On the prediction of the turbulent flow behind cylinder arrays via echo state networks. - In: Machine learning: science and technology, ISSN 2632-2153, Bd. 5 (2024), 3, 035005, S. 1-18
This study aims at the prediction of the turbulent flow behind cylinder arrays by the application of Echo State Networks (ESN). Three different arrangements of arrays of seven cylinders are chosen for the current study. These represent different flow regimes: single bluff body flow, transient flow, and co-shedding flow. This allows the investigation of turbulent flows that fundamentally originate from wake flows yet exhibit highly diverse dynamics. The data is reduced by Proper Orthogonal Decomposition (POD) which is optimal in terms of kinetic energy. The Time Coefficients of the POD Modes (TCPM) are predicted by the ESN. The network architecture is optimized with respect to its three main hyperparameters, Input Scaling (INS), Spectral Radius (SR), and Leaking Rate (LR), in order to produce the best predictions in terms of Weighted Prediction Score (WPS), a metric leveling statistic and deterministic prediction. In general, the ESN is capable of imitating the complex dynamics of turbulent flows even for longer periods of several vortex shedding cycles. Furthermore, the mutual interdependencies of the TCPM are well preserved. However, optimal hyperparameters depend strongly on the flow characteristics. Generally, as flow dynamics become faster and more intermittent, larger LR and INS values result in better predictions, whereas less clear trends for SR are observable.
https://doi.org/10.1088/2632-2153/ad5414
So far and yet so near - accelerating distributed joins with CXL. - In: 20th International Workshop on Data Management on New Hardware (DaMoN 2024), (2024), 7, insges. 9 S.
Distributed partitioned joins are one of the most expensive operators in distributed DBMSs where a major part of the execution is attributed to network transfer costs. Although high-speed network technologies, such as RDMA, can lower this cost, they still come with significantly higher latency than local DRAM access. The emerging CXL interconnect protocol promises to provide direct and cache-coherent access to remote memory while offering byte-addressable memory access without CPU intervention. For short-distance communication in distributed DBMSs, CXL represents an interesting alternative for low-latency requirements. In this work, we explore how CXL can be leveraged for engine-internal communication and data exchange. We discuss and apply communication strategies to distributed joins. We emulate various CXL characteristics based on optimistic and pessimistic assumptions on the real performance of upcoming CXL devices and evaluate their impact on the execution of distributed joins. Our results show that CXL has the potential to improve distributed join performance.
https://doi.org/10.1145/3662010.3663449
A robust, simple, and efficient convergence workflow for GW calculations. - In: npj computational materials, ISSN 2057-3960, Bd. 10 (2024), 135, S. 1-9
A robust, simple, and efficient convergence workflow for GW calculations in plane-wave-based codes is derived from more than 7000 GW calculations on a diverse dataset of 70 semiconducting and insulating solids divided into 60 bulk and 10 2D materials. The workflow can significantly accelerate material screening projects and high-precision single-system studies. Our method is based on two main results: The convergence of the two interdependent parameters in the numerical implementation of the dynamically screened Coulomb interaction W in a plane-wave basis set is accelerated by a ‘cheap first, expensive later’ coordinate search that maintains the same accuracy as a state-of-the-art convergence algorithm, but converges faster. In addition, we empirically establish the practical independence of the k-point grid and the aforementioned parameterization of W. Incorporating both results into one workflow dramatically speeds up convergence.
https://doi.org/10.1038/s41524-024-01311-9
Influence of additional intermediate thick Al layers on the reaction propagation and heat flow of Al/Ni reactive multilayers. - In: Advanced engineering materials, ISSN 1527-2648, Bd. 0 (2024), 0, 2400522, S. 1-8
This study investigates the effects of sputtering and electron beam evaporation (e-beam) on the microstructure and reactive properties of Al/Ni reactive multilayers (RMLs). The intermixing zone, a critical factor influencing reaction kinetics, is characterized using high-resolution transmission electron microscopy and found to be consistently 3 nm for both fabrication methods. Differential scanning calorimetry reveals that e-beam samples, with thicker Al layers, exhibit slightly higher total molar enthalpy and maintain high reaction temperatures despite reduced reaction velocities in comparison to sputtered samples. X-ray diffraction confirms the formation of both Al3Ni2 and AlNi phases in the e-beam samples. These findings indicate that while thicker bilayer structures reduce reaction velocity, they keep thermal output and mitigate the impact of intermixing zones, leading to similar total molar enthalpy. This analysis underscores the significance of deposition technique and bilayer thickness in optimizing the performance of Al/Ni RML, offering the possibility to establish different phase formations in thicker RML. It advances the control over the reactive properties of RMLs in their applications, for example, reactive bonding.
