Timely detection of early-stage fatigue microdamage remains a challenge in structural health monitoring (SHM) of safety-critical components. Here, we present a nonlinear ultrasonic (NLU) detection approach to assess early-stage microstructural degradation in cold-rolled AISI 304 stainless steels under high-cycle fatigue. An NLU system with controllable excitation and harmonic isolation was developed, aiming to correlate ultrasonic nonlinearity responses accumulation with microstructural evolution through the nonlinear metric beta'. NLU measurements revealed a bimodal trajectory of beta' during fatigue progression, with peaks at approximately 30% and 70% of fatigue life. Correlative metallographic characterisations validated the NLU response and elucidated its physical origin. Fatigue-induced changes, including dislocation activity, PSB formation, and microdamage accumulation, aligned closely with the beta' trajectory. Among them, dislocation (geometrically necessary dislocations, GNDs) dynamics played a dominant role in nonlinearity response fluctuations. A framework is proposed to major attribute early beta' evolution to lattice-elastic nonlinearity, while at later stages, crack-initiation-induced contact nonlinearity serves as an additional amplifier, boosting beta' towards its global maximum. These findings demonstrate the efficacy of harmonic-based NLU for capturing microstructural deterioration and early-damage, offering a scalable path towards physics-informed SHM in safety-critical metallic materials (as stainless steel) components.

ABSTRACT Many metallic components are frequently subjected to coupled environments of friction and vibration, stemming from interfacial contact and relative motion between mating parts or external stimuli. Consequently, it is imperative for these components to exhibit a synergy of high wear resistance and superior damping properties. In this study, a Ti 2 Ni/NiTi dual‐phase alloy with exceptional wear resistance and damping performance was in situ alloyed using the electron beam directed energy deposition (EB‐DED) process, employing pure Ti and Ni wires as raw materials. Within the three‐dimensional configuration, the NiTi martensite phase exhibits an island‐like morphology, embedded within a continuous network of the Ti 2 Ni phase, featuring an approximately equal phase ratio. This dual‐phase structure effectively harnesses the respective advantages of the Ti 2 Ni phase and the NiTi martensite phase. In friction scenarios, the hard Ti 2 Ni phase acts as a load‐bearing framework to resist the wear, whereas the NiTi martensite phase can deform to minimize wear and recover from deformation upon heating, demonstrating self‐healing capabilities. Under vibration conditions, the NiTi martensite phase dissipates energy through the movement and rearrangement of twin boundaries, resulting in effective damping. This dual‐phase structure shows promising potential for applications where friction and vibration coexist. Moreover, the phase ratio of Ti 2 Ni and NiTi can be flexibly tailored by EB‐DED technology through the regulation of wire feeding speeds, allowing for the on‐demand balance between wear resistance and damping performance.

High-energy beam oscillation can directly modulate the solidification behavior in the molten pool and has shown great potential for enhancing material performance in metal additive manufacturing. However, researches on electron beam oscillation remain limited, and no work has been reported on applying electron beam oscillation to Al-Cu alloys. In this study, Electron Beam Directed Energy Deposition (EB-DED) was employed to fabricate 2319/2A14 structures and systematically investigated the influence of oscillation amplitudes (non-oscillating, 2.0 mm, and 4.0 mm) on microstructures and mechanical properties using infinite (infinity) mode oscillation. Multiphysics numerical simulations were conducted to analyze the thermal histories, molten pool flow, and keyhole dynamics. Results demonstrate that increasing oscillation amplitude markedly reduced the porosity due to enhanced molten pool flow and keyhole stability. Notably, the porosity in the A = 4.0 mm sample decreased by approximately two orders of magnitude to 0.0018%, compared to the non-oscillating sample. Beam oscillation also promotes grain refinement and the columnar-to-equiaxed transition (CET) by agitating the molten pool and fragmenting dendrites. Furthermore, oscillation extends the molten pool length and liquid lifetime, intensifying the micro-segregation of Cu element. This leads to the formation of Cu-rich bands adjacent to alpha+theta eutectics during solidification, creating favorable sites for the subsequent precipitation of theta ' phase. The ultimate tensile strength increased by 5.3% and 14.6% for oscillation amplitude of 2.0 mm and 4.0 mm, respectively, compared to the non-oscillating condition. The improvement in strength is attributed to the synergistic effect of grain refinement and theta ' precipitation strengthening. The established relationship among "oscillation amplitude-molten pool dynamics-microstructures-mechanical properties" not only deepens the understanding of metallurgical mechanisms in additively manufactured Al-Cu alloys, but also provides a transferable framework for developing beam oscillation strategies for other alloy systems.

