Refine
Year of publication
- 2024 (49)
- 2023 (114)
- 2022 (143)
- 2021 (154)
- 2020 (171)
- 2019 (194)
- 2018 (171)
- 2017 (154)
- 2016 (158)
- 2015 (176)
- 2014 (167)
- 2013 (173)
- 2012 (163)
- 2011 (189)
- 2010 (185)
- 2009 (189)
- 2008 (156)
- 2007 (149)
- 2006 (160)
- 2005 (130)
- 2004 (161)
- 2003 (106)
- 2002 (130)
- 2001 (106)
- 2000 (108)
- 1999 (109)
- 1998 (99)
- 1997 (99)
- 1996 (81)
- 1995 (78)
- 1994 (86)
- 1993 (59)
- 1992 (54)
- 1991 (29)
- 1990 (39)
- 1989 (45)
- 1988 (57)
- 1987 (32)
- 1986 (19)
- 1985 (34)
- 1984 (22)
- 1983 (20)
- 1982 (29)
- 1981 (20)
- 1980 (36)
- 1979 (24)
- 1978 (34)
- 1977 (14)
- 1976 (13)
- 1975 (12)
- 1974 (3)
- 1973 (2)
- 1972 (2)
- 1971 (1)
- 1968 (1)
Institute
- Fachbereich Medizintechnik und Technomathematik (1689)
- Fachbereich Elektrotechnik und Informationstechnik (718)
- IfB - Institut für Bioengineering (624)
- Fachbereich Energietechnik (588)
- INB - Institut für Nano- und Biotechnologien (557)
- Fachbereich Chemie und Biotechnologie (551)
- Fachbereich Luft- und Raumfahrttechnik (497)
- Fachbereich Maschinenbau und Mechatronik (279)
- Fachbereich Wirtschaftswissenschaften (217)
- Solar-Institut Jülich (165)
Language
- English (4909) (remove)
Document Type
- Article (3277)
- Conference Proceeding (1162)
- Part of a Book (191)
- Book (144)
- Doctoral Thesis (30)
- Conference: Meeting Abstract (28)
- Patent (25)
- Other (10)
- Report (9)
- Conference Poster (6)
Keywords
- Biosensor (25)
- Finite-Elemente-Methode (12)
- Einspielen <Werkstoff> (10)
- CAD (8)
- civil engineering (8)
- Bauingenieurwesen (7)
- Blitzschutz (6)
- FEM (6)
- Gamification (6)
- Limit analysis (6)
We present an automated pipeline for the generation of synthetic datasets for six-dimension (6D) object pose estimation. Therefore, a completely automated generation process based on predefined settings is developed, which enables the user to create large datasets with a minimum of interaction and which is feasible for applications with a high object variance. The pipeline is based on the Unreal 4 (UE4) game engine and provides a high variation for domain randomization, such as object appearance, ambient lighting, camera-object transformation and distractor density. In addition to the object pose and bounding box, the metadata includes all randomization parameters, which enables further studies on randomization parameter tuning. The developed workflow is adaptable to other 3D objects and UE4 environments. An exemplary dataset is provided including five objects of the Yale-CMU-Berkeley (YCB) object set. The datasets consist of 6 million subsegments using 97 rendering locations in 12 different UE4 environments. Each dataset subsegment includes one RGB image, one depth image and one class segmentation image at pixel-level.
The eVTOL industry is a rapidly growing mass market expected to start in 2024. eVTOL compete, caused by their predicted missions, with ground-based transportation modes, including mainly passenger cars. Therefore, the automotive and classical aircraft design process is reviewed and compared to highlight advantages for eVTOL development. A special focus is on ergonomic comfort and safety. The need for further investigation of eVTOL’s crashworthiness is outlined by, first, specifying the relevance of passive safety via accident statistics and customer perception analysis; second, comparing the current state of regulation and certification; and third, discussing the advantages of integral safety and applying the automotive safety approach for eVTOL development. Integral safety links active and passive safety, while the automotive safety approach means implementing standardized mandatory full-vehicle crash tests for future eVTOL. Subsequently, possible crash impact conditions are analyzed, and three full-vehicle crash load cases are presented.
