Search Results (1,429)

Penetration depth is a critical factor determining joint strength in butt welding; however, it is difficult to monitor in keyhole-mode laser welding due to the dynamic nature of the keyhole. Recently, optical coherence tomography (OCT) has been introduced for real-time keyhole depth measurement, though accurate results require meticulous calibration. In this study, deep learning-based models were developed to estimate penetration depth in laser welding of 780 dual-phase (DP) steel. The models utilized coaxial weld pool images and spectrometer signals as inputs, with OCT signals serving as the output reference. Both uni-sensor models (based on coaxial pool images) and multi-sensor models (incorporating spectrometer data) were developed using transfer learning techniques based on pre-trained convolutional neural network (CNN) architectures including MobileNetV2, ResNet50V2, EfficientNetB3, and Xception. The coefficients of determination values (R²) of the uni-sensor CNN transfer learning models without fine-tuning ranged from 0.502 to 0.681, and the mean absolute errors (MAEs) ranged from 0.152 mm to 0.196 mm. In the fine-tuning models, R² decreased by more than 17%, and MAE increased by more than 11% compared to the previous models without fine-tuning. In addition, in the multi-sensor model, R² ranged from 0.900 to 0.956, and MAE ranged from 0.058 mm to 0.086 mm, showing better performance than uni-sensor CNN transfer learning models. This study demonstrated the potential of using CNN transfer learning models for predicting penetration depth in laser welding of 780DP steel. Full article

The growing use of weathering steel in bridge engineering has highlighted the increasing impact of fatigue damage caused by the combined effects of welding residual stress and vehicular loading. This study investigates the fatigue performance of Q500qENH weathering steel bridge decks by proposing a coupled analysis method for residual stress and fatigue crack growth, utilizing collaborative simulations with Abaqus 2023 and Franc3D 7.0. An interaction model integrating welding-induced residual stress fields and dynamic vehicular loads is developed to systematically examine crack propagation patterns in critical regions, including the weld toes of the top plate and the weld seams of the U-ribs. The results indicate that the crack propagation rate at the top plate weld toe exhibits the most rapid progression, reaching the critical dimension (two-thirds of plate thickness) at 6.98 million cycles, establishing this location as the most vulnerable failure point. Residual stresses significantly amplify the stress amplitude under tension–compression cyclic loading, with life degradation effects showing 48.9% greater severity compared to pure tensile stress conditions. Furthermore, parametric analysis demonstrates that increasing the top plate thickness to 16 mm effectively retards crack propagation, while wheel load pressures exceeding 1.0 MPa induce nonlinear acceleration of life deterioration. Based on these findings, engineering countermeasures including welding defect control, optimized top plate thickness (≥16 mm), and wheel load pressure limitation (≤1.0 MPa) are proposed, providing theoretical support for fatigue-resistant design and maintenance of weathering steel bridge decks. Full article

A longitudinal magnetic field provides a new method for regulating the plasma velocity, pressure field, and temperature field of the TIG welding arc. However, the mechanism of action of the longitudinal magnetic field remains poorly understood. In order to address this problem, this paper develops a numerical model based on continuum mechanics. The mechanism of how magnetic field strength affects temperature, pressure field, plasma velocity, and potential was investigated. The geometric shape, temperature, pressure, and plasma velocity of the TIG welding arc under different magnetic fields were predicted. The results indicate that as magnetic field strength increases, the arc shape is compressed under the influence of magnetic forces, with the degree of compression increasing with magnetic field strength; plasma velocity gradually increases from 74 m/s at 0 mT to 296 m/s at 150 mT, but the velocity along the arc’s central axis first decreases and then increases with increasing magnetic field strength. As the magnetic field strength increases, a negative pressure first appears near the cathode, then expands toward the cathode, and finally toward the anode. During the expansion of the negative pressure, the maximum absolute value of the arc pressure increases by 12.72 times. Full article