https://doi.org/10.1002/adem.202400522
Transfer of self-sustained reactions between thermally coupled reactive material elements. - In: Advanced engineering materials, ISSN 1527-2648, Bd. 0 (2024), 0, 2400870, S. 1-14
System engineering requires the implementation of assembly methods allowing the formation of functional elements according to a desired system design. This is possible by joining prefabricated elements on a desired place in the system or by forming functional materials at wanted locations. Both approaches need temperature treatment. A typical system in part consists of materials restricting the use of annealing procedures above 300 ˚C. For this reason, higher-temperature operations have to be localized to the point of use and transferred to the next reactive element. A further requirement is the reduction of the thermal budget of the localized joining process implementing ultrashort heat treatment operations. Reactive metallic multilayer offers the combined possibility of implementing a localized heat source and the formation of a functional intermetallic alloy. The article demonstrates the feasibility of a heat and reaction transfer chain in a topological structured pattern consisting of localized reactive multilayer materials under conditions of their gasless combustion. For the used Ni/Al multilayer material system, the critical (obstacle thickness) dimensions for the reaction and heat transfer are determined using a high-speed camera and pyrometer measurements.
https://doi.org/10.1002/adem.202400870
Properties of AlCl3 guanidine deep eutectic solvent with addition of polyethylene oxide of different molecular weights supported by quantum-chemical calculation. - In: Journal of energy storage, ISSN 2352-152X, Bd. 96 (2024), 112665, S. 1-8
Deep eutectic solvent of AlCl3 with guanidine HCl was synthesized in the molar ratio 3:1 in the presence of polyethylene glycols of various molecular weights from 1500 to 5&hahog;106 g/mol. Concentration of polymer was 2.5 %, 5 % and 10 %. Temperature dependence between 25 and 70 ˚C of conductivity of each electrolyte was calculated from the cyclic voltammetry of aluminum deposition-stripping measurements. It was found that conductivity increases with molecular weight of polymer in the interval between 2 and 10 kD. This result was interpreted as binding of AlCl3 guanidine to two different sites which have hydroxyl (chain ends) and ether (middle chain) oxygen with relatively higher and lower binding energy respectively. Polymer with longer chain has more low energy binding sites which causes lesser binding and consequently higher conductivity of electrolyte. B3LYP-D3 and DLPNO-CCSD(T) computations support experimental results. DES are discussed in terms of Pearson HSAB principle.
https://doi.org/10.1016/j.est.2024.112665
C60 fullerene reduces the level of liver damage in chronic alcohol intoxication of rats. - In: Molecules, ISSN 1420-3049, Bd. 29 (2024), 13, 2951, S. 1-15
The liver is the main organ responsible for the metabolism of ethanol, which suffers significantly as a result of tissue damage due to oxidative stress. It is known that C60 fullerenes are able to efficiently capture and inactivate reactive oxygen species in in vivo and in vitro systems. Therefore, the purpose of this study is to determine whether water-soluble C60 fullerene reduces the level of pathological process development in the liver of rats induced by chronic alcohol intoxication for 3, 6, and 9 months, depending on the daily dose (oral administration; 0.5, 1, and 2 mg/kg) of C60 fullerene throughout the experiment. In this context, the morphology of the C60 fullerene nanoparticles in aqueous solution was studied using atomic force microscopy. Such biochemical parameters of experimental animal blood as ALT (alanine aminotransferase), AST (aspartate aminotransferase), GGT (gamma-glutamyl transferase) and ALP (alkaline phosphatase) enzyme activities, CDT (carbohydrate-deficient transferrin) level, values of pro-antioxidant balance indicators (concentrations of H2O2 (hydrogen peroxide) and GSH (reduced glutathione), activities of CAT (catalase), SOD (superoxide dismutase) and GPx (selenium-dependent glutathione peroxidase)), and pathohistological and morphometric features of liver damage were analyzed. The most significant positive change in the studied biochemical parameters (up to 29 ± 2% relative to the control), as markers of liver damage, was recorded at the combined administration of alcohol (40% ethanol in drinking water) and water-soluble C60 fullerenes in the optimal dose of 1 mg/kg, which was confirmed by small histopathological changes in the liver of rats. The obtained results prove the prospective use of C60 fullerenes as powerful antioxidants for the mitigation of pathological conditions of the liver arising under prolonged alcohol intoxication.
