SEARCH · Engineering Papers
Results for “SCALING”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Cation ratio fluctuations in Cu 2 ZnSnS 4 at the 20 nm length scale investigated by analytical electron microscopy
Kesterite Cu 2 ZnSn(S,Se) 4 (CZTSSe) is a sustainable material for thin‐film photovoltaics with device efficiencies greater than 12% have been demonstrated. Despite similar crystal structure and polycrystalline film microstructures, there is widespread evidence for larger‐amplitude potential and bandgap fluctuations in CZTS than in the analogous Cu(In,Ga)Se 2 (CIGSe) chalcopyrite material. This disorder is believed to account for a sizable part of the larger open‐circuit voltage ( V OC ) deficit in CZTS devices, yet the detailed origins and length scales of these fluctuations have not been fully elucidated. Herein, we present a transmission electron microscopy study focusing on composition variation within bulk multicrystals of CZTS grown by the travelling heater method (THM). In these slow‐cooled, solution grown crystals we find direct evidence for spatial composition fluctuations of amplitude <1 at.% (∼5 × 10 20 cm −3 ) and thus, explainable by point defects. However, rather than being homogeneously‐distributed we find a characteristic 20 nm length scale for these fluctuations, which sets a definite length scale for band gap and potential fluctuations. At Σ3 grain boundaries, we find no evidence of composition variation compared to the bulk. The finding highlights such variations reported at grain boundaries in polycrystalline thin‐films are direct consequences of processing methods and not intrinsic properties of CZTS itself.
Linear and Nonlinear Solvers for Simulating Multiphase Flow within Large-Scale Engineered Subsurface Systems
Simulation of multiphase flow in the subsurface is well-known to be computationally challenging. While there have been many studies that have explored approaches to overcoming these challenges, they often utilize relatively simple case studies. In this paper, we focus on the unique numerical challenges posed by modeling large-scale engineered subsurface systems, characterized by discrete features embedded in a heterogeneous natural subsurface setting. The man-made features such as shafts, tunnels, and barriers often cause multiple challenges in modeling the domain for multiphase porous media flow. This flow scenario can have a wide range of applications such as nuclear waste repositories, enhanced recovery of a petroleum reservoir, geothermal engineering, and carbon sequestration. An example of these severe numerical challenges is the case of performance assessment (PA) for Waste Isolation Pilot Plant (WIPP), the only operating deep geological repository in the US, which simulates extreme material properties of bedded salt rock formation and extreme contrast due to open excavation next to the formation. The models have extremes not only of permeability and porosity but also of the constitutive models needed for multiphase flow; additionally, they have process models like salt creep closure reducing porosity over time, fracturing in clay and anhydrite interbeds of the bedded salt, gas generation from the waste materials, and unintentional human borehole intrusions in some scenarios. Numerical simulations require the solution of coupled systems of nonlinear PDEs; in our work, we use the open-source simulator PFLOTRAN which is based on Finite Volume discretization. The solution of the nonlinear equations requires use of the Newton-Raphson iteration at each time step, which entails the solution of the linearized Jacobian system at each iteration. The effects of all the processes (i.e., large number of unknowns, highly nonlinear constitutive relations, large contrasts in material properties in short distances) lead to an ill-conditioned Jacobian matrix that severely challenges traditional linear solver, i.e., stabilized biconjugate gradient with block Jacobi incomplete LU preconditioner (BCGS-ILU) leading to non-convergence for traditional Newton-Raphson nonlinear solver causing unacceptably long computation time for each model. This paper presents linear solvers such as constrained pressure residual (CPR) two-stage preconditioner with alternate-block-factorization (ABF) and quasi- implicit pressure and explicit saturation (QIMPES) decouplers and flexible generalized residual solver (FGMRES). The new general-purpose nonlinear solver, Newton trust-region dogleg Cauchy (NTRDC), is also introduced to resolve extreme nonlinearities in the models. We demonstrate the effectiveness of each method relative to the default BCGS-Newton solver. The two best cases had nearly 50 times speed-up and achieved completion of a simulation in 14 hours that never completed due to non-convergence with the default solver. We also investigate the strong scalability of each method and discuss some of the deficiencies found for Block Jacobi preconditioner using parallel domain decomposition, and node packing effects of modern processor architecture.