Electron beam freeform fabrication (EBF3) technology has attracted considerable attention due to its unique high-vacuum environment, high deposition efficiency, and exceptional space adaptability. However, the multiparameters complexity and time-varying deposition conditions usually make it difficult to ensure the stability of the molten pool width, which in turn imposes a huge burden on the geometric consistency of the thin-wall part. Therefore, this study presents a real-time vision-based monitoring and control system designed to regulate the molten pool geometry for enhanced precision. The results show that the developed image processing subsystem can quickly and accurately extract the molten pool width. The processing time for a single frame is approximately 17.65 ms, and the extraction accuracy is less than 0.2 mm. Based on the developed closed-loop control subsystem, the molten pool width can be stably and effectively controlled at the target value within +/- 0.5 mm, and the need for post-machining was largely eliminated. Moreover, variable speed and heterogeneous substrate deposition experiments demonstrate that the system developed in this study can effectively resist disturbances and exhibits good robustness. This work demonstrates a robust and cost-effective strategy for highprecision AM, representing a significant advancement towards intelligent and reliable manufacturing systems for the aerospace industry.

Electron beam directed energy deposition (EB‑DED) has significant potential for rapid near-net-shape manufacturing of large components with its high deposition rate and favorable mechanical properties. However, process stability and dimensional accuracy are hindered by complex transferred‑metal behaviors and layer‑accumulated geometric errors, making real‑time monitoring and control essential. While deep learning-based monitoring has gained traction in additive manufacturing, most existing approaches focus on static, single-frame features and singular monitoring tasks, limiting their effectiveness in fully characterizing the deposition process. To address these gaps, this study proposes a multi-task visual monitoring framework for the EB-DED process. A Temporal-Spatial Fusion Deposition Process Monitoring Network (TSF-DPMnet) is designed with a dual-branch backbone to extract features from both individual frames and image sequences. An attention-guided temporal-spatial feature fusion mechanism is designed to enhance performance in three tasks: molten pool/transferred metal segmentation, molten pool edge detection, and transfer state classification. A task incremental learning strategy, incorporating optimized training sequences and phased knowledge distillation, is further introduced to alleviate catastrophic forgetting and negative transfer, thus extending the model from single‑task to multi‑task monitoring. Based on the network outputs, post-processing algorithms for the identification or calculation of transferred‑metal transition distance, metal transfer state, and deposited layer height are designed. Experimental validation on thin-wall deposition demonstrates errors of approximately 0.1 mm in transition distance and layer height, with over 96% accuracy in metal transfer state recognition. The proposed method provides monitoring-oriented technical support for the future advancement of multi-variable and multi-loop closed-loop control in the EB-DED process, contributing to improved process stability and forming accuracy.

Non-destructive testing based on radiographic images is the key to the quality control of welded steel pipes. However, high annotation costs and the difficulty in dealing with unknown defects limit supervised methods. Existing unsupervised radiographic defect detection methods for welds mostly rely on reconstruction errors or single-feature distribution modeling, and they are still prone to false and missed detections under complex backgrounds and low-contrast subtle defect scenarios. Therefore, an unsupervised radiographic defect detection method for steel pipe welds based on intra- and inter-guided reverse distillation was proposed. Based on the reverse distillation framework, an intra- and inter-guided mechanism was introduced as a learnable prototype. Structured constraints were applied to the feature distribution from two levels of intra-sample and inter-sample to enhance the difference between local anomalies and contexts, thereby improving the discrimination ability for subtle defects under complex backgrounds. Experimental results indicate that compared with mainstream unsupervised anomaly detection methods, the proposed method achieves better results, and it provides a solution with engineering application value for the automation and intelligence of radiographic inspection for steel pipe welds.