Welding, as a critical process for achieving permanent material joining through localized heating or pressure, is extensively applied in mechanical manufacturing and transportation industries, significantly enhancing the assembly efficiency of complex structures. However, the associated localized high temperatures and rapid cooling often induce uneven thermal expansion and contraction, leading to complex stress evolution and residual stress distributions that compromise dimensional accuracy and structural integrity. In this study, we propose a combined heat source model based on the geometric characteristics of the weld pool to simulate the arc welding process of large flange shafts made of Fe-C-Mn-Cr low-alloy medium carbon steel. Simulations were performed under different welding durations and shaft diameters, and the model was validated through experimental welding tests. The results demonstrate that the proposed model accurately predicts weld pool geometry (depth error of only 2.2%) and temperature field evolution. Meanwhile, experimental and simulated deformations are presented with 95% confidence intervals (95% CI), showing good agreement. Residual stresses were primarily concentrated in the weld and heat-affected zones, exhibiting a typical “increase–steady peak–decrease” distribution along the welding direction. A welding duration of 90 s effectively reduced residual stress differentials perpendicular to the welding direction by 19%, making it more suitable for medium carbon steel components of this scale. The close agreement between simulation and experimental data verifies the model’s reliability and indicates its potential applicability to the welding simulation of other large-scale critical components, thereby providing theoretical support for process optimization. Full article

Laser beam welding (LBW) involves complex and rapid interactions between the laser and material, often resulting in defects such as pore formation. Emissions collected during the process offer valuable insight but are difficult to interpret directly for defect detection. In this study, we propose a data-driven framework to interpret electromagnetic emissions in LBW using both supervised and unsupervised learning. Our framework is implemented in the post-process monitoring stage and can be used as a real-time framework. The supervised approach uses labeled data corresponding to predefined defects (in this work, pore formation is an example of a defined defect). Meanwhile, the unsupervised method is used to identify anomalies without using predefined labels. Supervised and unsupervised learning aims to find reference values in the emissions data to determine the values of signals that lead to defects in welding (enabling quantitative monitoring). A total of 81 welding experiments were conducted, recording real-time emission data across 42 spectral channels. From these signals, statistical, temporal, and shape-based features were extracted, and dimensionality was reduced using Principal Component Analysis (PCA). The LSTM model achieved an average mean squared error (MSE) of 0.0029 and mean absolute error (MAE) of 0.0288 on the testing set across five folds. The Isolation Forest achieved 80% accuracy and 85.7% precision in detecting anomalous welds on a subset with validated defect labels. The proposed framework enhances the interpretability of 4D photonic data and enables both post-process analysis and potential real-time monitoring. It provides a scalable, data-driven approach to weld quality assessment for industrial applications. Full article

Ship repair is hazardous, often presenting unsuitable working areas and risks due to the ship’s configuration. Welding tasks are particularly dangerous due to the high temperatures generated, high enough to melt the metal in structural elements, bulkheads, linings, and tanks. This study investigates the consequences of temperature distribution during the welding of naval plates and proposes some accident prevention measures. Industry working conditions were reproduced, including the materials, procedures, and tools used, as well as the certified personnel employed. DH 36-grade naval steel, with a composition of C max. 0.18%, Mn 0.90–1.60%, P 0.035%, S 0.04%, Si 0.10–0.50%, Ni max 0.4%, Cr max 0.25%, Mo 0.08%, Cu max 0.35%, Cb (Nb) 0.05%, and V 0.1%, was welded via FCAW-G (Gas-Shielded Flux-Cored Arc Welding), selected for this study because it is one of the most widely practiced in the naval industry. The main sensor used in the experiments was an FLIR model E50 thermographic camera, and thermal waxes were employed. The results for each thickness case are presented in both graphical and tabular form to provide accurate and actionable guidelines, prioritizing safety. After studying the butt jointing of naval plates of various thicknesses (8, 10, and 15 mm), safe distances to maintain were proposed to avoid risks in the most unfavorable cases: 350 mm from the welding seam to avoid burn injuries to unprotected areas of the body and 250 mm from the welding seam to avoid producing flammable gases. These numbers are less accurate but easier to remember, which prevents errors in the face of hazards throughout a long working day. Full article

