Technische Universität Darmstadt

About: Technische Universität Darmstadt is a education organization based out in Darmstadt, Germany. It is known for research contribution in the topics: Neutron & Finite element method. The organization has 17316 authors who have published 40619 publications receiving 937916 citations. The organization is also known as: Darmstadt University of Technology & University of Darmstadt.

Impact of a drop onto a wetted wall: description of crown formation and propagation

Abstract: The impact of a drop onto a liquid film with a relatively high impact velocity, leading to the formation of a crown-like ejection, is studied theoretically The motion of a kinematic discontinuity in the liquid film on the wall due to the drop impact, the formation of the upward jet at this kinematic discontinuity and its elevation are analysed Four main regions of the drop and film are considered: the perturbed liquid film on the wall inside the crown, the unperturbed liquid film on the wall outside the crown, the upward jet forming a crown, and the free rim bounding this jet The theory of Yarin & Weiss (1995) for the propagation of the kinematic discontinuity is generalized here for the case of arbitrary velocity vectors in the inner and outer liquid films on the wall Next, the mass, momentum balance and Bernoulli equations at the base of the crown are considered in order to obtain the velocity and the thickness of the jet on the wall Furthermore, the dynamic equations of motion of the crown are developed in the Lagrangian form An analytical solution for the crown shape is obtained in the asymptotic case of such high impact velocities that the surface tension and the viscosity effects can be neglected in comparison to inertial effects The edge of the crown is described by the motion of a rim, formed due to the surface tensionThree different cases of impact are considered: normal axisymmetric impact of a single drop, oblique impact of a single drop, and impact and interaction of two drops The theoretical predictions of the height of the crown in the axisymmetric case are compared with experiments The agreement is quite good in spite of the fact that no adjustable parameters are used

Athena MIMOS II Mössbauer spectrometer investigation

Abstract: [1] Mossbauer spectroscopy is a powerful tool for quantitative mineralogical analysis of Fe-bearing materials. The miniature Mossbauer spectrometer MIMOS II is a component of the Athena science payload launched to Mars in 2003 on both Mars Exploration Rover missions. The instrument has two major components: (1) a rover-based electronics board that contains power supplies, a dedicated central processing unit, memory, and associated support electronics and (2) a sensor head that is mounted at the end of the instrument deployment device (IDD) for placement of the instrument in physical contact with soil and rock. The velocity transducer operates at a nominal frequency of ∼25 Hz and is equipped with two 57Co/Rh Mossbauer sources. The reference source (∼5 mCi landed intensity), reference target (α-Fe2O3 plus α-Fe0), and PIN-diode detector are configured in transmission geometry and are internal to the instrument and used for its calibration. The analysis Mossbauer source (∼150 mCi landed intensity) irradiates Martian surface materials with a beam diameter of ∼1.4 cm. The backscatter radiation is measured by four PIN-diode detectors. Physical contact with surface materials is sensed with a switch-activated contact plate. The contact plate and reference target are instrumented with temperature sensors. Assuming ∼18% Fe for Martian surface materials, experiment time is 6–12 hours during the night for quality spectra (i.e., good counting statistics); 1–2 hours is sufficient to identify and quantify the most abundant Fe-bearing phases. Data stored internal to the instrument for selectable return to Earth include Mossbauer and pulse-height analysis spectra (512 and 256 channels, respectively) for each of the five detectors in up to 13 temperature intervals (65 Mossbauer spectra), engineering data for the velocity transducer, and temperature measurements. The total data volume is ∼150 kB. The mass and power consumption are ∼500 g (∼400 g for the sensor head) and ∼2 W, respectively. The scientific measurement objectives of the Mossbauer investigation are to obtain for rock, soil, and dust (1) the mineralogical identification of iron-bearing phases (e.g., oxides, silicates, sulfides, sulfates, and carbonates), (2) the quantitative measurement of the distribution of iron among these iron-bearing phases (e.g., the relative proportions of iron in olivine, pyroxenes, ilmenite, and magnetite in a basalt), (3) the quantitative measurement of the distribution of iron among its oxidation states (e.g., Fe2+, Fe3+, and Fe6+), and (4) the characterization of the size distribution of magnetic particles. Special geologic targets of the Mossbauer investigation are dust collected by the Athena magnets and interior rock and soil surfaces exposed by the Athena Rock Abrasion Tool and by trenching with rover wheels.

General State Channel Networks

Abstract: One of the fundamental challenges that hinder further adaption of decentralized cryptocurrencies is scalability. Because current cryptocurrencies require that all transactions are processed and stored on a distributed ledger -- the so-called blockchain -- transaction throughput is inherently limited. An important proposal to significantly improve scalability are off-chain protocols, where the massive amount of transactions is executed without requiring the costly interaction with the blockchain. Examples of off-chain protocols include payment channels and networks, which are currently deployed by popular cryptocurrencies such as Bitcoin and Ethereum. A further extension of payment networks envisioned for cryptocurrencies are so-called state channel networks. In contrast to payment networks that only support off-chain payments between users, state channel networks allow execution of arbitrary complex smart contracts. The main contribution of this work is to give the first full specification for general state channel networks. Moreover, we provide formal security definitions and prove the security of our construction against powerful adversaries. An additional benefit of our construction is the use of channel virtualization, which further reduces latency and costs in complex channel networks.

DeepRED – Rule Extraction from Deep Neural Networks

Abstract: Neural network classifiers are known to be able to learn very accurate models. In the recent past, researchers have even been able to train neural networks with multiple hidden layers (deep neural networks) more effectively and efficiently. However, the major downside of neural networks is that it is not trivial to understand the way how they derive their classification decisions. To solve this problem, there has been research on extracting better understandable rules from neural networks. However, most authors focus on nets with only one single hidden layer. The present paper introduces a new decompositional algorithm – DeepRED – that is able to extract rules from deep neural networks.