Optimizing checkpoint data placement with guaranteed burst buffer endurance in large-scale hierarchical storage systems
Non-volatile devices, such as SSDs, will be an integral part of the deepening storage hierarchy on large-scale HPC systems. These devices can be on the compute nodes as part of a distributed burst buffer service or they can be external. Wherever they are located in the hierarchy, one critical design issue is the SSD endurance under the write-heavy workloads, such as the checkpoint I/O for scientific applications. For these environments, it is widely assumed that checkpoint operations can occur once every 60 min and for each checkpoint step as much as half of the system memory can be written out. Unfortunately, for large-scale HPC applications, the burst buffer SSDs can be worn out much more quickly given the extensive amount of data written at every checkpoint step. One possible solution is to control the amount of data written by reducing the checkpoint frequency. However, a direct effect caused by reduced checkpoint frequency is the increased vulnerability window of system failures and therefore potentially wasted computation time, especially for large-scale compute jobs.
The magnetic shear-current effect: Generation of large-scale magnetic fields by the small-scale dynamo
A novel large-scale dynamo mechanism, the magnetic shear-current effect, is discussed and explored. Here, the effect relies on the interaction of magnetic fluctuations with a mean shear flow, meaning the saturated state of the small-scale dynamo can drive a large-scale dynamo – in some sense the inverse of dynamo quenching. The dynamo is non-helical, with the mean field ${\it\alpha}$ coefficient zero, and is caused by the interaction between an off-diagonal component of the turbulent resistivity and the stretching of the large-scale field by shear flow. Following up on previous numerical and analytic work, this paper presents further details of the numerical evidence for the effect, as well as an heuristic description of how magnetic fluctuations can interact with shear flow to produce the required electromotive force. The pressure response of the fluid is fundamental to this mechanism, which helps explain why the magnetic effect is stronger than its kinematic cousin, and the basic idea is related to the well-known lack of turbulent resistivity quenching by magnetic fluctuations. As well as being interesting for its applications to general high Reynolds number astrophysical turbulence, where strong small-scale magnetic fluctuations are expected to be prevalent, the magnetic shear-current effect is a likely candidate for large-scale dynamo in the unstratified regions of ionized accretion disks. Evidence for this is discussed, as well as future research directions and the challenges involved with understanding details of the effect in astrophysically relevant regimes.
Local-scale Arctic tundra heterogeneity affects regional-scale carbon dynamics
In northern Alaska nearly 65% of the terrestrial surface is composed of polygonal ground, where geomorphic tundra landforms disproportionately influence carbon and nutrient cycling over fine spatial scales. Process-based biogeochemical models used for local to Pan-Arctic projections of ecological responses to climate change typically operate at coarse-scales (1km 2 –0.5°) at which fine-scale (<1km 2 ) tundra heterogeneity is often aggregated to the dominant land cover unit. Here, we evaluate the importance of tundra heterogeneity for representing soil carbon dynamics at fine to coarse spatial scales. We leveraged the legacy of data collected near Utqiagvik, Alaska between 1973 and 2016 for model initiation, parameterization, and validation. Simulation uncertainty increased with a reduced representation of tundra heterogeneity and coarsening of spatial scale. Hierarchical cluster analysis of an ensemble of 21 st -century simulations reveals that a minimum of two tundra landforms (dry and wet) and a maximum of 4km 2 spatial scale is necessary for minimizing uncertainties (<10%) in regional to Pan-Arctic modeling applications.
