Refine
Institute
- Fakultät Informatik (597)
- Fakultät Elektro- und Informationstechnik (463)
- THI Business School (412)
Version
- published (858)
"Kirjasto heißt Bibliothek"
(2019)
Recent developments in maintenance modelling fuelled by data-based approaches such as machine learning (ML), have enabled a broad range of applications. In the automotive industry, ensuring the functional safety over the product life cycle while limiting maintenance costs has become a major challenge. One crucial approach to achieve this, is predictive maintenance (PdM). Since modern vehicles come with an enormous amount of operating data, ML is an ideal candidate for PdM. While PdM and ML for automotive systems have both been covered in numerous review papers, there is no current survey on ML-based PdM for automotive systems. The number of publications in this field is increasing — underlining the need for such a survey. Consequently, we survey and categorize papers and analyse them from an application and ML perspective. Following that, we identify open challenges and discuss possible research directions. We conclude that (a) publicly available data would lead to a boost in research activities, (b) the majority of papers rely on supervised methods requiring labelled data, (c) combining multiple data sources can improve accuracies, (d) the use of deep learning methods will further increase but requires efficient and interpretable methods and the availability of large amounts of (labelled) data.
A Dynamic Multinode Model for Component-Oriented Thermal Analysis of Flat-Plate Solar Collectors
(2015)
A mathematical model of a flat-plate solar collector was developed on the basis of the physical principles of optics and heat transfer in order to determine collector’s component temperatures as well as collector efficiency. In contrast to many available models, the targeted use of this dynamic model is the detailed, theoretical investigation of the thermal behaviour of newly developed or adjusted collector designs on component level, for example, absorber, casing, or transparent cover. The defined model is based on a multinode network (absorber, fluid, glazing, and backside insulation) containing the relevant physical equations to transfer the energy. The heat transfer network covers heat conduction, convection, and radiation. Furthermore, the collector optics is defined for the plane glazing and the absorber surface and also considers interactions between them. The model enables the variation of physical properties considering the geometric parameters and materials. Finally, the model was validated using measurement data and existing efficiency curve models. Both comparisons proved high accuracy of the developed model with deviation of up to 3% in collector efficiency and 1 K in component temperatures.
Today the concept of smart cities is discussed in scientific society and politics. A core function of smart cities is transportation. This paper gives a short overview on the concepts for Intelligent Transport Systems (ITS) for smart cities and proposes a framework for the design of an autonomic transportation system that provides personalized mobility services to its users in a smart city setting. The transportation poses extreme environment for ICT systems due to fast moving vehicles and users, requiring real-time acquisition and high performance processing of large scale data, and rapidly changing communication networks topologies and node densities. The aim of this paper is to propose a framework that will act as a reference for the design of future transportation systems that are able to cope with the ever rising system complexities and users’ demands. Therefore, a backbone system providing information at different levels was designed following the principles of a corporative ICT that were proposed in [1]. The framework fulfills the main requirement providing suitable information about the local decision engines in vehicles and infrastructure interacting in smart cities traffic systems.
The new equity regulations of Basel III are more restrictive. Therefore the research objective of this article is to show the impact of equity regulations and to provide a proposal for professional preparation for the rating procedure for credit seeking companies. The research method applied is to give an overview of the relevant equity regulations of Basel III. From this research, we can derive that risk is the most important issue to be considered. In consequence, the outcome is a proposal for a contemporary rating approach for companies, containing an appropriate way how to control risk and perform in an excellent way. We can conclude, that the new equity regulations will not only have an impact on banks, but also on credit taking companies. Especially small and medium-sized companies will be affected. Furthermore in future banks intend to get an exhaustive insight into companies. The risks identified in the business will serve as a measure to calculate a risk adequate interest rate, depending on the rating class. Therefore it won’t be sufficient anymore to use only backward looking statements from the balance sheet or profit and loss statement. In this article we propose an evidence based approach for executing a professional and sustainable rating. Supporting this, we provide six dimensions for a contemporary rating approach.
A classification system for the segmentation of driving maneuvers and its validation in autonomous parking using a small-scale vehicle are presented in this work. The classifiers are designed to detect points that are crucial for the path-planning task, thus enabling the implementation of efficient autonomous parking maneuvers. The training data set is generated by simulations using appropriate vehicle–dynamics models and the resulting classifiers are validated with the small-scale autonomous vehicle. To achieve both a high classification performance and a classification system that can be implemented on a microcontroller with limited computational resources, a two-stage design process is applied. In a first step an ensemble classifier, the Random Forest (RF) algorithm, is constructed and based on the RF-kernel a General Radial Basis Function (GRBF) classifier is generated. The GRBF-classifier is integrated into the small-scale autonomous vehicle leading to excellent performance in parallel-, cross- and oblique- parking maneuvers. The work shows that segmentation using classifies and open-loop control are an efficient approach in autonomous driving for the implementation of driving maneuvers.
