Filtern
Erscheinungsjahr
- 2023 (3) (entfernen)
Dokumenttyp
Sprache
- Englisch (3)
Referierte Publikation
- ja (3) (entfernen)
Schlagworte
- Artificial Intelligence (3) (entfernen)
Organisationseinheit der BAM
It is shown that preconditioning of experimental X‐ray computed tomography (XCT) data is critical to achieve high‐precision segmentation scores. The challenging experimental XCT datasets and deep convolutional neural networks (DCNNs) are used that are trained with low‐resemblance synthetic XCT data. The material used is a 6‐phase Al–Si metal matrix composite‐reinforced with ceramic fibers and particles. To achieve generalization, in our past studies, specific data augmentation techniques were proposed for the synthetic XCT training data. In addition, two toolsets are devised: (1) special 3D DCNN architecture (3D Triple_UNet), slicing the experimental XCT data from multiple views (MultiView Forwarding), the i.S.Sy.Da.T.A. iterative segmentation algorithm, and (2) nonlocal means (NLM) conditioning (filtering) for the experimental XCT data. This results in good segmentation Dice scores across all phases compared to more standard approaches (i.e., standard UNet architecture, single view slicing, standard single training, and NLM conditioning). Herein, the NLM filter is replaced with the deep conditioning framework BAM SynthCOND introduced in a previous publication, which can be trained with synthetic XCT data. This leads to a significant segmentation precision increase for all phases. The proposed methods are potentially applicable to other materials and imaging techniques.
AbstractLaser beam welding has become widely applied in many industrial fields in recent years. Solidification cracks remain one of the most common welding faults that can prevent a safe welded joint. In civil engineering, convolutional neural networks (CNNs) have been successfully used to detect cracks in roads and buildings by analysing images of the constructed objects. These cracks are found in static objects, whereas the generation of a welding crack is a dynamic process. Detecting the formation of cracks as early as possible is greatly important to ensure high welding quality. In this study, two end-to-end models based on long short-term memory and three-dimensional convolutional networks (3D-CNN) are proposed for automatic crack formation detection. To achieve maximum accuracy with minimal computational complexity, we progressively modify the model to find the optimal structure. The controlled tensile weldability test is conducted to generate long videos used for training and testing. The performance of the proposed models is compared with the classical neural network ResNet-18, which has been proven to be a good transfer learning model for crack detection. The results show that our models can detect the start time of crack formation earlier, while ResNet-18 only detects cracks during the propagation stage.
The new crack luminescence method offers the possibility of making fatigue surface cracks in metallic materials more visible during inspections through a special coating system. This coating system consists of two layers, whereby the first layer has fluorescent properties and emits visible light as soon as it is irradiated by UV light. The top layer is black and is designed to prevent the fluorescent layer from emitting if no crack develops in the underlying material. The technique proved particularly useful in a wide variety of fatigue tests of steel components under laboratory conditions. Moreover, it has the potential to be used in various industrial applications. To enable industrial deployment and integration into maintenance strategies, a concept study is developed in this contribution, resulting in a qualification framework that can serve as a foundation for determining the reliability of the crack luminescence system in terms of a probability of detection curve. Within this study, factors causing measurement variability and uncertainty are being determined and their influences assessed. Due to the extension of the system by a moving computer vision system for automated crack detection using artificial intelligence, additional long-term effects associated with structural health monitoring systems need to be incorporated into an extended probability of detection study as part of the technical justification. Finally, important aspects and findings related to design of experiments are discussed, and a framework for reliability assessment of a new optical crack monitoring method is presented, emphasizing the influence of various uncertainty parameters, including long-term effects such as system ageing.