Traditional knowledge-augmented deep learning methods show their potential in vision-based welding quality monitoring under small samples dataset. However, there remains a challenge in actual welding that existing methods by utilizing expert knowledge transfer are highly dependent on abundant modal information, especially in precision foils joining scenario with difficulties in stably obtaining high-quality welding images. Inspired by welder’s attention mechanism and context-aware hierarchy, this paper presents formal expert knowledge-guided welding, a novel framework that applies knowledge-augmented deep learning to monitor weld forming quality with microscopic molten pool images captured from high-speed vision imaging. The proposed framework consists of two modules: multi-scale visual fusion (MSVF) and interaction-knowledge semantic integration (IKSI). The molten pool region and neighboring near molten pool region are extracted from the welding image and then separately fed to MSVF module and IKSI module. Then, the discriminative features derived from MSVF module are learned through logic rule regularization term and semantic constraint regularization term obtained from IKSI module to implement the classification of five typical welding statuses during foils joining process. The ultrathin foils microwelding experimental results indicate that the proposed model achieves a classification accuracy of 93.58%. Compared with typical deep learning networks, the proposed model has an obvious superiority. Moreover, it achieves a test accuracy of 82.26% on an untrained generalization set. Available: https://github.com/Reskaine/FKEGW.

In-situ visual monitoring (ISVM) of weld pools and defects is critical for maintaining weld geometry consistency and ensuring defect-free quality in kilometer-scale laser beam welding (LBW) of large ultra-thin-walled aerospace structures. Although end-to-end deep learning (DL) offers high adaptability for ISVM, existing single-task models exhibit limited reliability under complex welding conditions due to their inability to comprehensively depict the welding state. To address this limitation, we propose a unified ISVM framework that integrates multivision-task (MVT) learning with stereo vision for simultaneous high-precision geometric perception of the weld pool, online weld reconstruction, and real-time defect identification. We proposed MT-Segformer, an efficient MVT architecture featuring a shared encoder and task-specific decoders for parallel pixel-level weld pool segmentation and defect recognition from stereo image pairs. This MVT design enhances computational efficiency and leverages segmentation-derived spatial priors to enhance defect discrimination under background interference, achieving an Intersection over Union (IoU) of 0.936 for weld pool segmentation and 0.984 accuracy for defect recognition. The segmentation output provides robust geometric constraints for stereo matching, enabling accurate 3D reconstruction, while the defect recognition branch enables real-time intervention for timely suppression of defect propagation. Experiments on liquid rocket engine nozzle welding validate the framework's capability for simultaneous real-time defect detection and accurate 3D reconstruction, with a mean reconstruction error below 0.06 mm. The proposed approach offers a systematic and effective solution for real-time quality assurance and reliability enhancement in LBW.

To address the condition monitoring problem characterized by weak detectability and poor traceability of early-stage degradation and micro-damage in weak regions of welded pressure vessels during service, a nonlinear ultrasonics (NLU) detection and evaluation strategy based on harmonic features was proposed, in which the equivalent nonlinear parameter β′ = A2/A12 of the fundamental amplitude A1 and the second harmonic amplitude A2 was adopted as a uniformly calculated and defined state trend characterization metric. An explicit finite element model incorporating mechanisms such as material elastic response nonlinearity and closed microcrack/micro-interface contact acoustic nonlinearity was first established, and the differential effects and coupling characteristics of multi-source mechanisms on harmonic generation and parameter feature responses were systematically analyzed. The investigation shows that contact acoustic nonlinearity exhibits stronger nonlinear effect characteristics; under fixed ultrasonic propagation path and excitation conditions, β′ is appropriately extended and interpreted as an equivalent indicator of multi-source nonlinear coupling intensity to clarify its physical meaning and applicability boundaries. Subsequently, an NLU detection link platform was constructed and calibrated, and repeated coupling detection validations were conducted on gradient fatigue specimens. The results show that β′ exhibits a reproducible “double-peak” evolution characteristic with the fatigue process (the ratio of fatigue life N to the fatigue life at failure Nf, i.e., N0 = N/Nf), providing higher sensitivity and discriminability to early-stage state variations of structural materials compared to linear acoustic indicators. This study provides an interpretable parametric feature baseline for the engineering application of “periodic inspection and trend interpretation” in welded pressure vessel structures and lays a foundation for subsequent data-driven/intelligent condition assessment.

