Refine
Year of publication
Institute
- ECSM European Center for Sustainable Mobility (113) (remove)
Has Fulltext
- no (113) (remove)
Document Type
- Conference Proceeding (59)
- Article (35)
- Part of a Book (7)
- Patent (3)
- Report (3)
- Conference: Meeting Abstract (2)
- Conference Poster (2)
- Book (1)
- Doctoral Thesis (1)
Keywords
- Energy storage (4)
- Power plants (4)
- Associated liquids (3)
- Concentrated solar power (3)
- Hybrid energy system (3)
- Electricity generation (2)
- Smart Building (2)
- Solar thermal technologies (2)
- UAV (2)
- electro mobility (2)
In this paper we investigate the use of deep neural networks for 3D object detection in uncommon, unstructured environments such as in an open-pit mine. While neural nets are frequently used for object detection in regular autonomous driving applications, more unusual driving scenarios aside street traffic pose additional challenges. For one, the collection of appropriate data sets to train the networks is an issue. For another, testing the performance of trained networks often requires tailored integration with the particular domain as well. While there exist different solutions for these problems in regular autonomous driving, there are only very few approaches that work for special domains just as well. We address both the challenges above in this work. First, we discuss two possible ways of acquiring data for training and evaluation. That is, we evaluate a semi-automated annotation of recorded LIDAR data and we examine synthetic data generation. Using these datasets we train and test different deep neural network for the task of object detection. Second, we propose a possible integration of a ROS2 detector module for an autonomous driving platform. Finally, we present the performance of three state-of-the-art deep neural networks in the domain of 3D object detection on a synthetic dataset and a smaller one containing a characteristic object from an open-pit mine.
The recovery of waste heat requires heat exchangers to extract it from a liquid or gaseous medium into another working medium, a refrigerant. In Organic Rankine Cycles (ORC) on Combustion Engines there are two major heat sources, the exhaust gas and the water/glycol fluid from the engine’s cooling circuit. A heat exchanger design must be adapted to the different requirements and conditions resulting from the heat sources, fluids, system configurations, geometric restrictions, and etcetera. The Stacked Shell Cooler (SSC) is a new and very specific design of a plate heat exchanger, created by AKG, which allows with a maximum degree of freedom the optimization of heat exchange rate and the reduction of the related pressure drop. This optimization in heat exchanger design for ORC systems is even more important, because it reduces the energy consumption of the system and therefore maximizes the increase in overall efficiency of the engine.
The paper presents an aerodynamic investigation of 70 different streamlined bodies with fineness ratios ranging from 2 to 10. The bodies are chosen to idealize both unmanned and small manned aircraft fuselages and feature cross-sectional shapes that vary from circular to quadratic. The study focuses on friction and pressure drag in dependency of the individual body’s fineness ratio and cross section. The drag forces are normalized with the respective body’s wetted area to comply with an empirical drag estimation procedure. Although the friction drag coefficient then stays rather constant for all bodies, their pressure drag coefficients decrease with an increase in fineness ratio. Referring the pressure drag coefficient to the bodies’ cross-sectional areas shows a distinct pressure drag minimum at a fineness ratio of about three. The pressure drag of bodies with a quadratic cross section is generally higher than for bodies of revolution. The results are used to derive an improved form factor that can be employed in a classic empirical drag estimation method. The improved formulation takes both the fineness ratio and cross-sectional shape into account. It shows superior accuracy in estimating streamlined body drag when compared with experimental data and other form factor formulations of the literature.
The results of a statistical investigation of 42 fixed-wing, small to medium sized (20 kg−1000 kg) reconnaissance unmanned air vehicles (UAVs) are presented. Regression analyses are used to identify correlations of the most relevant geometry dimensions with the UAV’s maximum take-off mass. The findings allow an empirical based geometry-build up for a complete unmanned aircraft by referring to its take-off mass only. This provides a bridge between very early design stages (initial sizing) and the later determination of shapes and dimensions. The correlations might be integrated into a UAV sizing environment and allow designers to implement more sophisticated drag and weight estimation methods in this process. Additional information on correlation factors for a rough drag estimation methodology indicate how this technique can significantly enhance the accuracy of early design iterations.
A hybrid-electric propulsion system combines the advantages of fuel-based systems and battery powered systems and offers new design freedom. To take full advantage of this technology, aircraft designers must be aware of its key differences, compared to conventional, carbon-fuel based, propulsion systems. This paper gives an overview of the challenges and potential benefits associated with the design of aircraft that use hybrid-electric propulsion systems. It offers an introduction of the most popular hybrid-electric propulsion architectures and critically assess them against the conventional and fully electric propulsion configurations. The effects on operational aspects and design aspects are covered. Special consideration is given to the application of hybrid-electric propulsion technology to both unmanned and vertical take-off and landing aircraft. The authors conclude that electric propulsion technology has the potential to revolutionize aircraft design. However, new and innovative methods must be researched, to realize the full benefit of the technology.
The predictive control of commercial vehicle energy management systems, such as vehicle thermal management or waste heat recovery (WHR) systems, are discussed on the basis of information sources from the field of environment recognition and in combination with the determination of the vehicle system condition.