Permeability scaling relationships of volcanic tuff from core to field scale measurements
A recent chemical explosive test in P-Tunnel at the Nevada National Security Site, Nevada, USA, was conducted to better understand how signals propagate from explosions in the subsurface. A primary signal of interest is the migration of gases that can be used to differentiate chemical from nuclear explosions. Gas migration is highly dependent on the rock permeability which is notoriously difficult to determine experimentally in the field due to a potentially large dependence on the scale over which measurements are made. Here, we present pre-explosion permeability estimates to characterize the geologic units surrounding the recent test. Permeability measurements were made at three scales of increasing size: core samples (≈2 cm), borehole packer system tests (≈1 m), and a pre-shot cavity pressurization test (> 10 m) across ten tuff units. Permeability estimates based on core measurements showed little difference from borehole packer tests. However, permeability in most rock units calibrated from cavity pressurization tests resulted in higher permeability estimates by up to two orders of magnitude. Here, we demonstrate that the scale of the measurement significantly impacts the characterization efforts of hydraulic properties in volcanic tuff, and that local-scale measurements (< 10 m scale) do not incorporate enough heterogeneity to accurately predict field-scale flow and mass transport.
Fast response of electron-scale turbulence to auxiliary heating cessation in National Spherical Torus Experiment
In this letter, we report the first observation of the fast response of electron-scale turbulence to auxiliary heating cessation in National Spherical Torus eXperiment [Ono et al., Nucl. Fusion 40, 557 (2000)]. The observation was made in a set of RF-heated L-mode plasmas with toroidal magnetic field of 0.55 T and plasma current of 300 kA. It is observed that electron-scale turbulence spectral power (measured with a high-k collective microwave scattering system) decreases significantly following fast cessation of RF heating that occurs in less than 200 μs. The large drop in the turbulence spectral power has a short time delay of about 1–2 ms relative to the RF cessation and happens on a time scale of 0.5–1 ms, much smaller than the energy confinement time of about 10 ms. Power balance analysis shows a factor of about 2 decrease in electron thermal diffusivity after the sudden drop of turbulence spectral power. Measured small changes in equilibrium profiles across the RF cessation are unlikely able to explain this sudden reduction in the measured turbulence and decrease in electron thermal transport, supported by local linear stability analysis and both local and global nonlinear gyrokinetic simulations. The observations imply that nonlocal flux-driven mechanism may be important for the observed turbulence and electron thermal transport.
Radiation from particles moving in small-scale magnetic fields created in solid-density laser-plasma laboratory experiments
Plasmas created by high-intensity lasers are often subject to the formation of kinetic-streaming instabilities, such as the Weibel instability, which lead to the spontaneous generation of high-amplitude, tangled magnetic fields. These fields typically exist on small spatial scales, i.e., “sub-Larmor scales.” Radiation from charged particles moving through small-scale electromagnetic (EM) turbulence has spectral characteristics distinct from both synchrotron and cyclotron radiation, and it carries valuable information on the statistical properties of the EM field structure and evolution. Consequently, this radiation from laser-produced plasmas may offer insight into the underlying electromagnetic turbulence. Here, we investigate the prospects for, and demonstrate the feasibility of, such direct radiative diagnostics for mildly relativistic, solid-density laser plasmas produced in lab experiments.
Characterization of laser-induced plasmas as a complement to high-explosive large-scale detonations
Experimental investigations into the characteristics of laser-induced plasmas indicate that LIBS provides a relatively inexpensive and easily replicable laboratory technique to isolate and measure reactions germane to understanding aspects of high-explosive detonations under controlled conditions. Furthermore, we examine spectral signatures and derived physical parameters following laser ablation of aluminum, graphite and laser-sparked air as they relate to those observed following detonation of high explosives and as they relate to shocked air. Laser-induced breakdown spectroscopy (LIBS) reliably correlates reactions involving atomic Al and aluminum monoxide (AlO) with respect to both emission spectra and temperatures, as compared to small- and large-scale high-explosive detonations. Atomic Al and AlO resulting from laser ablation and a cited small-scale study, decay within ~10 -5 s, roughly 100 times faster than the Al and AlO decay rates (~10 -3 s) observed following the large-scale detonation of an Al-encased explosive. Temperatures and species produced in laser-sparked air are compared to those produced with laser ablated graphite in air. With graphite present, CN is dominant relative to N 2 + . Thus, in studies where the height of the ablating laser's focus was altered relative to the surface of the graphite substrate, CN concentration was found to decrease with laser focus below the graphite surface, indicating that laser intensity is a critical factor in the production of CN, via reactive nitrogen.
