20005 research outputs found
Sort by
Design challenges for high-speed digital links for automotive applications
No full textThis paper outlines the state of the art and future directions for high-speed links in modern automotive applications. The technological constraints of signal integrity (SI) are discussed in the context of challenges of the mechanical and atmospheric environment. The requirements are illustrated by example of automotive Ethernet and its use to establish communication between electronic control units (ECUs). The overview covers design elements and parts inside the ECU enclosure as well as the link segment in between control units. Moreover, the compound system is subject to operational testing and verification of compliance with applicable standards
Propeller-ice interaction experiments on a scaled S.A. Agulhas II propeller
No full textBetween 2011 and 2020, over 16% more incidents of machinery damage and failure were reported in the waters within the Arctic Circle compared to the rest. In addition, Allianz (AGCS) also reported that machinery damages and failures account for more than 45% of all incidents in Arctic waters, which are caused by the exposure of the propulsion system to ice contact. A better understanding of the propeller-ice interaction process would allow a more efficient propeller design and a safer operation in ice-covered waters. However, full-scale measurements of propeller-ice loads are unique. In addition, propeller-ice loads determined in model tanks are hard to transfer to full scale due to the model ice properties and the large scaling of the propeller. Moreover, holistic propeller-ice simulations have yet to be created, capturing the complex ice failure and representing realistic loads. This paper presents propeller-ice interaction experiments on the propeller of S.A. Agulhas II on a scale of 1:3. The experiments focus on contact milling and impact loads. The test matrix consists of different interaction velocities up to 4 m/s using a drop-tower, cylindrical ice specimen with diameters up to 200 mm, and three different contact conditions from milling loads to impact loads. The specimens consist of granular freshwater ice with high uniaxial strength, and the reports of the measured interaction force and the strains measured on the propeller should provide valuable information for future simulations of the propeller-ice interaction process
Constitutive Kolmogorov–Arnold Networks (CKANs): Combining accuracy and interpretability in data-driven material modeling
No full textHybrid constitutive modeling integrates two complementary approaches for describing and predicting a material's mechanical behavior: purely data-driven black-box methods and physically constrained, theory-based models. While black-box methods offer high accuracy, they often lack interpretability and extrapolability. Conversely, physics-based models provide theoretical insight and generalizability but may not capture complex behaviors with the same accuracy. Traditionally, hybrid modeling has required a trade-off between these aspects. In this paper, we show how recent advances in symbolic machine learning — specifically Kolmogorov–Arnold Networks (KANs) — help to overcome this limitation. We introduce Constitutive Kolmogorov–Arnold Networks (CKANs) as a new class of hybrid constitutive models. By incorporating a post-processing symbolification step, CKANs combine the predictive accuracy of data-driven models with the interpretability and extrapolation capabilities of symbolic expressions, bridging the gap between machine learning and physical modeling
Modeling the impact of bursty traffic on end-to-end delays in 6TiSCH wireless sensor networks
No full textNew applications emerge for Wireless Sensor Networks (WSNs), e.g., monitoring and surveillance, producing bursty traffic which greatly impacts communication network performance. The latter can be evaluated through simulations or experiments, both of which are more time-consuming and costly than analytical models. Reliable communication in WSNs is achievable through IEEE 802.15.4 Timeslotted Channel Hopping (TSCH). Accurately evaluating its performance under bursty traffic is critical to plan future networks. This letter proposes a framework for modeling the impact of bursty traffic on end-to-end delays in TSCH networks, using queuing theory and Markov chains. Simulations in OMNeT++ validate framework accuracy with <1% normalized Root Mean Squared Error (RMSE)
Elimination of dispersion ripples in a waveguide cavity using two slightly dispersion-shifted Bragg gratings
No full textChip-integrated chirped or apodized Bragg gratings (CBGs) are key components for on-chip dispersion compensation. They are increasingly sought after for applications such as dispersion compensation for telecommunications, integrated chirped pulse amplifiers [1], mode-locked lasers [2], and ultrashort pulse soliton generation [3] on a silicon chip. However, maintaining flat group-delay dispersion (GDD) over a large bandwidth without strong ripples is challenging. These ripples typically arise from imperfect apodization profiles, and reflections at the grating boundaries, which can introduce phase inconsistencies and disrupt the desired dispersion profile
Predicting route-specific energy demand under realistic environmental behaviour