In the automotive industry, sensors and sensor systems are one of the most important components in upcoming challenges like highly automated and autonomous driving. Forward-looking sensors (radar, lidar and cameras) have the technical capability to already provide important (pre-)crash information, such as the position of contact, relative crash velocity and overlap (width of contact) before the crash occurs. Future safety systems can improve crash mitigation with sophisticated vehicle safety strategies based on this information. One such strategy is an early activation of restraint systems compared with conventional passive safety systems. These integrated safety systems consist of a combination of predictive forward-looking sensors and occupant restraint systems (airbags, belt tensioners, etc.) to provide the best occupant safety in inevitable crash situations. The activation of the restraint systems is the most critical decision process and requires a very robust validation system to avoid false activation. Hence, the information provided by the forward-looking sensor needs to be highly reliable. A validation sensor is required to check the plausibility of crucial information from forward-looking sensors used in integrated safety systems for safe automated and autonomous driving.
This work presents a CFRP-based (carbon-fiber-reinforced plastic) validation sensor working on the principle of change in electrical resistance when a contact occurs. This sensor detects the first contact, gives information on impact position (where the contact occurs) and provides information on the overlap. The aim is to activate the vehicle restraint systems at near T0 (time of first contact). Prototypes of the sensor were manufactured in house and manually and were evaluated. At first, the sensor and its working principle were tested with a pendulum apparatus. In the next stage, the sensor was tested in a real crash test. The comparison of the signals from the CFRP-based sensor with presently used crash sensors in the vehicle highlights its advantages. The crash event can be identified at 0.1 ms after the initial contact. The sensor also provides information on impact position at 1.2 ms and enables a validation of the overlap development. Finally, a possible algorithm for the vehicle safety system using forward-looking sensors with a validation sensor is described.
Lithium-ion traction battery systems of hybrid and electric vehicles must have a high level of durability and reliability like all other components and systems of a vehicle. Battery systems get heated while in the application. To ensure the desired life span and performance, most systems are equipped with a cooling system. The changing environmental condition in daily use may cause water condensation in the housing of the battery system. In this study, three system designs were investigated, to compare different solutions to deal with pressure differences and condensation: (1) a sealed battery system, (2) an open system and (3) a battery system equipped with a pressure compensation element (PCE). These three designs were tested under two conditions: (a) in normal operation and (b) in a maximum humidity scenario. The amount of the condensation in the housing was determined through a change in relative humidity of air inside the housing. Through PCE and available spacing of the housing, moisture entered into the housing during the cooling process. While applying the test scenarios, the gradient-based drift of the moisture into the housing contributed maximum towards the condensation. Condensation occurred on the internal surface for all the three design variants.
Wir stellen einer festen Zinsbindung von 10 Jahren bei Wohnungsbaukrediten eine monatliche an einem Kurzfristzins angepasste Zinsbelastung bei unterschiedlichen Rückzahlungsfristen gegenüber. Dafür werten wir Monatsdaten seit den 1950er Jahren für Deutschland aus. Damit greifen wir mehrere Zinszyklen, Zinserhöhungs- und Zinssenkungsphasen ab. Im Ergebnis finden wir, dass in der überwiegenden Mehrzahl der Fälle kurzfristig variable Zinsen eine geringere Zinsbelastung nach sich ziehen. Und diese weisen auch noch geringere Schwankungen auf.
Cut in greenhouse gas emissions, increment of energy from renewables and improvement in energy efficiency represent the three key targets for future energy systems. Among the available bioenergy technologies, biogas production via biodegradation and anaerobic digestion is a widely applied approach, not only to produce biofuels but also to manage industrial and domestic organic waste. Within the biogas production, a sufficient mixing of the organic mass is a crucial step to ensure high biogas yields by bacteria and enzymes. Measurements of the electric power consumption of biogas plants revealed that the electrical energy demand of the stirrer system has a high share of the total electricity consumption of a biogas plant. Investigations on real biogas digesters to optimize the mixing process are cost and time intensive. Therefore, laboratory prototypes and computational simulations represent promising alternatives to analyse and improve the efficiency of mixing systems. In this paper, a computational fluid dynamics (CFD) model is presented, which is applied to commercial stirring systems. The case of two propeller stirrers, located in diametrically opposite positions in a tank filled with ca. 1400 m3 of substrate is described in detail. For the simulation, the rheology of the fluid is adapted to a biomass with 12 wt % dry matter content and obeying the non-Newtonian generalized Ostwald-de Waele power law. The developed simulation procedure considers the rotation angle of each propeller and its height. A total of 441 mixing configurations are calculated and evaluated in terms of the technical benefit. The investigation reveals that locations of the rotors far away from the bottom and high rotational angles cause advantageous fluid dynamics.