Appropriate weld penetration is of vital significance for ensuring the welding quality of gas tungsten arc welding (GTAW). Visual monitoring based on deep learning has been widely applied in weld penetration monitoring. However, deep learning requires a large number of labeled samples to achieve satisfactory performance. Deep transfer learning (DTL) is an effective technique to address this issue, but the famous ImageNet dataset may not be suitable for pre-training a deep learning model for weld penetration prediction. In this study, a visual monitoring approach for weld penetration of aluminum alloy GTAW based on DTL enhanced by task-specific pre- training and semi-supervised learning (SSL) is proposed to obtain better prediction accuracy of the backside bead width with limited labeled data. Firstly, an active vision method is used to capture images of the weld pool. Next, a task-specific pre-training method is designed by constructing a keypoint localization task to pre-train a deep learning model with an encoder-decoder architecture, and SSL is introduced to reduce the required number of labeled data in pre-training. Finally, an encoder-based regression model is constructed and fine-tuned to predict the backside bead width. It is found that by using SSL in task-specific pre-training, the keypoint localization model trained with only 40 labeled samples can achieve ideal performance, and the performance of SSL outperforms fully-supervised learning (FSL) in terms of both keypoint localization accuracy and robustness to the randomness of labeled training samples. Moreover, the mean prediction error of backside bead width after finetuning is only 0.176 mm, which is reduced by 29.9 % compared to using ImageNet for pre-training. The proposed method also has good real-time performance and thus has the capability to be applied in the real-time monitoring and control of weld penetration.

Climbing helium arc welding is one of the key technologies in aerospace, military and other high-end equipment industries. Due to the time-varying position and pose of molten pool, the heat and mass transfer process during such welding are complicated and pose a challenge to ensure molten pool stability and weld penetration consistency. Recently, the weld penetration prediction based on molten pool images has shown great application potential in online weld monitoring. However, to accurately predict weld penetration is still a difficult task because of the high similarity of molten pool surface morphology under different weld penetration, especially during non-flat welding process of medium-thickness aluminum alloy plates. This paper proposes a novel vision-based two-phase framework for weld penetration prediction and applies it to climbing helium arc welding. The framework consists of multi-level characteristics extraction phase and deep time series forecasting phase. Firstly, the Deeplabv3+ is used to segment multi-region from molten pool image sequences, extracting the Molten Pool Region (MPR) and the Exposed Aluminum Liquid Region (EALR) within each image. Then, a time series comprising the extracted multi-level characteristics is constructed, and a Kansformer model is subsequently utilized to predict the weld back width based on this characteristic time series. The validity of the proposed method was verified by using data retained from an actual production platform of 2219 aluminum alloy rocket propellant canisters, and the experimental results showed that it was superior to existing methods. Moreover, its generalization ability is verified under varying process parameters, and the average inference speed of a single frame can reach 17.12 fps. In particular, SHAP analysis is utilized to explain the key role of the extracted dynamic characteristics in weld back width prediction.

