Search Results (340)

Accurate weld seam recognition is essential in automated welding systems, as it directly affects path planning and welding quality. With the rapid advancement of industrial vision, weld seam instance segmentation has emerged as a prominent research focus in both academia and industry. However, existing approaches still face significant challenges in boundary perception and structural representation. Due to the inherently elongated shapes, complex geometries, and blurred edges of weld seams, current segmentation models often struggle to maintain high accuracy in practical applications. To address this issue, a novel structure-aware and boundary-enhanced YOLO (SABE-YOLO) is proposed for weld seam instance segmentation. First, a Structure-Aware Fusion Module (SAFM) is designed to enhance structural feature representation through strip pooling attention and element-wise multiplicative fusion, targeting the difficulty in extracting elongated and complex features. Second, a C2f-based Boundary-Enhanced Aggregation Module (C2f-BEAM) is constructed to improve edge feature sensitivity by integrating multi-scale boundary detail extraction, feature aggregation, and attention mechanisms. Finally, the inner minimum point distance-based intersection over union (Inner-MPDIoU) is introduced to improve localization accuracy for weld seam regions. Experimental results on the self-built weld seam image dataset show that SABE-YOLO outperforms YOLOv8n-Seg by 3 percentage points in the AP(50–95) metric, reaching 46.3%. Meanwhile, it maintains a low computational cost (18.3 GFLOPs) and a small number of parameters (6.6M), while achieving an inference speed of 127 FPS, demonstrating a favorable trade-off between segmentation accuracy and computational efficiency. The proposed method provides an effective solution for high-precision visual perception of complex weld seam structures and demonstrates strong potential for industrial application. Full article

Real-time quality monitoring using molten pool images is a critical focus in researching high-quality, intelligent automated welding. To address interference problems in molten pool images under complex welding scenarios (e.g., reflected laser spots from spatter misclassified as porosity defects) and the limited interpretability of deep learning models, this paper proposes a multi-granularity spatiotemporal representation learning algorithm based on the hybrid enhancement of handcrafted and deep learning features. A MobileNetV2 backbone network integrated with a Temporal Shift Module (TSM) is designed to progressively capture the short-term dynamic features of the molten pool and integrate temporal information across both low-level and high-level features. A multi-granularity attention-based feature aggregation module is developed to select key interference-free frames using cross-frame attention, generate multi-granularity features via grouped pooling, and apply the Convolutional Block Attention Module (CBAM) at each granularity level. Finally, these multi-granularity spatiotemporal features are adaptively fused. Meanwhile, an independent branch utilizes the Histogram of Oriented Gradient (HOG) and Scale-Invariant Feature Transform (SIFT) features to extract long-term spatial structural information from historical edge images, enhancing the model’s interpretability. The proposed method achieves an accuracy of 99.187% on a self-constructed dataset. Additionally, it attains a real-time inference speed of 20.983 ms per sample on a hardware platform equipped with an Intel i9-12900H CPU and an RTX 3060 GPU, thus effectively balancing accuracy, speed, and interpretability. Full article

This study investigates the use of YOLOv8 deep learning models to segment and classify thermal images acquired from the rear of the weld pool during the Gas Tungsten Arc Welding (GTAW) process. Thermal data were acquired using a two-color pyrometer under three welding current levels (160 A, 180 A, and 200 A). Models of sizes from nano to extra-large were trained on 66 annotated frames and evaluated with and without data augmentation. The results demonstrate that the YOLOv8m model achieved the best classification performance, with a precision of 83.25% and an inference time of 21.4 ms per frame by using GPU, offering the optimal balance between accuracy and speed. Segmentation accuracy also remained high across all current levels. The YOLOv8n model was the fastest (15.9 ms/frame) but less accurate (75.33%). Classification was most reliable at 160 A, where the thermal field was more stable. The arc reflection class was consistently identified with near-perfect precision, demonstrating the model’s robustness against non-relevant thermal artifacts. These findings confirm the feasibility of using lightweight, dual-task neural networks for reliable weld pool analysis, even with limited training data. Full article