In this article, a mathematical method for predicting the exhaust gas mass flow and the exhaust gas temperature is presented based on driving data of a heavy-duty vehicle. The prediction refers to the conditions of the exhaust gas at the inlet of the exhaust gas recirculation (EGR) cooler and at the outlet of the exhaust gas aftertreatment system (EAT). The heavy-duty vehicle was operated on the motorway to investigate the characteristic operational profile. In addition to the use of road gradient profile data, an evaluation of the continuously recorded distance signal, which represents the distance between the test vehicle and the road user ahead, is included in the prediction model. Using a Fourier analysis, the trajectory of the vehicle speed is determined for a defined prediction horizon.
To verify the method, a holistic simulation model consisting of several hierarchically structured submodels has been developed. A map-based submodel of a combustion engine is used to determine the EGR and EAT exhaust gas mass flows and exhaust gas temperature profiles. All simulation results are validated on the basis of the recorded vehicle and environmental data. Deviations from the predicted values are analyzed and discussed.
It is investigated whether a nonrotating lifting fan remaining uncovered during cruise flight, as opposed to being covered by a shutter system, can be realized with limited additional drag and loss of lift during cruise flight. A wind-tunnel study of a wing-embedded lifting fan has been conducted at the Side Wind Test Facility Göttingen of DLR, German Aerospace Center in Göttingen using force, pressure, and stereoscopic particle image velocimetry techniques. The study showed that a step on the lower side of the wing in front of the lifting fan duct increases the lift-to-drag ratio of the whole model by up to 25% for all positive angles of attack. Different sizes and inclinations of the step had limited influence on the surface pressure distribution. The data indicate that these parameters can be optimized to maximize the lift-to-drag ratio. A doubling of the curvature radius of the lifting fan duct inlet lip on the upper side of the wing affected the lift-to-drag ratio by less than 1%. The lifting fan duct inlet curvature can therefore be optimized to maximize the vertical fan thrust of the rotating lifting fan during hovering without affecting the cruise flight performance with a nonrotating fan.
To meet the challenges of manufacturing smart products, the manufacturing plants have been radically changed to become smart factories underpinned by industry 4.0 technologies. The transformation is assisted by employment of machine learning techniques that can deal with modeling both big or limited data. This manuscript reviews these concepts and present a case study that demonstrates the use of a novel intelligent hybrid algorithms for Industry 4.0 applications with limited data. In particular, an intelligent algorithm is proposed for robust data modeling of nonlinear systems based on input-output data. In our approach, a novel hybrid data-driven combining the Group-Method of Data-Handling and Singular-Value Decomposition is adapted to find an offline deterministic model combined with Pareto multi-objective optimization to overcome the overfitting issue. An Unscented-Kalman-Filter is also incorporated to update the coefficient of the deterministic model and increase its robustness against data uncertainties. The effectiveness of the proposed method is examined on a set of real industrial measurements.
To meet the challenges of manufacturing smart products, the manufacturing plants have been radically changed to become smart factories underpinned by industry 4.0 technologies. The transformation is assisted by employment of machine learning techniques that can deal with modeling both big or limited data. This manuscript reviews these concepts and present a case study that demonstrates the use of a novel intelligent hybrid algorithms for Industry 4.0 applications with limited data. In particular, an intelligent algorithm is proposed for robust data modeling of nonlinear systems based on input-output data. In our approach, a novel hybrid data-driven combining the Group-Method of Data-Handling and Singular-Value Decomposition is adapted to find an offline deterministic model combined with Pareto multi-objective optimization to overcome the overfitting issue. An Unscented-Kalman-Filter is also incorporated to update the coefficient of the deterministic model and increase its robustness against data uncertainties. The effectiveness of the proposed method is examined on a set of real industrial measurements.
With the many achievements of Machine Learning in the past years, it is likely that the sub-area of Deep Learning will continue to deliver major technological breakthroughs [1]. In order to achieve best results, it is important to know the various different Deep Learning frameworks and their respective properties. This paper provides a comparative overview of some of the most popular frameworks. First, the comparison methods and criteria are introduced and described with a focus on computer vision applications: Features and Uses are examined by evaluating papers and articles, Adoption and Popularity is determined by analyzing a data science study. Then, the frameworks TensorFlow, Keras, PyTorch and Caffe are compared based on the previously described criteria to highlight properties and differences. Advantages and disadvantages are compared, enabling researchers and developers to choose a framework according to their specific needs.
Neue Perspektiven für die Bahn in der Produktions- und Distributionslogistik durch Prozessautomation
(2020)
The number of case studies focusing on hybrid-electric aircraft is steadily increasing, since these configurations are thought to lead to lower operating costs and environmental impact than traditional aircraft. However, due to the lack of reference data of actual hybrid-electric aircraft, in most cases, the design tools and results are difficult to validate. In this paper, two independently developed approaches for hybrid-electric conceptual aircraft design are compared. An existing 19-seat commuter aircraft is selected as the conventional baseline, and both design tools are used to size that aircraft. The aircraft is then re-sized under consideration of hybrid-electric propulsion technology. This is performed for parallel, serial, and fully-electric powertrain architectures. Finally, sensitivity studies are conducted to assess the validity of the basic assumptions and approaches regarding the design of hybrid-electric aircraft. Both methods are found to predict the maximum take-off mass (MTOM) of the reference aircraft with less than 4% error. The MTOM and payload-range energy efficiency of various (hybrid-) electric configurations are predicted with a maximum difference of approximately 2% and 5%, respectively. The results of this study confirm a correct formulation and implementation of the two design methods, and the data obtained can be used by researchers to benchmark and validate their design tools.