Search

Wire-arc additive manufacturing (WAAM) is a promising industrial production technique. Without optimization, inherent temperature gradients can cause powerful residual stresses and microstructural defects. There is therefore a need for data-driven methods allowing real-time process optimization for WAAM. This study focuses on machine learning (ML)-based prediction of temperature history for WAAM-produced aluminum bars with different geometries and process parameters, including bar length, number of deposition layers, and heat source movement speed. Finite element (FE) simulations are used to provide training and prediction data. The ML models are based on a simple multilayer perceptron (MLP) and performed well during baseline training and testing, giving a testing mean absolute percentage error (MAPE) of less than 0.7% with an 80/20 train–test split, with low variation in model performance. When using the trained models to predict results from FE simulations with greater length or number of layers, the MAPE increased to an average of 3.22% or less, with greater variability. In the cases of greatest difference, some models still returned a MAPE of less than 1%. For different scanning speeds, the performance was worse, with some outlier models giving a MAPE of up to 14.91%. This study demonstrates the transferability of temperature history for WAAM with a simple MLP approach. [ABSTRACT FROM AUTHOR]

The multi-material design of modern car bodies requires joining technologies for dissimilar materials. Adhesive bonding in combination with self-piercing riveting is widely used for joining steel and aluminium structures. To guarantee crashworthiness and reliability of a car body, accurate and effcient numerical models of its materials and connections are required. Suitable component test setups are necessary for development and validation of such models. In this work, a novel test setup for adhesively bonded and point-wise connected components is presented. Here, load combinations comparable to a vehicle crash are introduced into the connections. The developed setup facilitates successive failure of multiple connections and enables a broad validation of numerical connection models.

Forests are one of the key elements in ecological transition policies in Europe. Sustainable forest management is needed in order to optimise wood harvesting, while preserving carbon storage, biodiversity and other ecological functions. Forest managers and public bodies need improved and cost-effective forest monitoring tools. Research studies have been carried out to assess the use of optical and radar images for producing forest height or biomass maps. The main limitations are the quantity, quality and representativeness of the reference data for model training. The Global Ecosystem Dynamics Investigation (GEDI) mission (full waveform LiDAR on board the International Space Station) has provided an unprecedented number of forest canopy height samples from 2019. These samples could be used to improve reference datasets. This paper aims to present and validate a method for estimating forest dominant height from open access optical and radar satellite images (Sentinel-1, Sentinel-2 and ALOS-2 PALSAR-2), and then to assess the use of GEDI samples to replace field height measurements in model calibration. Our approach combines satellite image features and dominant height measurements, or GEDI metrics, in a Support Vector Machine regression algorithm, with a feature selection process. The method is tested on mixed uneven-aged broadleaved and coniferous forests in France. Using dominant height measurements for model training, the cross-validation shows 7.3 to 11.6% relative Root Mean Square Error (RMSE) depending on the forest class. When using GEDI height metrics instead of field measurements for model training, errors increase to 12.8–16.7% relative RMSE. This level of error remains satisfactory; the use of GEDI could allow the production of dominant height maps on large areas with better sample representativeness. Future work will focus on confirming these results on new study sites, improving the filtering and processing of GEDI data, and producing height maps at regional or national scale. The resulting maps will help forest managers and public bodies to optimise forest resource inventories, as well as allow scientists to integrate these cartographic data into climate models. [ABSTRACT FROM AUTHOR]

The Mesoproterozoic Rivard minette dyke, Grenville Province, Quebec, has undergone dyke-parallel fracturing due to the movement of low-velocity melts at deep crustal levels. The brittle-to-ductile deformation and big dense fragments in the dyke's xenoliths suggest turbulence and upward movement of magma. The xenoliths were probably formed by dyke wall fragmentation and delamination of the wall rock due to magma intrusion from the fractures. An increase in the xenolith level increased the velocity of the liquid-solid mixture, and promoted magma flow and xenolith suspension.

• Implementation of a non-local GTN model to avoid spurious mesh dependency. • Applied to model damage evolution and crack propagation until complete fracture. • Validated against corresponding local GTN model. • Over-aggressive regularization is avoided by evaluating on current rather than initial configuration. In this study, we follow the work of Tvergaard and Needleman (1995, 1997) and Needleman and Tvergaard (1998) and present the numerical implementation and initial applications of a non-local Gurson-Tvergaard-Needleman (GTN) model for explicit finite element (FE) analysis. The delocalization relates to the damage mechanism and is incorporated in terms of an integral condition on the rate of change of the porosity. To demonstrate the mesh independence during all stages of ductile damage and fracture, several material test specimens were simulated using different mesh sizes until full fracture occurred. For comparison purposes, the results are also obtained for the corresponding local GTN model in all cases. The effect of the material characteristic length on the ductile damage and fracture behavior and on the mesh sensitivity of the results is discussed. The numerical study shows that simulation results obtained in all stages of the ductile fracture process, including void growth, fracture initiation by coalescence and crack propagation all the way to a fully fractured specimen, are mesh independent for a certain mesh size ratio related to the material characteristic length, provided the non-local integral is evaluated on the current configuration. This ratio is unique for each individually simulated specimen as it depends on the spatial gradients of the porosity and the material parameters adopted for the problem at hand. It is shown that excessive averaging occurs at large deformations if the non-local integral is evaluated on the reference configuration, i.e., without updating the element interaction matrix resulting from the discretization of the non-local integral. [ABSTRACT FROM AUTHOR]

This work examines the effects of loading rate on the plastic flow and ductile failure of porous solids exhibiting rate-dependent behavior relevant to many structural metals. Two different modeling approaches for ductile failure are employed and numerical analyses are performed over a wide range of strain rates. Finite element unit cell simulations are carried out to evaluate the macroscopic mechanical response and ductile failure by void coalescence for various macroscopic strain rates. The unit cell results are then used to assess the accuracy of a rate-dependent porous plasticity model, which is subsequently used in strain localization analyses based on the imperfection band approach. Strain localization analyses are conducted for (i) proportional loading paths and (ii) non-proportional loading paths obtained from finite element simulations of axisymmetric and flat tensile specimens. The effects of strain rate are most apparent on the stress-strain response, whereas the effects of strain rate on ductile failure is found to be small for the adopted rate-dependent constitutive model. However, the rate-dependent constitutive formulation tends to regularize the plastic strain field when the strain rate increases. In the unit cell simulations, this slightly increases the strain at coalescence with increasing strain rate compared to a rate-independent constitutive formulation. When the strain rate is sufficiently high, the strain at coalescence becomes constant. The strain localization analyses show a negligible effect of strain rate under proportional loading, while the effect of strain rate is more pronounced when non-proportional loading paths are assigned. [ABSTRACT FROM AUTHOR]