Reliable methods for automatic readability assessment have the potential to impact a variety of fields, ranging from machine translation to self-informed learning. Recently, large language models for the German language (such as GBERT and GPT-2-Wechsel) have become available, allowing to develop Deep Learning based approaches that promise to further improve automatic readability assessment. In this contribution, we studied the ability of ensembles of fine-tuned GBERT and GPT-2-Wechsel models to reliably predict the readability of German sentences. We combined these models with linguistic features and investigated the dependence of prediction performance on ensemble size and composition. Mixed ensembles of GBERT and GPT-2-Wechsel performed better than ensembles of the same size consisting of only GBERT or GPT-2-Wechsel models. Our models were evaluated in the GermEval 2022 Shared Task on Text Complexity Assessment on data of German sentences. On out-of-sample data, our best ensemble achieved a root mean squared error of 0:435.
A concept for the analysis and optimal design of reinforced concrete structures is described. It is based on a nonlinear optimization algorithm and a finite element program for linear and nonlinear analysis of structures. With the aim of minimal cost design a two stage optimization using efficient gradient algorithm is developed. The optimization problems on global (structural) and local (crosssectional) level are formulated. A parallelization concept for solving the two stage optimization problem in minimal time is presented. Examples are included to illustrate the practical use and the effectively of the parallelization in the area of engineering design.
In collaborative research projects, both researchers and practitioners work together solving business-critical challenges. These projects often deal with ETL processes, in which humans extract information from non-machine-readable documents by hand. AI-based machine learning models can help to solve this problem.
Since machine learning approaches are not deterministic, their quality of output may decrease over time. This fact leads to an overall quality loss of the application which embeds machine learning models. Hence, the software qualities in development and production may differ.
Machine learning models are black boxes. That makes practitioners skeptical and increases the inhibition threshold for early productive use of research prototypes. Continuous monitoring of software quality in production offers an early response capability on quality loss and encourages the use of machine learning approaches. Furthermore, experts have to ensure that they integrate possible new inputs into the model training as quickly as possible.
In this paper, we introduce an architecture pattern with a reference implementation that extends the concept of Metrics Driven Research Collaboration with an automated software quality monitoring in productive use and a possibility to auto-generate new test data coming from processed documents in production.
Through automated monitoring of the software quality and auto-generated test data, this approach ensures that the software quality meets and keeps requested thresholds in productive use, even during further continuous deployment and changing input data.
Reliable automation of the labor-intensive manual task of scoring animal sleep can facilitate the analysis of long-term sleep studies. In recent years, deep-learning-based systems, which learn optimal features from the data, increased scoring accuracies for the classical sleep stages of Wake, REM, and Non-REM. Meanwhile, it has been recognized that the statistics of transitional stages such as pre-REM, found between Non-REM and REM, may hold additional insight into the physiology of sleep and are now under vivid investigation. We propose a classification system based on a simple neural network architecture that scores the classical stages as well as pre-REM sleep in mice. When restricted to the classical stages, the optimized network showed state-of-the-art classification performance with an out-of-sample F1 score of 0.95 in male C57BL/6J mice. When unrestricted, the network showed lower F1 scores on pre-REM (0.5) compared to the classical stages. The result is comparable to previous attempts to score transitional stages in other species such as transition sleep in rats or N1 sleep in humans. Nevertheless, we observed that the sequence of predictions including pre-REM typically transitioned from Non-REM to REM reflecting sleep dynamics observed by human scorers. Our findings provide further evidence for the difficulty of scoring transitional sleep stages, likely because such stages of sleep are under-represented in typical data sets or show large inter-scorer variability. We further provide our source code and an online platform to run predictions with our trained network.
