In this work, an insight into the premature Type IV cracking was undertaken to clarify its mechanisms in Grade 91 steel pipe weldments. High-resolution microscopy observations of the as-welded ...heat-affected zone (HAZ) reveal that the commonly recognized fine-grained region susceptible of cracking on the edge of HAZ belongs to the inter-critical HAZ (ICHAZ), rather than the fine-grained HAZ (FGHAZ). Instrumented indentation tests uncover that the ICHAZ is the weakest region across the weld, exhibiting the largest displacement and the lowest hardness in three thermal stages. Localized deformation of matrix grains and high stress triaxiality in the ICHAZ promoted nucleation of creep cavities along grain boundaries. This localized deformation was induced by the creep strength mismatch of matrix grains with different Cr concentrations. Cavity-free regions exhibit a relatively homogenous Cr distribution, whereas, an inhomogeneous Cr distribution is observed in the cavity-containing regions. It is believed that this local Cr inhomogeneity in the ICHAZ is caused by the partial dissolution of Cr-rich M23C6 carbides and an insufficient homogenization during rapid welding thermal cycles.
Display omitted
•Heterogenous fine-grained strucuture belonging to the intercritical heat-affected zone degraded faster than the other regions.•Localized C and Cr inhomogeneities in the matrix weakened intercritical heat-affected zone.•Localized high stress triaxiality around coarsened precipitates induced cavity nucleation at the triple juctions of grain boundaries.
High-precision positioning and multi-target detection have been proposed as key technologies for robotic path planning and obstacle avoidance. First, the Cartographer algorithm was used to generate ...high-quality maps. Then, the iterative nearest point (ICP) and the occupation probability algorithms were combined to scan and match the local point cloud, and the positions and attitudes of the robot were obtained. Furthermore, Sparse Matrix Pose Optimization was carried out to improve the positioning accuracy. The positioning accuracy of the robot in x and y directions was kept within 5 cm, the angle error was controlled within 2°, and the positioning time was reduced by 40%. An improved timing elastic band (TEB) algorithm was proposed to guide the robot to move safely and smoothly. A critical factor was introduced to adjust the distance between the waypoints and the obstacle, generating a safer trajectory, and increasing the constraint of acceleration and end speed; thus, smooth navigation of the robot to the target point was achieved. The experimental results showed that, in the case of multiple obstacles being present, the robot could choose the path with fewer obstacles, and the robot moved smoothly when facing turns and approaching the target point by reducing its overshoot. The proposed mapping, positioning, and improved TEB algorithms were effective for high-precision positioning and efficient multi-target detection.
With the rapid development of vision sensing, artificial intelligence, and robotics technology, one of the challenges we face is installing more advanced vision sensors on welding robots to achieve ...intelligent welding manufacturing and obtain high-quality welding components. Depth perception is one of the bottlenecks in the development of welding sensors. This review provides an assessment of active and passive sensing methods for depth perception and classifies and elaborates on the depth perception mechanisms based on monocular vision, binocular vision, and multi-view vision. It explores the principles and means of using deep learning for depth perception in robotic welding processes. Further, the application of welding robot visual perception in different industrial scenarios is summarized. Finally, the problems and countermeasures of welding robot visual perception technology are analyzed, and developments for the future are proposed. This review has analyzed a total of 2662 articles and cited 152 as references. The potential future research topics are suggested to include deep learning for object detection and recognition, transfer deep learning for welding robot adaptation, developing multi-modal sensor fusion, integrating models and hardware, and performing a comprehensive requirement analysis and system evaluation in collaboration with welding experts to design a multi-modal sensor fusion architecture.
During steel production, various defects often appear on the surface of the steel, such as cracks, pores, scars, and inclusions. These defects may seriously decrease steel quality or performance, so ...how to timely and accurately detect defects has great technical significance. This paper proposes a lightweight model based on multi-branch dilated convolution aggregation and multi-domain perception detection head, DAssd-Net, for steel surface defect detection. First, a multi-branch Dilated Convolution Aggregation Module (DCAM) is proposed as a feature learning structure for the feature augmentation networks. Second, to better capture spatial (location) information and to suppress channel redundancy, we propose a Dilated Convolution and Channel Attention Fusion Module (DCM) and Dilated Convolution and Spatial Attention Fusion Module (DSM) as feature enhancement modules for the regression and classification tasks in the detection head. Third, through experiments and heat map visualization analysis, we have used DAssd-Net to improve the receptive field of the model while paying attention to the target spatial location and redundant channel feature suppression. DAssd-Net is shown to achieve 81.97% mAP accuracy on the NEU-DET dataset, while the model size is only 18.7 MB. Compared with the latest YOLOv8 model, the mAP increased by 4.69%, and the model size was reduced by 23.9 MB, which has the advantage of being lightweight.