No full textTo meet current and forthcoming international regulations, it is essential to reduce emissions and increase the efficiency of ships. A prevailing approach is to install hybrid drive systems to optimize the engine load. Evaluating the actual efficiency gains of a battery-supported ship compared to a conventional system requires detailed knowledge of its operational proőle. Battery support proves advantageous primarily in applications with highly variable power demands. However, the data basis is often limited to statistical power distributions rather than continuous data. Since the state of charge of the battery is time-dependent, a continuous time series of the power demand is crucial for accurate sizing of system components. Additionally, the operational proőle is inŕuenced by environmental factors that vary statistically and regionally. To capture realistic power demand patterns, the environmental forces must also be incorporated in time domain. Therefore, this paper presents a manoeuvering simulation for predicting power demand along a ship route under realistically changing environmental conditions. The calculations utilize a fast, in-house, force-based manoeuvering approach that accounts for environmental factors such as seaway, wind, currents, and shallow water effects. Arbitrary ship routes are deőned by waypoints on a map, which are automatically navigated using the line-of-sight concept. Environmental data is sourced from the ERA5 database, providing information at speciőc geographical grid points. A spatial and temporal interpolation process generates a realistic continuous proőle of these ambient parameters based on the ship’s current position along its route. The simulated results align strongly with measured data, conőrming the model’s validity
Effective nudging as a booster for agile collaboration in innovation
No full textAgile methods such as Scrum enable structured and iterative work processes in product development. However, cognitive biases often lead to planning inaccuracies, particularly during sprint planning. This study investigates whether nudges can improve planning quality in agile teams. In cooperation with an industrial partner developing mechatronic safety systems, tailored nudges were designed and implemented in two Scrum teams. Using a before-and-after design, the effects on story granularity and sprint completion rates were analysed. The results show that, especially in less mature teams, nudges led to a reduction in average story size and an increase in sprint fulfilment. Rather than causing abrupt changes, nudges supported ongoing learning processes and fostered more consistent planning. The study provides initial empirical evidence that nudging can serve as a practical tool to enhance decision-making quality in agile development environments
An all-pairs shortest path algorithm for bipartite graphs
No full textAbstract Bipartite graphs are widely used for modeling of complex structures in biology, engineering, and computer science. The search for shortest paths in such structures is a highly demanded procedure that requires optimization. This paper presents a variant of the all-pairs shortest path algorithm for bipartite graphs. The method is based on the distance matrix product and improves the general algorithm by exploiting the graph topology. The space complexity is reduced by a factor of at least four and the time complexity decreased by almost an order of magnitude when compared with the basic APSP algorithm
Treatment of singular integrals for boundary element methods in hydro- and aerodynamics - a short review
No full textBoundary element methods are widely employed in engineering and research to solve partial differential equations in the form of boundary integral equations numerically. In hydro- and aerodynamics, these methods provide fast solution to potential flow problems, especially useful in early design phases of ships or aircraft, e.g. for the computation of ship motion and added wave resistance or of aircraft loads and acoustic analyses. However, the fundamental solution of the underlying boundary integral equations as well as their derivatives are characterized by an increasingly singular nature, so the assembly of the required boundary integral operators is non-trivial. Therefore, suitable treatment of the singular integrals is crucial for the boundary element method, since it requires the evaluation of the singular boundary integrals in the near- and self-influence regimes. In this paper, we classify the different methods for singular integration, detail the theory behind these techniques, give examples of existing approaches and sort them according to the presented classification. The present review for singular integration methods for Laplace boundary element methods aims to give an overview of existing frameworks and the related theory, intended as a starting point for choosing appropriate methods by considering the advantageous characteristics or identifying fields of further research
A coordinated approach to apply value stream analysis to the one-of-a-kind shipbuilding supply chain
No full textIn one-of-a-kind shipbuilding, shipyards rely on highly diversified supply chains involving multiple suppliers and subcontractors. The high complexity of the processes and the unique nature of the ships present challenges for the planning and coordination of the assembly at the shipyard and for production and logistics at the suppliers. The current organization leads to long delivery times, and high inventory costs. In addition, productivity losses occur due to extra work, e.g., searching for and modifying pre-manufactured components or delayed material supply at the assembly. Synchronizing the provision of materials from suppliers with the actual assembly offers promising potential for improvement. In one-of-a-kind production, Just-in-Sequence delivery is applicable only to a limited extent due to a lack of data and insufficient communication. The systematic use of digital assistance systems in the shipbuilding supply chain offers the potential to establish a consistent database and to provide user-oriented real-time information. As a first step, this paper performs an exhaustive analysis of the shipyard and supply chain processes regarding material and information flow, and the key figures from production planning and control. For this purpose, a value stream analysis adapted to the one-of-a-kind shipbuilding process was carried out at a shipyard and three of its suppliers.Bundesministerium für Wirtschaft und Klimaschut