Publications without fulltext

Computer Aided Engineering (CAE) methods such as finite element based simulation and optimization techniques have become invaluable for product development in the automotive industry. Due to the continuously growing computational power and the introduction of cloud infrastructures, CAE tools are becoming more powerful, faster and cheaper to deploy. However, they still require a lot of expert knowledge in order to be used correctly and effectively. This is why larger corporations usually employ specially trained simulation engineers who help design engineers to set-up, run and post-process simulation scenarios. In order to make simulation technology ubiquitous even in small and medium sized enterprises, it is necessary to develop CAE tools that are equipped with intuitive interfaces and that can be used by non-experts (CAE democratization).Recent advances in artificial intelligence have sparked the development of novel computer-based assistance systems in many domains such as autonomous vehicles, robotics and medicine. These systems demonstrate how knowledge-based approaches can either augment the capabilities of human experts or even replace them. Currently, these methods have seen very limited use within the CAE community.

[Präsentationsfolien]

Energy generation is moving away from centralized fossil fuel based generators towards renewable energy to provide clean and reliable sources. Hydrogen-based generation such as solid oxide fuel cell is one of the promising solution. For efficient and optimized operations of the overall system, e.g. frequency or voltage support actions, accurate dynamic models of the generators can be highly beneficial. Those are often not provided by manufacturers in sufficient detail. Since the dynamics of fuel cells are non-linear and depend on a high number of hard-to-measure parameters, white-box models are often hard or impossible to implement. The goal of this work is to develop and implement methods for data-driven physics-based model identification for partially unknown solid oxide fuel cells, that function with minimal measurement data. A mechanistic gray box model, a pre-trained feed forward neural network and long short-term memory neural network are implemented. They are evaluated by comparing their output to that of a simulated fuel cell stack in different scenarios. For large variations in operating conditions, the feed forward network shows the best performance. Close to the maximum power point, the long-short term memory based model performs best.

The climate change and the shortage of petroleum resources are two critical global issues that have led to an increased focus on sustainability, circular economy, and the use of renewable resources. Bioplastics, such as poly(lactic acid/lactide) (PLA), which are derived from agricultural resources, have emerged as a potential carbon neutral solution to these challenges. Moreover, one promising approach is the chemical recycling of PLA, closing the loop of its circular economy by using PLA-wastes as a feedstock for new and high-quality plastic products. Within the framework of this Call-for-Transfer project the potential of chemical recycling of PLA and its scale-up were investigated.

Monitoring of batch production processes is complex and existing solutions do not offer good performance in providing real-time feedback about the state of the process. Therefore, we introduce an AI system that monitors a fermentation process and detects deviations from the normal process execution directly on the edge and provides real-time feedback to the operator, allowing intervention before the process gets out of control. We analyze the accuracy of the novel AI-based approach by carrying out several experiments and compare the outcome with statistical methods as a baseline. The experiments show that the AI-based approach performs significantly better at detecting anomalies in a fermentation process than the statistical methods.