https://doi.org/10.3390/molecules29132951
Generative convective parametrization of a dry atmospheric boundary layer. - In: Journal of advances in modeling earth systems, ISSN 1942-2466, Bd. 16 (2024), 6, e2023MS004012, S. 1-20
Turbulence parametrizations will remain a necessary building block in kilometer-scale Earth system models. In convective boundary layers, where the mean vertical gradients of conserved properties such as potential temperature and moisture are approximately zero, the standard ansatz which relates turbulent fluxes to mean vertical gradients via an eddy diffusivity has to be extended by mass-flux parametrizations for the typically asymmetric up- and downdrafts in the atmospheric boundary layer. We present a parametrization for a dry and transiently growing convective boundary layer based on a generative adversarial network. The training and test data are obtained from three-dimensional high-resolution direct numerical simulations. The model incorporates the physics of self-similar layer growth following from the classical mixed layer theory of Deardorff by a renormalization. This enhances the training data base of the generative machine learning algorithm and thus significantly improves the predicted statistics of the synthetically generated turbulence fields at different heights inside the boundary layer, above the surface layer. Differently to stochastic parametrizations, our model is able to predict the highly non-Gaussian and transient statistics of buoyancy fluctuations, vertical velocity, and buoyancy flux at different heights thus also capturing the fastest thermals penetrating into the stabilized top region. The results of our generative algorithm agree with standard two-equation mass-flux schemes. The present parametrization provides additionally the granule-type horizontal organization of the turbulent convection which cannot be obtained in any of the other model closures. Our proof of concept-study also paves the way to efficient data-driven convective parametrizations in other natural flows.
https://doi.org/10.1029/2023MS004012
Port-Hamiltonian descriptor systems are relative generically controllable and stabilizable. - In: Mathematics of control, signals, and systems, ISSN 1435-568X, Bd. 0 (2024), 0, insges. 37 S.
The present work is a successor of Ilchmann and Kirchhoff (Math Control Signals Syst 33:359-377, 2021) on generic controllability and of Ilchmann and Kirchhoff (Math Control Signals Syst 35:45-76, 2022) on relative generic controllability of linear differential-algebraic equations. We extend the result from general, unstructured differential-algebraic equations to differential-algebraic equations of port-Hamiltonian type. We derive results on relative genericity. These findings are the basis for characterizing relative generic controllability of port-Hamiltonian systems in terms of dimensions. A similar result is proved for relative generic stabilizability.
https://doi.org/10.1007/s00498-024-00392-7
Minimizing the homogeneous L2-gain of homogeneous differentiators. - In: European journal of control, ISSN 1435-5671, Bd. 0 (2024), 0, 101039, S. 1-8
The differentiation of noisy signals using the family of homogeneous differentiators is considered. It includes the high-gain (linear) as well as robust exact (discontinuous) differentiator. To characterize the effect of noise and disturbance on the differentiation estimation error, the generalized, homogeneous L2-gain is utilized. Analog to the classical Lp-gain, it is not defined for the discontinuous case w.r.t. disturbances acting on the last channel. Thus, only continuous differentiators are addressed. The gain is estimated using a differential dissipation inequality, where a scaled Lyapunov function acts as storage function for the homogeneous L2 supply rate. The fixed differentiator gains are scaled with a gain-scaling parameter similar to the high-gain differentiator. This paper shows the existence of an optimal scaling which minimizes the homogeneous L2-gain estimate and provides a procedure to obtain it locally. Differentiators of dimension two are considered and the results are illustrated via numerical evaluation and a simulation example.
https://doi.org/10.1016/j.ejcon.2024.101039