spot welding quality detection system based on ultrasonic monitoring is applied to a resistance spot welding quality detection method based on ultrasonic monitoring as described in any one of claims 1-9. The resistance spot welding quality detection system based on ultrasonic monitoring includes: An ultrasonic parameter module for determining a plurality of ultrasonic parameters based on ultrasonic monitoring of a welded joint by an ultrasonic probe. The plurality of ultrasonic parameters include the propagation speed, attenuation coefficient, and reflection coefficient of ultrasonic waves; A welding defect area module for locating the weld seam of a welded joint, determining weld seam features based on a dynamic image of the weld seam of the welded joint, and determining the welding defect area of the welded joint according to the weld seam features and the plurality of ultrasonic parameters; A welding defect grade module for determining a plurality of welding defect features based on the welding defect area, and determining the welding defect grade based on the mapping relationship between the plurality of welding defect features, corresponding spatial positions, and corresponding defect grades; An abnormal welding parameter module for collecting a corresponding welding data set according to the traceability of the welding defect area, and determining abnormal welding parameters according to the detection of the welding data set; A welding quality module for determining the welding abnormality grade according to the abnormal welding parameters and the resistance spot welding head, and determining the welding quality of the resistance spot welding according to the mapping relationship between the welding defect grade, welding abnormality grade, and welding quality.

Images