How to measure flexibility
(2019)
Flexible power provision from biogas can significantly contribute to energy systems with high shares of renewables. However, the characteristics and demands for this flexibility are not clearly defined or measured. In this paper eight indicators are defined to shape “flexibility” and perform a downstream investigation of eight research projects focusing on flexible energy provision of biogas plants. The indicators are structured in three dimensions (1) velocity (ramps) by which the system can be modulated, (2) power range (bandwidth) and (3) duration for specific load conditions. Based on these indicators bottlenecks for the flexibility potential were identified. One crucial result shows that short-term flexibility of biogas plants is mainly driven by properties of the combined heat and power unit (velocity and bandwidth). The long-term flexibility depends mainly on gas storage, mode of operation and ability for modulation of the target gas production.
Global Navigation Satellite Systems (GNSS) are crucial for applications that demand very accurate positioning. Tensor-based time-delay estimation methods, such as CPD-GEVD, DoA/KRF, and SECSI, combined with the GPS3 L1C signal, are capable of, significantly, mitigating the positioning degradation caused by multipath components. However, even though these schemes require an estimated model order, they assume that the number of multipath components is constant. In GNSS applications, the number of multipath components is time-varying in dynamic scenarios. Thus, in this paper, we propose a tensor-based framework with model order selection and high accuracy factor decomposition for time-delay estimation in dynamic multipath scenarios. Our proposed approach exploits the estimates of the model order for each slice by grouping the data tensor slices into sub-tensors to provide high accuracy factor decomposition. We further enhance the proposed approach by incorporating the tensor-based Multiple Denoising (MuDe).
Contribution: Prior studies comparing the effectiveness of different laboratory learning modes do not allow one to draw a universally valid conclusion, as other influences are mixed with the learning modes. In order to contribute to the existing body of work and to add another piece to the puzzle, this article demonstrates an improved methodology to evaluate the effectiveness of computer-simulated laboratories in comparison to hands-on exercises using a battery basics practical course as a case study.
Background: Computer-simulated experiments are becoming increasingly popular for conducting laboratory exercises in higher education and vocational training institutions. To ensure the consistent quality of laboratory learning, an accurate comparison between the results of simulated experiments and practical hands-on experiments is required.
Intended Outcomes: In this article, the achievement of the following learning objectives were compared between the two laboratory modes: 1) comprehension of the most important parameters of battery cells and 2) knowledge on how these parameters can be determined using adequate experimental procedures.
Application Design: To avoid interference of factors other than laboratory mode on the learning, laboratory instructions and experimental interfaces ensured identical execution of the experiments in the compared modes. Using a counterbalanced methodology, the two laboratory modes alternated by the session, while the experimental procedures remained constant regardless of the respective modes.
Findings: Tests taken by the participants after conducting the laboratory experiments revealed that hands-on laboratories resulted in statistically significantly better student performance than simulated laboratories. This difference was even more pronounced for the participants that finished a vocational education and training program before the university studies.
Drowsy driving imposes a high safety risk. Current systems often use driving behavior parameters for driver drowsiness detection. The continuous driving automation reduces the availability of these parameters, therefore reducing the scope of such methods. Especially, techniques that include physiological measurements seem to be a promising alternative. However, in a dynamic environment such as driving, only non- or minimal intrusive methods are accepted, and vibrations from the roadbed could lead to degraded sensor technology. This work contributes to driver drowsiness detection with a machine learning approach applied solely to physiological data collected from a non-intrusive retrofittable system in the form of a wrist-worn wearable sensor. To check accuracy and feasibility, results are compared with reference data from a medical-grade ECG device. A user study with 30 participants in a high-fidelity driving simulator was conducted. Several machine learning algorithms for binary classification were applied in user-dependent and independent tests. Results provide evidence that the non-intrusive setting achieves a similar accuracy as compared to the medical-grade device, and high accuracies (>92%) could be achieved, especially in a user-dependent scenario. The proposed approach offers new possibilities for human–machine interaction in a car and especially for driver state monitoring in the field of automated driving.