To achieve online monitoring of welding penetration status, ensure weld quality, and promote the development of robotic intelligent technology, a knowledge-enhanced prediction method for Gas Tungsten Arc Welding (GTAW) penetration status based on molten pool dynamic deformation is proposed. High-speed, high-dynamic-range industrial cameras are employed to capture molten pool images, and the DeepLabv3+ semantic segmentation model is utilized for dynamic segmentation of the molten pool to obtain precise molten pool regions. On this basis, multi-frame molten pool contour image fusion is performed to describe the dynamic deformation of the molten pool during the welding process. The fused molten pool contour images and original molten pool images are combined and input into a CNN to learn pixel-level changes in the molten pool contours at the same location, enabling the CNN to predict penetration status. Experimental results demonstrate that the CNN enhanced with molten pool dynamic deformation knowledge can accurately identify three typical weld states: partial penetration, adequate penetration, and excessive penetration, achieving a classification accuracy of 97.1% with a single-frame prediction time of 0.86 ms. Compared to deep learning methods without integration of expert knowledge on molten pool dynamic deformation features, this method exhibits higher robustness and accuracy in scenarios with limited sample data.

Wire-based electron beam directed energy deposition (DED) is acclaimed for its high deposition efficiency, optimal material utilization, and the ambient conditions of vacuum deposition. Nonetheless, the inherent stresses resulting from wire circular stockpiling, coupled with the thermal-induced deformation, readily lead to the deviation of feeding wire from the molten pool, which drastically impacts the forming quality and stability during the deposition process. Therefore, it is imperative to delve into the influence mechanisms of the wire deviation response and develop the corresponding online monitoring method. In this study, the wire deviation simulation model was originally established, and the experiment method was also combined to reveal the effects of wire deviation on the wire melting process, molten pool dynamics, and the as-printed part morphology. Furthermore, a visual sensing system and corresponding image extraction algorithms were also developed, specifically designed to monitor and analyze this behavior. Results indicate with increasing deviation distance, the wire melting pattern shifts from droplet to liquid bridge mode until it fails to melt. When the deviation distance is on a small-scale, it can cause molten pool liquid outflow (liquid transition mode) and a deviation in the deposition path location (droplet transition mode) despite the existence of obvious reflux behavior. In addition, the monitoring system developed in this study can effectively protect the camera lens from being contaminated by the metal vapor and the issue of unclear wire regions caused by the overexposure of the molten pool. The gray-level co-occurrence matrix was adopted to effectively overcome the issue of unclear boundaries at the wire center, and the texture entropy feature's noise ratio only increased from 1.0 to 1.3, demonstrating good noise resistance. Based on the developed algorithm, the wire's deflection distance can be detected with an error below 0.1 mm and a response time under 10 ms. The newly revealed mechanisms and the developed monitoring technologies lay a solid foundation for the subsequent closed-loop control of wire deviation behavior, making a significant enhancement of forming stability and automation level of wire-based DED technology.

Radiography is a primary non-destructive testing method and is crucial for ensuring the quality of steel pipe welds. Although deep learning-based methods have shown significant achievements, they still encounter subjective biases during the network configuration. In addition, the problems of quantitative segmentation of small defects and motion blur remains challenging. To address these issues, this study constructed a pipe weld defect detection system based on CCD camera and image intensifier. Defect data are systematically collected and enhanced to establish a comprehensive pixel-level dataset of steel pipe weld defects. On this basis, a self-configuring transformer-based deep network (SCTnet) is proposed. It is capable of self-configuring parameters for any new dataset to generate a segmentation pipeline, reducing the subjectivity of the network. A multi-scale transformer module is proposed to integrate discriminative features across various semantic levels from a global perspective, effectively addressing the challenge of detecting small defects. Additionally, a hybrid loss function is proposed to effectively reduce the impact of motion blur. The results of using SCTnet on the steel pipe welds dataset demonstrate that the proposed segmentation method achieves competitive outcomes compared to other segmentation algorithms, with an average accuracy of 99.55% and an average dice coefficient of 78.91%.