The microstructure and mechanical properties of welded joints of API 5L X-52 steel plates cladded with AISI 316L-Si austenitic stainless steel were evaluated. The gas metal arc welding process with pulsed arc (GMAW-P) and controlled arc oscillation were used to join the bimetallic plates. After the root welding pass, buttering with an ERNiCrMo-3 filler wire was performed and multi-pass welding followed using an ER70S-6 electrode. The results obtained by optical and scanning electron microscopy indicated that the shielding atmosphere, welding parameters, and electric arc oscillation enabled good arc stability and proper molten metal transfer from the filler wire to the sidewalls of the joint during welding. Vickers microhardness (HV) and tensile tests were performed for correlating microstructural and mechanical properties. The mixture of ERNiCrMo-3 and ER70S-6 filler materials presented fine interlocked grains with a honeycomb network shape of the Ni–Fe mixture with Ni-rich grain boundaries and a cellular-dendritic and equiaxed solidification. Variation of microhardness at the weld metal (WM) in the middle zone of the bimetallic welded joints (BWJ) is associated with the manipulation of the welding parameters, promoting precipitation of carbides in the austenitic matrix and formation of martensite during solidification of the weld pool and cooling of the WM. The BWJ exhibited a mechanical strength of 380 and 520 MPa for the yield stress and ultimate tensile strength, respectively. These values are close to those of the as-received API 5L X-52 steel. Full article

Process control and quality assurance of wire + arc additive manufacturing (WAAM) and automated welding rely heavily on in-process monitoring videos to quantify variables such as melt pool geometry, location and size of droplet transfer, arc characteristics, etc. To enable feedback control based upon this information, an automatic and robust segmentation method for monitoring of videos and images is required. However, video segmentation in WAAM and welding is challenging due to constantly fluctuating arc brightness, which varies with deposition and welding configurations. Additionally, conventional computer vision algorithms based on greyscale value and gradient lack flexibility and robustness in this scenario. Deep learning offers a promising approach to WAAM video segmentation; however, the prohibitive time and cost associated with creating a well-labelled, suitably sized dataset have hindered its widespread adoption. The emergence of large computer vision models, however, has provided new solutions. In this study a semi-automatic annotation tool for WAAM videos was developed based upon the computer vision foundation model SAM and the video object tracking model XMem. The tool can enable annotation of the video frames hundreds of times faster than traditional manual annotation methods, thus making it possible to achieve rapid quantitative analysis of WAAM and welding videos with minimal user intervention. To demonstrate the effectiveness of the tool, three cases are demonstrated: online wire position closed-loop control, droplet transfer behaviour analysis, and assembling a dataset for dedicated deep learning segmentation models. This work provides a broader perspective on how to exploit large models in WAAM and weld deposits. Full article

Traditional narrow gap welding of thick titanium alloy plates easily produces dynamic molten pool flow instability, poor sidewall fusion, and excessive residual stress after welding, which leads to defects such as pores, cracks, and large welding deformations. In view of the above problems, this study takes 16-mm-thick TC4 titanium alloy as the research object, uses low-power pulsed laser-GTA flexible heat source welding technology, and uses the flexible regulation of space between the laser, arc, and wire to promote good fusion of the molten pool and side wall metal. By implementing instant ultrasonic impact treatment on the weld surface, the residual stress of the welded specimen is controlled within a certain range to reduce deformation after welding. The results show that the new welding process makes the joint stable, the side wall is well fused, and there are no defects such as pores and cracks. The weld zone is composed of a large number of α′ martensites interlaced with each other to form a basketweave structure. The tensile fracture of the joint occurs at the base metal. The joint tensile strength is 870 MPa, and the elongation after fracture can reach 17.1%, which is 92.4% of that of the base metal. The impact toughness at the weld is 35 J/cm², reaching 81.8% of that of the base metal. After applying ultrasound, the average residual stress decreased by 96% and the peak residual stress decreased by 94.8% within 10 mm from the weld toe. The average residual stress decreased by 95% and the peak residual stress decreased by 95.5% within 10 mm from the weld root. The residual stress on the surface of the whole welded test plate could be controlled within 200 MPa. Finally, a high-performance thick Ti-alloy plate welded joint with good forming and low residual stress was obtained. Full article