Crack-tip opening displacement (CTOD) tests were conducted on girth welds of two API 5L X70 pipeline steels (pipe A and pipe B) to investigate the influence of base metal composition on the fracture ...toughness of the joint. CTOD measurements across the weld showed that the weld fusion zone had the lowest CTOD values for both pipes, with pipe B having a higher CTOD value than pipe A. Detailed microstructure characterization of the multi-pass weld showed that the fusion zone in both pipes consisted of three distinct zones: the columnar zone, the coarse equiaxed zone, and the fine equiaxed zone. Both the columnar zone and coarse-grained equiaxed zone had acicular ferrite and grain boundary ferrite microstructures, whereas the fine-grained equiaxed zone had a finer ferrite microstructure compared to the other two zones. The main difference between the two pipes was the variation in ferrite grain sizes and the volume fractions of grain boundary ferrite and acicular ferrite. In comparison to pipe B, pipe A, with a higher concentration of Mo, Ni, and Cu in both the base metal and the weld fusion zones, consisted of a higher volume fraction of grain boundary ferrite and a lower volume fraction of acicular ferrite in the columnar and coarse-grained equiaxed zones. The lower concentration of Mo, Ni, and Cu in pipe B likely resulted in the formation of a predominantly acicular ferrite microstructure in the fusion zone, thereby improving the toughness of the weld joint in comparison to pipe A.
Flow-accelerated corrosion (FAC) preferentially attacks the downstream heat-affected zone of the root-pass weld in steam pipe systems. A detailed characterization identifies the fusion boundary as ...the initiation location for the attack. Alloying elements are found depleted along the weld fusion boundary, and multiple welding thermal cycles and repetitive austenite-to-ferrite phase transformations result in an increased proportion of grains with Goss {110} texture along the fusion boundary. The synergistic effects of chemical segregation and the Schmid factor may contribute to the preferential initiation of FAC cracks along the root weld fusion boundary, making it the weakest link for FAC attack in steam pipe girth welds.
Periodic inspection, commonly performed by a technician, of weld seam quality is important for assessing equipment reliability. To save labor costs and improve efficiency, an autonomous navigation ...and inspection robot is developed. The development process involves the design of chassis damping, target detection mechanism, control system, and algorithms. For performing weld inspection in complex, outdoor, environments, an algorithm is developed for the robot to avoid any obstacles. This algorithm for planning the inspection route is based on an improved timed-elastic-band (TEB) algorithm. The developed robot is capable of conducting inspection tasks in complex and dangerous environments efficiently and autonomously.
Ensemble learning has attracted considerable attention owing to its good generalization performance. The main issues in constructing a powerful ensemble include training a set of diverse and accurate ...base classifiers, and effectively combining them. Ensemble margin, computed as the difference of the vote numbers received by the correct class and the another class received with the most votes, is widely used to explain the success of ensemble learning. This definition of the ensemble margin does not consider the classification confidence of base classifiers. In this work, we explore the influence of the classification confidence of the base classifiers in ensemble learning and obtain some interesting conclusions. First, we extend the definition of ensemble margin based on the classification confidence of the base classifiers. Then, an optimization objective is designed to compute the weights of the base classifiers by minimizing the margin induced classification loss. Several strategies are tried to utilize the classification confidences and the weights. It is observed that weighted voting based on classification confidence is better than simple voting if all the base classifiers are used. In addition, ensemble pruning can further improve the performance of a weighted voting ensemble. We also compare the proposed fusion technique with some classical algorithms. The experimental results also show the effectiveness of weighted voting with classification confidence.
•We give a new definition of margin based on classification confidence of base classifiers in ensemble learning.•We construct optimal objective functions based on margin distribution for obtaining weights of base classifiers.•Difference strategies to utilize the weights and classification confidence in the final decision are tried.•Extensive experiments are conducted to compare different solutions and an optimal solution is derived.
GTD-111 directionally solidified superalloy was repaired by the Nd:YAG laser based laser-engineered net shaping (LENS) process. Selection of process parameters are discussed against the formation of ...deposits and occurrence of defects. Formation of microfissures, porosity, lack of fusion, and stray crystals in the deposit zone was studied. It was identified that formation of stray crystals in the directionally solidified alloy could be due to two types of sources: the shape of the melt pool, and the existence of second-phase particles at the melt pool fusion boundary.
Display omitted
•Cruciform biaxial tensile test were performed in rolling/transverse and 45°/135° sampling directions under different loading paths.•Priority of calibrating the material parameters ...with the equi-biaxial data should be ensured.•The introduction of plane strain yield stresses are positive for the accurate prediction of plastic anisotropy.•Non-integer exponent is helpful to improve the description accuracy of anisotropy yield behavior.•Biaxial tensile data under principal stress state are not enough to evaluate the best identification method.
It is necessary to identify differences among biaxial tensile mechanical properties to describe the plastic anisotropy and potential adjustment ability of yield criteria with the non-integer exponent for the yield surface. Therefore, in this study, uniaxial and cruciform biaxial tensile tests were performed under 17 different loading paths: uniaxial tension in seven different directions, cruciform biaxial tension in rolling/transverse and 45°/135° sampling directions with seven and three different stress ratios, respectively. Based on the BBC2008 yield criterion, the uniaxial yield stresses, rθ-values, yield loci on the normal plane, and shear yield loci on the diagonal plane, predicted using six parameter identification strategies, were quantitatively evaluated for MP980, DP490, 6016-T4, and 5182-O. Results show the constraining and regulation ability of the equi-biaxial tensile data for yield loci to be better than that of near-plane strain state data. The parameter identification strategy considering the non-integer exponent was observed to significantly improved the ability of the yield criterion to describe the anisotropic yield behavior. For a simplified evaluation system that considers only the prediction accuracy of the yield locus under the principal stress state, neglecting the prediction accuracy for the shear yield locus may lead to incorrect judgments regarding the best identification strategy.