Electron beam directed energy deposition (EB-DED) has been successfully employed to additively manufacture and repair Al-Cu alloys, however, the effects of thermal histories on the microstructure and properties have been little investigated. In this study, a specially developed Al-Si-Cu wire (380D) was deposited on the Al-Cu substrates (2A14-T6) via EB-DED to fabricate hybrid components for additive repair purposes. By using both experimental and numerical methods, the effects of substrate thickness on the thermal histories experienced and the microstructures resulted across the structural interfaces, as well as the mechanical properties of the hybrid parts, are comprehensively investigated. Results show that the columnar to equiaxed transition (CET) occurs in the deposits when the temperature gradient G and solidification velocity R satisfy the critical condition of G(2)/R < 5.075 x 10(12). A thicker substrate leads to a higher G within the molten pool, which postpones the CET and increases the area of columnar grains. The columnar grains show lower strength than the equiaxed grains and are more susceptible to fractures under tensile loads. In the bottom layer of deposits, the theta-Al2Cu and Si phases are apt to precipitate under the extremely high solidification cooling rate, forming very fine networks, and resulting in much higher strengths than the upper deposited layers. Multiple thermal cycles act as in-situ heat treatments to promote the precipitation of theta' and theta phases in the bottom deposit regions and the 2A14-T6 substrates. During the EB-DED deposition, the 2A14-T6 substrates are over-aged due to the theta'' phase coarsening and transforming into theta' and theta phases, and the decrease in strength is less when the substrate is thicker and the temperature is lower. Thus, there is a trade-off between the strengths of deposit and substrate in the hybrid parts. It is suggested that the CET in the deposit should be facilitated while maintaining the substrate temperature below 200 degrees C during the EB-DED process, so as to achieve a hybrid component with higher overall strength.

Vision-based weld seam tracking has become one of the key technologies to realize intelligent robotic welding, and weld deviation detection is an essential step. However, accurate and robust detection of weld deviations during the microwelding of ultrathin metal foils remains a significant challenge. This challenge can be attributed to the fusion zone at the mesoscopic scale and the complex time-varying interference (pulsed arcs and reflected light from the workpiece surface). In this paper, an intelligent seam tracking approach for foils joining based on spatial-temporal deep learning from molten pool serial images is proposed. More specifically, a microscopic passive vision sensor is designed to capture molten pool and seam trajectory images under pulsed arc lights. A 3D convolutional neural network (3DCNN) and long short-term memory (LSTM)-based welding torch offset prediction network (WTOP-net) is established to implement highly accurate deviation prediction by capturing long-term dependence of spatial-temporal features. Then, expert knowledge is further incorporated into the spatio-temporal features to improve the robustness of the model. In addition, the slime mould algorithm (SMA) is used to prevent local optima and improve accuracy, efficiency of WTOPnet. The experimental results indicate that the maximum error detected by our method fluctuates within +/- 0.08 mm and the average error is within +/- 0.011 mm when joining two 0.12 mm thickness stainless steel diaphragms. The proposed approach provides a basis for automated robotic seam tracking and intelligent precision manufacturing of ultrathin sheets welded components in aerospace and other fields.

During active brazing of alumina ceramics, active elements react with the ceramic to form a reaction layer, which has significant influence on the mechanical property of the brazed joint. However, the composition and formation mechanism of this layer remain unclear among researchers. To fill this gap, different brazing temperatures (900–1100 °C) and heating rates (2.5 °C/min and 10 °C/min) were used to braze 95% Al2O3 ceramics and a Kovar 4J34 alloy using a Ag-Cu-2Ti active brazing filler, and the microstructure and mechanical properties of the joints were investigated. The results show that the joint could be divided into five layers: Al2O3, ceramic-side reaction layer, filler layer, Kovar-side reaction layer, and Kovar. The ceramic-side reaction layer could be further divided into a Ti-O-rich layer and an intermetallics (IMC)-rich layer, and the Kovar-side reaction layer consists of TiFe2 particles, Ag-Cu eutectic, and the remaining Kovar. A belt-like TiFe2+TiNi3 IMC could be found in the filler layer. Increasing the brazing temperature enlarged the belt-like TiFe2+TiNi3 IMC in the filler layer and increased the thickness of the IMC-rich layer in the ceramic-side reaction layer, but had no significant effect on the thickness of the Ti-O-rich layer in the ceramic-side reaction layer. A lower heating rate (2.5 °C/min) was found to suppress the formation of the IMC-rich layer and shift the fracture location in shear tests from the ceramic-side reaction layer to the filler layer, indicating that the strength of the ceramic-side reaction layer was enhanced by controlling the formation of the IMC-rich layer. A maximum shear strength of 170 ± 61 MPa was obtained at a heating rate of 2.5 °C/min and a brazing temperature of 940 °C.