This study presents numerical analyses of the thermal, metallurgical, and mechanical processes involved in welding. The temperature fields were computed by solving the transient heat transfer equation using the ABAQUS/Standard 2024 finite element solver. Two types of moving heat sources were applied: a surface Gaussian distribution and a volumetric model, both implemented via DFLUX subroutines to simulate welding on butt-jointed plates. The simulation accounted for key welding parameters, including current, voltage, welding speed, and plate dimensions. The thermophysical properties of the INC 738 LC nickel superalloy were used in the model. Solidification characteristics, such as dendritic arm spacing, were estimated based on cooling rates around the weld pool. The model also calculated transverse residual stresses and applied a hot cracking criterion to identify regions vulnerable to cracking. The peak transverse stress, recorded in the heat-affected zone (HAZ), reached 1.1 GPa under Goldak’s heat input model. Additionally, distortions in the welded plates were evaluated for both heat source configurations. Full article

This study investigates the spatter suppression mechanism in aluminum alloy welding using a hybrid fiber–semiconductor laser system. By integrating high-speed photography and three-dimensional thermal-fluid coupling numerical simulations, the spatter formation process and its suppression mechanisms were systematically analyzed. The results indicate that spatter formation is primarily governed by surface tension and recoil pressure. In single fiber laser welding, concentrated laser energy induces a steep temperature gradient on the molten pool surface, triggering a strong Marangoni effect and subsequent spatter generation. In contrast, the hybrid laser system optimizes energy distribution, reducing the temperature gradient and weakening the Marangoni effect, thereby suppressing spatter. Additionally, the hybrid laser stabilizes molten pool flow through uniform recoil pressure distribution, further inhibiting spatter formation. Experimental results demonstrate that the hybrid fiber–semiconductor laser system significantly reduces spatter, improving welding quality and stability. This study provides theoretical and technical support for optimizing aluminum alloy laser welding. Full article

During the cold metal transfer plus pulse (CMT+P) welding process of medium-thick plates, problems such as incomplete penetration (IP) and burn-through (BT) are prone to occur, and weld pool morphology is important information reflecting the penetration states. In order to acquire high-quality weld pool images under complex welding conditions, such as smoke and arc light, a welding monitoring system was designed. For the purpose of predicting weld penetration states, the improved Inception-ResNet prediction model was proposed. Squeeze-and-Excitation (SE) block was added after each Inception-ResNet block to further extract key feature information from weld pool images, increasing the weight of key features beneficial for predicting the penetration states. The model has been trained, validated, and tested. The results demonstrate that the improved model has an accuracy of over 96% in predicting penetration states of aluminum alloy medium-thick plates compared to the original model. The model was applied in welding experiments and achieved an accurate prediction. Full article

Automated and robotic welding have become standard practices in manufacturing, requiring precise control to maintain weld quality without relying on skilled welders. In Tungsten Inert Gas (TIG) welding, monitoring the weld pool is crucial for ensuring the necessary weld penetration, which is vital for maintaining weld integrity. Real-time observation is essential to prevent defects and improve weld quality. Various sensing technologies have been developed to address this need, with vision-based systems showing particular effectiveness in enhancing welding quality and productivity within the framework of Industry 4.0. This review looks at the latest technologies for monitoring weld pools and bead shapes. It covers methods like using Complementary Metal-Oxide Semiconductors (CMOS) to take clear images of the melt pool for better process identification, Active Appearance Model (AAM) to capture 3D images of the weld pool for accurate penetration measurement, and Charge-Coupled Devices (CCD) and Laser-Induced Breakdown Spectroscopy (LIBS) to analyze plasma spectra and create material composition graphs. Full article

In the present study, a thermal–electromagnetic hydrodynamics model has been used to study welding temperature and melt flow characteristics during the laser welding of 316L steel. This welding was performed using an assisted electromagnetic field. In addition, a Monte Carlo model was used to study grain growth during solidification with the purpose of achieving a better understanding of the control of the microstructure. Based on the numerical model, which has been validated by experimental data, the effects of the current intensity of the electromagnetic field on the temperature distribution, melt flow characteristics, and grain growth are discussed here in detail. The simulation results showed that both Marangoni convection and welding temperature could be controlled by the magnetic damping effect, and that they increased due to the electromagnetic heating effect when using an electromagnetic field. Furthermore, when controlling the temperature distribution and melt flow velocity in the laminar flow of the melt pool, which was assisted by a 30 A current intensity of the electromagnetic field, the temperature gradient decreased by 13.5%. This decrease could be even larger than 50% when a turbulent flow was formed in the melt pool, which has here been demonstrated for a current intensity of 100 A. In addition, undercooling was found to decrease because of the increase in the melt flow velocity when using an assistive electromagnetic field. This led to a longer nucleation time in the melt pool. Furthermore, more and larger directional columnar grains, grown by the driving force of the temperature gradient, could be formed after the consumption of the small, nucleated grains near the solid–liquid interface. In short, by controlling the temperature distribution and melt flow velocity, the required grain morphology (equiaxed or columnar) and dimension (radius, length, or width) can be controlled by coarsening and epitaxial growth. Full article