An important part of the industry 4.0 concept is the horizontal and vertical integration of manufacturing systems.
Information exchange in traditional production environments happens through interfaces that are connections between strictly defined senders and receivers. This limits the possibility for changing and extending the manufacturing system. A possible approach to enable the information exchange between all system entities uniformly are information models. Such models are semantic descriptions of the available data. The creation of these models needs to follow the manufacturing process, but also requires certain standardization to improve efficiency. Another challenge is the actual technical integration of the information into a common address space.
This paper connects an approach for information modeling with a concept for dynamic aggregation. The approach is described with the help of a continuous example that uses OPC UA as a middleware technology.
Chemotherapy-induced polyneuropathy (CIPN), one of the most severe and incapacitating side effects of chemotherapeutic drugs, is a serious concern in breast cancer therapy leading to dose diminution, delay, or cessation. The reversibility of CIPN is of increasing importance since active chemotherapies prolong survival. Clinical assessment tools show that patients experiencing sensorimotor CIPN symptoms not only do they have to cope with loss in autonomy and life quality, but CIPN has become a key restricting factor in treatment. CIPN incidence poses a clinical challenge and has lacked established and efficient therapeutic options up to now. Complementary, non-opioid therapies are sought for both prevention and management of CIPN. In this perspective, we explore the potential that digital interventions have for sensorimotor CIPN rehabilitation in breast cancer patients. Our primary goal is to emphasize the benefits and impact that Virtual Reality (VR) avatars and Machine Learning have in combination in a digital intervention aiming at (1) assessing the complete kinematics of deficits through learning underlying patient sensorimotor parameters, and (2) parameterize a multimodal VR simulation to drive personalized deficit compensation. We support our perspective by evaluating sensorimotor effects of chemotherapy, the metrics to assess sensorimotor deficits, and relevant clinical studies. We subsequently analyse the neurological substrate of VR sensorimotor rehabilitation, with multisensory integration acting as a key element. Finally, we propose a closed-loop patient-centered design recommendation for CIPN sensorimotor rehabilitation. Our aim is to provoke the scientific community toward the development and use of such digital interventions for more efficient and targeted rehabilitation.
Background: The status of the data-driven management of cancer care as well as the challenges, opportunities, and recommendations aimed at accelerating the rate of progress in this field are topics of great interest. Two international workshops, one conducted in June 2019 in Cordoba, Spain, and one in October 2019 in Athens, Greece, were organized by four Horizon 2020 (H2020) European Union (EU)–funded projects: BOUNCE, CATCH ITN, DESIREE, and MyPal. The issues covered included patient engagement, knowledge and data-driven decision support systems, patient journey, rehabilitation, personalized diagnosis, trust, assessment of guidelines, and interoperability of information and communication technology (ICT) platforms. A series of recommendations was provided as the complex landscape of data-driven technical innovation in cancer care was portrayed.
Objective: This study aims to provide information on the current state of the art of technology and data-driven innovations for the management of cancer care through the work of four EU H2020–funded projects.
Methods: Two international workshops on ICT in the management of cancer care were held, and several topics were identified through discussion among the participants. A focus group was formulated after the second workshop, in which the status of technological and data-driven cancer management as well as the challenges, opportunities, and recommendations in this area were collected and analyzed.
Results: Technical and data-driven innovations provide promising tools for the management of cancer care. However, several challenges must be successfully addressed, such as patient engagement, interoperability of ICT-based systems, knowledge management, and trust. This paper analyzes these challenges, which can be opportunities for further research and practical implementation and can provide practical recommendations for future work.
Conclusions: Technology and data-driven innovations are becoming an integral part of cancer care management. In this process, specific challenges need to be addressed, such as increasing trust and engaging the whole stakeholder ecosystem, to fully benefit from these innovations.
Database management and monitoring is an inseparable part of any industry. A uniform scheme of monitoring relational databases without explicit user access to database servers is not much explored outside the database environment. In this paper, we present an information distribution scheme related to databases using Open Platform Communication Unified Architecture (OPC UA) servers to clients when multiple databases are involved in a factory. The aim is for external, but relevant clients, to be able to monitor this information mesh independent of explicit access to user schemas. A methodology to dispense data from, as well as check changes in databases using SQL queries and events is outlined and implemented using OPC UA servers. The structure can be used as a remote viewing application for multiple databases in one address space of an OPC UA server.