In order to ensure high weld qualities and structural integrity of engineering structures, it is crucial to detect areas of high stress concentrations along weld seams. Traditional inspection methods rely on visual inspection and manual weld geometry measurements. Recent advances in the field of automated measurement techniques allow virtually unrestricted numbers of inspections by laser measurements of weld profiles; however, in order to compare weld qualities of different welding processes and manufacturers, a deeper understanding of statistical distributions of stress concentrations along weld seams is required. Hence, this study presents an approach to statistically characterize different types of butt joint weld seams. For this purpose, an artificial neural network is created from 945 finite element simulations to determine stress concentration factors at butt joints. Besides higher quality of predictions compared to empirical estimation functions, the new approach can directly be applied to all types welded structures, including arc- and laser-welded butt joints, and coupled with all types of 3D-measurement devices. Furthermore, sheet thickness ranging from 1 mm to 100 mm can be assessed.

A digital twin describes the virtual representation of a real process. This twin is constantly updated with real data and can thus control and adapt the real model. Designing suitable digital twins for path planning of autonomous robots or drones is often challenging due to the large number of different dynamic environments and multi-task and agent systems. However, common path algorithms are often limited to two tasks and to finding shortest paths. In real applications, not only a short path but also the width of the passage with a path as centered as possible are crucial, since robotic systems are not ideal and require recalibration frequently. In this work, so-called homotopic shrinking is used to generate the digital twin, which can be used to extract all possible path proposals including their passage widths for 2D and 3D environments and multiple tasks and robots. The erosion of the environment is controlled by constraints such that the task stations, the robot or drone positions, and the topology of the environment are considered. Such a deterministic path algorithm can flexibly respond to changing environmental conditions and consider multiple tasks simultaneously for path generation. A distinctive feature of these paths is the central orientation to the non-passable areas, which can have significant benefits for worker and patient safety. The method is tested on 2D and 3D maps with different tasks, obstacles, and multiple robots. For example, the robust generation of the digital twin for a maze and also the dynamic adaptation in case of sudden changes in the environment is covered. This variety of use cases and the comparison with alternative methods result in significant advantages, such as high robustness, consideration of multiple targets, and high safety distances to obstacles and areas that cannot be traversed. Finally, it was shown that the environment for the digital twin can be reduced to reasonable paths by constrained shrinking, both for real 2D maps and for complex virtual 2D and 3D maps.

Multi-Objective Topology Optimization is a tool for finding low-weight solutions using a discretized geometry of several objective functions. In this contribution; the coupling of heat conduction and elastostatics is covered using the global criteria meth

Feature lines such as sharp edges are the main characteristic lines of a surface. These lines are suitable as a basis for surface reconstruction and reverse engineering [1]. A supervised deep learning approach based on graph convolutional networks on estimating local feature lines will be introduced in the following. We test this deep learning architecture on two provided data sets of which one covers sharp feature lines and the other arbitrary feature lines based on unnoisy meshed constructive solid geometry [CSG]. Furthermore. we use a data balancing strategy by classifying different feature line types. We then compare the selected architecture with classical machine learning models. Finally. we show the detection of these lines on noisy and deformed meshes.