Plasma arc welding (PAW) is commonly employed for welding medium and thick plates due to its capability of single-side welding and double-side forming. Ensuring welding quality necessitates real-time precise identification of the melting state. However, the intricate interaction between the plasma arc and the molten pool, along with substantial signal noise, poses a significant technical hurdle for achieving accurate real-time melting state identification. This study introduces a magnetically controlled method for identifying plasma arc melt-through, which integrates arc voltage and arc pool pressure. The application of an alternating transverse magnetic field induces regular oscillations in the melt pool by the plasma arc. The frequency characteristics of the arc voltage and pressure signals during these oscillations exhibit distinct mapping relationships with various fusion states. A hybrid feature extraction model combining gray correlation analysis (GRA) and the Pearson correlation coefficient (PCC) is devised to disentangle the nonlinear, non-smooth, and high-dimensional repetitive features of the signals. This model extracts features highly correlated with the fusion state to construct a feature vector. Subsequently, this vector serves as input for the fusion classification model, CNN-SVM, facilitating fusion state identification. The experimental results of melt-through under various welding speeds demonstrate the robustness of the proposed method for identifying melt-through through magnetic field-assisted melt pool oscillation, achieving an accuracy of 96%. This method holds promise for integration into the closed-loop quality control system of plasma arc welding, enabling real-time monitoring and control of melt pool quality. Full article

The quality of manufacturing processes largely depends on applying modern design methods and technologies. Much progress has been made in the field of metallurgy and the physics of welding, including the weld pool hydrodynamics, the surface and volumetric forces of different origins, the modeling of the SP crystallization process, and the structural transformation morphology in SWC. Additionally, attempts have been made to use the normalized parameters of fracture mechanics to evaluate the material SU. The above-mentioned solutions have also been given a more specific character by establishing SINTAP procedures and computational welding mechanics (CWM). This study discusses a universal method for welded joint evaluation according to the most significant criteria and relevant descriptive features. Full article

In additive manufacturing processes, the metal melt pool is decisive for processing quality. A single sensor is incapable of fully capturing its physical characteristics and is prone to data inaccuracies. This study proposes a multi-sensor monitoring solution integrating an off-axis infrared thermal camera with an on-axis high-speed camera to address this issue; a multi-source data pre-processing procedure has been designed, a multi-source data fusion method based on video frame interpolation has been developed, and a self-supervised training strategy based on transfer learning has been introduced. Experimental results indicate that the proposed data fusion method can eliminate temperature anomalies caused by single emissivity and droplet splashing, generating highly credible fused data and significantly enhancing the stability of metal additive manufacturing and the quality of parts. Full article

A weld pool morphology acquisition and visualization system was designed in the current study, which can present real-time three-dimensional (3D) weld pool morphologies to welders. The underneath of the weld pool is calculated by utilizing an in situ calibrated analytical solution based on real-time collected welding voltage, current, and the surface boundary of the weld pool. In the meantime, the heat source distribution coefficients of the analytical solution were also calibrated through a scaling calibration method. Thus, the system updates a 3D weld pool instantaneously in weld diameter, and the error is 0.8% at the minimum, and the average value is 8.54%. Furthermore, a virtual environment was constructed by using virtual reality (VR) devices, and the visualization of the 3D weld pool model was realized by employing the hot-update technology. The experimental results demonstrate that this system is basically feasible except the update rates still need to be optimized. The current study facilitates the easier observation of weld pool morphology and is highly significant for enhancing the teleoperation skills of welders, especially in achieving precise teleoperation welding. Full article