Showing papers in "Welding in The World in 2022"

Automated defect detection in digital radiography of aerospace welds using deep learning

Abstract: Abstract Aerospace welds are non-destructively evaluated (NDE) during manufacturing to identify defective parts that may pose structural risks, often using digital radiography. The analysis of these digital radiographs is time consuming and costly. Attempts to automate the analysis using conventional computer vision methods or shallow machine learning have not, thus far, provided performance equivalent to human inspectors due to the high reliability requirements and low contrast to noise ratio of the defects. Modern approaches based on deep learning have made considerable progress towards reliable automated analysis. However, limited data sets render current machine learning solutions insufficient for industrial use. Moreover, industrial acceptance would require performance demonstration using standard metrics in non-destructive evaluation, such as probability of detection (POD), which are not commonly used in previous studies. In this study, data augmentation with virtual flaws was used to overcome data scarcity, and compared with conventional data augmentation. A semantic segmentation network was trained to find defects from computed radiography data of aerospace welds. Standard evaluation metrics in non-destructive testing were adopted for the comparison. Finally, the network was deployed as an inspector’s aid in a realistic environment to predict flaws from production radiographs. The network achieved high detection reliability and defect sizing performance, and an acceptable false call rate. Virtual flaw augmentation was found to significantly improve performance, especially for limited data set sizes, and for underrepresented flaw types even at large data sets. The deployed prototype was found to be easy to use indicating readiness for industry adoption.

Effect of the Heat Input on Wire-Arc Additive Manufacturing of Invar 36 Alloy: Microstructure and Mechanical Properties

Abstract: Abstract Invar, also known as FeNi36, is a material of great interest due to its unique properties, which makes it an excellent alternative for sectors such as tooling in aeronautics and aerospace. Its manufacture by means of wire arc additive manufacturing (WAAM) technology could extend its use. This paper aims to evaluate the comparison of two of the most widespread WAAM technologies: plasma arc welding (PAW) and gas metal arc welding (GMAW). This comparison is based on the analysis of wall geometry, metallography, and mechanical properties of the material produced by both technologies. The results show a slight increase in toughness and elongation before fracture and worse tensile strength data in the case of PAW, with average values of 485 MPa for ultimate tensile strength (UTS), 31% for elongation and 475 MPa, 40% in GMAW and PAW, respectively. All results gathered from the analysis show the possibility of successful manufacturing of Invar by means of WAAM technologies. The novelties presented in this paper allow us to establish relationships between the thermal input of the process itself and the mechanical and metallographic properties of the material produced.

EBSD characterisation of grain size distribution and grain sub-structures for ferritic steel weld metals

Abstract: Abstract Microstructural characterisation of engineering materials is required for understanding the relationships between microstructure and mechanical properties. Conventionally grain size is measured from grain boundary maps obtained using optical or electron microscopy. This paper implements EBSD-based linear intercept measurement of spatial grain size variation for ferritic steel weld metals, making analysis flexible and robust. While grain size has been shown to correlate with the strength of the material according to the Hall–Petch relationship, similar grain sizes in weld metals with different phase volume fractions can have significantly different mechanical properties. Furthermore, the solidification of the weld pool induces the formation of grain sub-structures that can alter mechanical properties. The recently developed domain misorientation approach is used in this study to provide a more comprehensive characterisation of the grain sub-structures for ferritic steel weld metals. The studied weld metals consist of varying mixtures of primary ferrite, acicular ferrite, and bainite/martensite, with large differences observed in hardness, grain size, grain morphology, and dislocation cell size. For the studied weld metals, the average dislocation cell size varied between 0.68 and 1.41 µm, with bainitic/martensitic weld metals showing the smallest sub-structures and primary ferrite the largest. In contrast, the volume-weighted average grain size was largest for the bainitic/martensitic weld metal. Results indicate that a Hall–Petch-type relationship exists between hardness and average dislocation cell size and that it partially corrects the significantly different grain size—hardness relationship observed for ferritic and bainitic/martensitic weld metals. The methods and datasets are provided as open access.

Laser beam welding setup for the coaxial combination of two laser beams to vary the intensity distribution

Abstract: Abstract Deep-penetration laser beam welding is highly dynamic and affected by many parameters. Several investigations using differently sized laser spots, spot-in-spot laser systems, and multi-focus optics show that the intensity distribution is one of the most influential parameters; however, the targeted lateral and axial intensity design remains a major challenge. Therefore, a laser processing optic has been developed that coaxially combines two separate laser sources/beams with different beam characteristics and a measuring beam for optical coherence tomography (OCT). In comparison to current commercial spot-in-spot laser systems, this setup not only makes it possible to independently vary the powers of the two laser beams but also their focal planes, thus facilitating the investigation into the influence of specific energy densities along the beam axis. First investigations show that the weld penetration depth increases with increasing intensities in deeper focal positions until the reduced intensity at the sample surface, due to the deep focal position, is no longer sufficient to form a stable keyhole, causing the penetration depth to drop sharply.