The molten pool behavior and weld formation in ultra-thin (thickness <= 0.3 mm) sheets edge welding is extremely sensitive to the variation of thermodynamic conditions, due to its unique heat transfer conditions and molten pool dynamics caused by special joint form and extremely small molten pool size. In this paper, microplasma arc source was applied to join the edge joint composed of two 0.12 mm thickness 304 stainless steel diaphragms. The typical molten pool bridging behavior was observed by a high-speed microphotography system. In addition, the formation mechanism of lack of fusion (LOF) defects was analyzed. The experimental results showed that common disturbances could affect the continuity and symmetry of melting process. Due to the instability raised by this melting process, the liquid bridge fails to form or to maintain, which is the major cause for undesirable weld and defects. Unlike sound weld formation process, the molten pool behaviors in LOF defects formation process could be classified into three states: temporarily discontinuous bridging (TDB), cyclically discontinuous bridging (CDB), and not only cyclically discontinuous but asymmetric bridging (CDAB). Comparing the TDB state, the backflow of molten pool under the CDB state tends to be more intense, leading to the occurrence of defects in succession. During the CDAB process, the molten pool is subject to lateral misalignment due to the gravitational component, resulting in asymmetric weld with defects. This study offers a comprehensive insight into molten pool behavior and weld formation process, which can enhance the understanding of ultra-thin sheets edge welding.

In this study, FeCoCrNiMo high-entropy alloy coatings were prepared using the high-speed laser cladding (HSLC) technique, and their microstructure, Vickers hardness, wear resistance, and corrosion resistance were examined. The results showed that the FeCoCrNiMo coating comprised a face-centered cubic (FCC) phase and a sub-micronsized sigma phase. The Vickers hardness of the FeCoCrNiMo coating increased with depth, which was attributed to the reduction in width of the plastically deformable FCC subcrystals. Additionally, the FeCoCrNiMo coating exhibited a maximum Vickers hardness values that were 77.1 %, 44.8 %, and 18.8 % higher than those of the 304, 431, and Ni60 coatings, respectively. The superior wear resistance of the FeCoCrNiMo coating was evidenced by its wear rate, which was 51 % to 511 % lower than that of the Ni60, 431, and 304 coatings. This improved wear resistance was mainly attributed to MoO3 functioning as a solid lubricant and MoO2 imparting a toughening effect to the oxide layer, resulting in predominantly oxidative wear in the coating. The Ecorr of the FeCoCrNiMo coating was 1.5-1.7 times higher than the 304, 431, and Ni60 coatings, while its Icorr was only a quarter of 304's, half of 431's, and slightly above that of Ni60. Therefore, the FeCoCrNiMo coating exhibited wear-corrosion resistance, making it a promising material for demanding service environments.

Multiwire submerged arc welding (MSAW) is an efficient joining technology widely used in manufacturing large-diameter oil and gas steel pipes. During spiral pipe MSAW, the long cantilever welding torch (LCWT) for inner welding is prone to vibration due to the magnetic effects from high welding currents, which decreases welding stability. This paper studies the magnetic field for the internal welding scenario in spiral pipes using a combination of experimental measurements and finite element methods. Welding experiments were performed on spiral pipe MSAW equipment, and the influence of the magnetized steel pipe was investigated. Finite element methods were then performed to simulate magnetic field distribution and the electromagnetic loads on the LCWT. Finally, the response of the LCWT to electromagnetic loads was computed using an analytical method. Based on magnetic field modeling, this study reveals the magnetization of steel pipes and its influence on magnetic distribution. Further, it confirms that the vibrations of LCWT are due to the electromagnetic load from the magnetized steel pipe. This study identifies the physical mechanism of torch vibration under electromagnetic excitation, providing a basis for designing stable and efficient spiral pipe final welding systems.