Ultrafast laser welding of glass/metal heterostructures has found extensive applications in sensors, medical devices, and optical systems. However, achieving high-stability, high-quality welds under non-optical contact conditions remains challenging due to severe internal damage within glass materials. This study addresses thermal management through synergistic control of thermal accumulation effects and material ablation thresholds. Using the sapphire/Invar alloy system as a model for glass/metal welding, we investigated thermal accumulation effects during ultrafast laser ablation of Invar alloy through theoretical simulations. Under a repetition rate of 1 MHz, the femtosecond laser raised the lattice equilibrium temperature by 700 K within 10 microseconds, demonstrating that high repetition rate femtosecond lasers can induce effective heat accumulation in Invar alloy. Furthermore, ablation thresholds for both materials were determined across varying repetition rates via the D² method, with corresponding threshold curves systematically constructed. Finally, based on the simulation and ablation threshold calculation results, laser parameters were selected for ultrafast laser single point welding of sapphire and Invar alloy. The experimental results demonstrate effective thermal effect mitigation, achieving a maximum shear strength of 63.37 MPa. Comparative analysis against traditional scan welding further validates the superiority of our approach in thermal management. This work provides foundational theoretical and methodological guidance for ultrafast laser welding of glass/metal heterostructures. Full article

While modern power sources have improved process stability, real-time monitoring and feedback control remain essential for ensuring consistent weld quality under dynamic conditions. To address this need, a vision-based closed-loop control system was developed for pulsed Metal-Active Gas (MAG) welding. The system dynamically adjusts the welding speed based on real-time visual feedback in the welding process. Otsu thresholding combined with morphological operations was applied to molten pool images for brightness-based feature extraction. These features, representing the dynamic behavior of the molten pool, were incorporated into a feedback loop for real-time control. Without relying on complex model-based prediction or sensor fusion, the proposed method reduces fluctuations in weld bead geometry and lowers the occurrence of defects. The experimental results showed that, under optimized control conditions and after a steady welding state was achieved, the weld bead’s height deviation exhibited an average standard deviation of 0.08 mm, and a process stability rate of 92%. The combination of conventional hardware and straightforward image processing makes the proposed approach practical for industrial implementation. Full article

Multi-layer multi-pass welding plays an important role in manufacturing industries such as nuclear power plants, pressure vessel manufacturing, and ship building. However, distortion or welding defects are still challenges; therefore, welding monitoring and quality control are essential tasks for the dynamic adjustment of execution during welding. The aim was to propose a machine vision system for monitoring and surface quality evaluation during multi-pass welding using a line scanner and infrared camera sensors. The cross-section modelling based on the line scanner data enabled the measurement of distortion and dynamic control of the welding plan. Lack of fusion, porosity, and burn-through defects were intentionally generated by controlling welding parameters to construct a defect inspection dataset. To reduce the influence of material surface colour, the proposed normal map approach combined with a deep learning approach was applied for inspecting the surface defects on each layer, achieving a mean average precision of 0.88. In addition to monitoring the temperature of the weld pool, a burn-through defect detection algorithm was introduced to track welding status. The whole system was integrated into a graphical user interface to visualize the welding progress. This work provides a solid foundation for monitoring and potential for the further development of the automatic adaptive welding system in multi-layer multi-pass welding. Full article

In welding applications, line-structured-light vision is widely used for seam tracking, but intense noise from arc glow, spatter, smoke, and reflections makes reliable laser-stripe segmentation difficult. To address these challenges, we propose EUFNet, an uncertainty-driven symmetrical two-stage segmentation network for precise stripe extraction under real-world welding conditions. In the first stage, a lightweight backbone generates a coarse stripe mask and a pixel-wise uncertainty map; in the second stage, a functionally mirrored refinement network uses this uncertainty map to symmetrically guide fine-tuning of the same image regions, thereby preserving stripe continuity. We further employ an uncertainty-weighted loss that treats ambiguous pixels and their corresponding evidence in a one-to-one, symmetric manner. Evaluated on a large-scale dataset of 3100 annotated welding images, EUFNet achieves a mean IoU of 89.3% and a mean accuracy of 95.9% at 236.7 FPS (compared to U-Net’s 82.5% mean IoU and 90.2% mean accuracy), significantly outperforming existing approaches in both accuracy and real-time performance. Moreover, EUFNet generalizes effectively to the public WLSD benchmark, surpassing state-of-the-art baselines in both accuracy and speed. These results confirm that a structurally and functionally symmetric, uncertainty-driven two-stage refinement strategy—combined with targeted loss design and efficient feature integration—yields high-precision, real-time performance for automated welding vision. Full article