Theory-inspired machine learning—towards a synergy between knowledge and data

Abstract: Abstract Most engineering domains abound with models derived from first principles that have beenproven to be effective for decades. These models are not only a valuable source of knowledge, but they also form the basis of simulations. The recent trend of digitization has complemented these models with data in all forms and variants, such as process monitoring time series, measured material characteristics, and stored production parameters. Theory-inspired machine learning combines the available models and data, reaping the benefits of established knowledge and the capabilities of modern, data-driven approaches. Compared to purely physics- or purely data-driven models, the models resulting from theory-inspired machine learning are often more accurate and less complex, extrapolate better, or allow faster model training or inference. In this short survey, we introduce and discuss several prominent approaches to theory-inspired machine learning and show how they were applied in the fields of welding, joining, additive manufacturing, and metal forming.

Modification of Co–Cr alloys to optimize additively welded microstructures and subsequent surface finishing

Abstract: Abstract Cobalt chromium alloys are often used in turbine and plant construction. This is based on their high thermal and mechanical stress resistance as well as their high wear resistance to corrosive and abrasive loads. However, cobalt is a cost-intensive material that is difficult to machine. Moreover, increasingly complex structures and the optimisation of resource efficiency also require additive manufacturing steps for the production or repair of components in many sectors. Concerning inhomogeneity and anisotropy of the microstructure and properties as well as manufacturing-related stresses, a lot of knowledge is still necessary for the economic use of additive welding processes in SMEs. As a result of the high stresses on the components and requirements for a high surface quality, a complementary use of additive and machining manufacturing processes is necessary. Thereby, Co–Cr alloys are extremely challenging for machining with geometrically defined cutting edges because of their low thermal conductivity combined with high strength and toughness. An approach to solve this problem is to refine and homogenise the microstructure. This is achieved by modifying the alloy with elements zirconium and hafnium, which are added up to a maximum of 1 wt.-%. A reduction of the process forces and stresses on the tool and work piece surface is also achievable via hybrid milling processes. There are already studies on the combined use of additive and machining manufacturing processes based on laser technology. However, knowledge based on powder and wire-based arc processes is important, as these processes are more widespread. Furthermore, the effects on the surface zone of additively manufactured components by hybrid finish milling have not yet been a subject of research. The results show that the structural morphology could be significantly influenced with the addition of zirconium and hafnium.

Laser welding of 16MnCr5 butt welds with gap: resulting weld quality and fatigue strength assessment

Abstract: Abstract For laser welding, gaps should normally be avoided to ensure stable weld processes and high weld quality. Nonetheless, sometimes, gaps are resulting from non-optimal weld preparation in industrial applications. Within this investigation, the effects of a defocussed beam and laser beam oscillation on gap bridging abilities at reduced ambient pressure were investigated. For reference purpose, conventional laser welding with zero gap at ambient pressure was performed, too. The resulting weld geometry was investigated and correlated to gap bridging strategies and weld quality groups according to ISO 13919–1. The welds were characterized regarding their hardness and weld microstructure. Residual stress was determined by means of X-ray diffraction, and tensile tests as well as fatigue tests were conducted. The fatigue tests were evaluated by the nominal stress approach, the critical distance approach, and the stress averaging approach and correlated to weld quality measures. Resulting from this, fatigue resistance of laser welded butt welds with gap can be estimated by the FAT160 design S–N curve. The stress evaluation parameters for the determination of k eff -values ( ρ* = 0.4 mm, a = 0.1 mm) were confirmed.

Determination of notch factors for transverse non-load carrying stiffeners based on numerical analysis and metamodeling

Abstract: Abstract Stress concentration factors (SCFs) at weld toes and weld roots as required for the effective notch stress concept (see [1, 2]) are usually computed using finite element analysis (FEA) which requires a certain amount of effort for model generation, the solving process, and postprocessing. Regression functions of many FEAs within given parameter bounds provide the possibility of a fast prediction of SCFs. This paper provides new and accurate regression formulae for the estimation of notch stresses at idealized weld geometries on the basis of multiple linear-elastic FEAs for the transverse stiffener (non-load carrying T-joint) under tension and bending of the load carrying slab. Regression of sampled finite element results has been performed using (a) second-order polynomial regression with coupling terms (PRC) and (b) artificial neural networks (ANN). The presented formulae are compared with several existing estimations of stress concentration factors. The new methods appear to show a higher quality of prognosis as well as apply to significant larger ranges of the geometrical parameters of the weld joint. The formulae presented here for the transverse stiffener add another welded joint to a series of similar surrogate models presented from Munich University of Applied Sciences in earlier publications and made available for use by the web-based tool SCF-Predictor .