Wind-induced operational variability is one of the major challenges for structural health monitoring of slender engineering structures like aircraft wings or wind turbine blades. Damage sensitive features often show an even bigger sensitivity to operational variability. In this study a composite cantilever was subjected to multiple mass configurations, velocities and angles of attack in a controlled wind tunnel environment. A small-scale impact damage was introduced to the specimen and the structural response measurements were repeated. The proposed damage detection methodology is based on automated operational modal analysis. A novel baseline preparation procedure is described that reduces the amount of user interaction to the provision of a single consistency threshold. The procedure starts with an indeterminate number of operational modal analysis identifications from a large number of datasets and returns a complete baseline matrix of natural frequencies and damping ratios that is suitable for subsequent anomaly detection. Mahalanobis distance-based anomaly detection is then applied to successfully detect the damage under varying severities of operational variability and with various degrees of knowledge about the present operational conditions. The damage detection capabilities of the proposed methodology were found to be excellent under varying velocities and angles of attack. Damage detection was less successful under joint mass and wind variability but could be significantly improved through the provision of the currently encountered operational conditions.
An approach to automatically generate a dynamic energy simulation model in Modelica for a single existing building is presented. It aims at collecting data about the status quo in the preparation of energy retrofits with low effort and costs. The proposed method starts from a polygon model of the outer building envelope obtained from photogrammetrically generated point clouds. The open-source tools TEASER and AixLib are used for data enrichment and model generation. A case study was conducted on a single-family house. The resulting model can accurately reproduce the internal air temperatures during synthetical heating up and cooling down. Modelled and measured whole building heat transfer coefficients (HTC) agree within a 12% range. A sensitivity analysis emphasises the importance of accurate window characterisations and justifies the use of a very simplified interior geometry. Uncertainties arising from the use of archetype U-values are estimated by comparing different typologies, with best- and worst-case estimates showing differences in pre-retrofit heat demand of about ±20% to the average; however, as the assumptions made are permitted by some national standards, the method is already close to practical applicability and opens up a path to quickly estimate possible financial and energy savings after refurbishment.
Combined with the use of renewable energy sources for
its production, Hydrogen represents a possible alternative gas
turbine fuel for future low emission power generation. Due to
its different physical properties compared to other fuels such
as natural gas, well established gas turbine combustion
systems cannot be directly applied for Dry Low NOx (DLN)
Hydrogen combustion. This makes the development of new
combustion technologies an essential and challenging task
for the future of hydrogen fueled gas turbines.
The newly developed and successfully tested “DLN
Micromix” combustion technology offers a great potential to
burn hydrogen in gas turbines at very low NOx emissions.
Aiming to further develop an existing burner design in terms
of increased energy density, a redesign is required in order to
stabilise the flames at higher mass flows and to maintain low
emission levels.
For this purpose, a systematic design exploration has
been carried out with the support of CFD and optimisation
tools to identify the interactions of geometrical and design
parameters on the combustor performance. Aerodynamic
effects as well as flame and emission formation are observed
and understood time- and cost-efficiently. Correlations
between single geometric values, the pressure drop of the
burner and NOx production have been identified as a result.
This numeric methodology helps to reduce the effort of
manufacturing and testing to few designs for single
validation campaigns, in order to confirm the flame stability
and NOx emissions in a wider operating condition field.
Having well-defined control strategies for fuel cells, that can efficiently detect errors and take corrective action is critically important for safety in all applications, and especially so in aviation. The algorithms not only ensure operator safety by monitoring the fuel cell and connected components, but also contribute to extending the health of the fuel cell, its durability and safe operation over its lifetime. While sensors are used to provide peripheral data surrounding the fuel cell, the internal states of the fuel cell cannot be directly measured. To overcome this restriction, Kalman Filter has been implemented as an internal state observer.
Other safety conditions are evaluated using real-time data from every connected sensor and corrective actions automatically take place to ensure safety. The algorithms discussed in this paper have been validated thorough Model-in-the-Loop (MiL) tests as well as practical validation at a dedicated test bench.