To reduce welding deformation during the automated welding of intersection seams on thick plate curved penetrations and thereby improve welding quality and efficiency, an optimized method for segmented and multi-layer multi-pass welding sequences, along with welding process improvement strategies, is proposed. First, based on the welding model of the curved penetrations, a multi-layer multi-pass welding trajectory equation is designed. Next, a Gaussian heat source model is selected, and numerical simulation theories for welding temperature and stress fields are established using finite-element theory. Then, for the intersection seams of curved components with three different thicknesses, four numerical tests of segmented welding sequence optimization are carried out using welding finite-element simulation theory. Finally, the optimal welding process for the welding sequence is improved using orthogonal experimental methods, and the optimal welding process parameters for curved components with different thicknesses are determined. The optimization of welding sequences for intersection seams on three types of thick plates shows that the optimal sequence for segmented welding is first to perform upper–lower diagonal symmetry, followed by left–right symmetry. Compared to other welding sequences, the proposed method reduces welding deformation by an average of 9.24% and welding stress by an average of 7.40%, which verifies the effectiveness of the welding sequence optimization presented in the paper. Full article

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

The article presents a comprehensive review of key issues and challenges related to enhancing the durability of hot forging tools. It discusses modern strategies aimed at increasing tool life, including modifications to tool materials, heat treatment, surface engineering, tool and die design, die geometry, tribological conditions, and lubrication. The review is based on extensive literature data, including recent publications and the authors’ own research, which has been implemented under industrial conditions at the modern forging facility in Forge Plant “Glinik” (Poland). The study introduces original design and technological solutions, such as an innovative concept for manufacturing forging dies from alloy structural steels with welded impressions, replacing traditional hot-work tool steel dies. It also proposes a zonal hardfacing approach, which involves applying welds with different chemical compositions to specific surface zones of the die impressions, selected according to the dominant wear mechanisms in each zone. General guidelines for selecting hardfacing material compositions are also provided. Additionally, the article presents technological processes for die production and regeneration. The importance and application of computer simulations of forging processes are emphasized, particularly in predicting wear mechanisms and intensity, as well as in optimizing tool and forging geometry. Full article

In the present work, the optimization of ceramic-based composite WC(Co,Ni) welds by laser cladding was carried out using response surface methodology based on finite element analysis. The heat distribution and temperature field of laser-melted WC(Co,Ni) ceramic coatings were simulated using ANSYS software, which allowed the computation of the distribution of residual stresses. The results show that the isotherms in the simulation of the temperature field are elliptical in shape, and that the isotherms in front of the moving heat source are dense with a larger temperature gradient, while the isotherms behind the heat source are sparse with a smaller temperature gradient. In addition, the observed microstructural evolution shows that the melting zone domains of WC(Co,Ni) are mainly composed of unmelted carbides. These carbides are dendritic, rod-like, leaf-like, or net-like, and are agglomerated into smaller groups. The W content of these unmelted carbides exceeds 80%, while the C content is around 1.5–3.0%. The grey areas are composed of WC, Co and Ni compounds. Based on the regression model, a quadratic model was successfully constructed. A three-dimensional profile model of the residual stress behaviour was further explored. The estimated values of the RSM-based FEA model for residual stress are very similar to the actual results, which shows that the model is effective in reducing residual stress by laser cladding. Full article

Accurate weld defect detection is critical for ensuring structural safety and evaluating welding quality in industrial applications. Manual inspection methods have inherent limitations, including inefficiency and inadequate sensitivity to subtle defects. Existing detection models, primarily designed for natural images, struggle to adapt to the characteristic challenges of weld X-ray images, such as high noise, low contrast, and inter-defect similarity, particularly leading to missed detections and false positives for small defects. To address these challenges, a multi-dimensional feature fusion model (MADet), which is a multi-branch deep fusion network for weld defect detection, was proposed. The framework incorporates two key innovations: (1) A multi-scale feature fusion network integrated with lightweight attention residual modules to enhance the perception of fine-grained defect features by leveraging low-level texture information. (2) An anchor-based feature-selective detection head was used to improve the discrimination and localization accuracy for five typical defect categories. Extensive experiments on both public and proprietary weld defect datasets demonstrated that MADet achieved significant improvements over the state-of-the-art YOLO variants. Specifically, it surpassed the suboptimal model by 7.41% in mAP@0.5, indicating strong industrial applicability. Full article