Self-induced flavor conversion of supernova neutrinos on small scales
Self-induced flavor conversion of supernova (SN) neutrinos is a generic feature of neutrino-neutrino dispersion. The corresponding run-away modes in flavor space can spontaneously break the original symmetries of the neutrino flux and in particular can spontaneously produce small-scale features as shown in recent schematic studies. However, the unavoidable “multi-angle matter effect” shifts these small-scale instabilities into regions of matter and neutrino density which are not encountered on the way out from a SN. The traditional modes which are uniform on the largest scales are most prone for instabilities and thus provide the most sensitive test for the appearance of self-induced flavor conversion. As a by-product we clarify the relation between the time evolution of an expanding neutrino gas and the radial evolution of a stationary SN neutrino flux. Our results depend on several simplifying assumptions, notably stationarity of the solution, the absence of a “backward” neutrino flux caused by residual scattering, and global spherical symmetry of emission.
Performance Analysis, Design Considerations, and Applications of Extreme-Scale In Situ Infrastructures
A key trend facing extreme-scale computational science is the widening gap between computational and I/O rates, and the challenge that follows is how to best gain insight from simulation data when it is increasingly impractical to save it to persistent storage for subsequent visual exploration and analysis. One approach to this challenge is centered around the idea of in situ processing, where visualization and analysis processing is performed while data is still resident in memory. This paper examines several key design and performance issues related to the idea of in situ processing at extreme scale on modern platforms: Scalability, overhead, performance measurement and analysis, comparison and contrast with a traditional post hoc approach, and interfacing with simulation codes. We illustrate these principles in practice with studies, conducted on large-scale HPC platforms, that include a miniapplication and multiple science application codes, one of which demonstrates in situ methods in use at greater than 1M-way concurrency.
Two key features influencing community assembly processes at regional scale: Initial state and degree of change in environmental conditions
Belowground microbial communities strongly influence ecosystem function such that predicting function may rely on understanding ecological processes that assemble communities. Uncertainty remains, however, in what governs the relative contributions of different ecological processes. To help fill this knowledge gap we test the general hypothesis that both initial state and degree of change in environmental conditions govern the relative contributions of different ecological assembly processes. To do so we leveraged regional-scale nutrient and organic matter addition experiments and used soil organic matter (SOM) as a proxy of integrated soil environmental conditions. Consistent with our hypothesis, we found that both the initial amount of SOM and the degree of change in SOM—in response to nutrient addition—influenced the relative contributions of different ecological assembly processes. These influences were most clearly observed at the regional scale, suggesting potential scale dependence. More specifically, nutrient additions homogenized bacterial community composition due to enhanced influences of homogenizing dispersal when SOM content was initially high. In contrast, nutrient additions led to divergence in community composition due to variable selection when initial SOM was low and/or when SOM increased significantly in response to nutrient additions. Our findings indicate important connections among initial conditions, degree of change in environmental variables, and microbial community assembly processes that may influence ecosystem processes. These conceptual inferences highlight a need to strengthen connections between ecological theory and biogeochemical modeling.
Combining scaling relationships overcomes rate vs. overpotential tradeoffs in O2 molecular electrocatalysis
The development of advanced chemical-to-electrical energy conversions requires fast and efficient electrocatalysis of multi-electron/multi-proton reactions, such as the oxygen reduction reaction 10 (ORR). Using molecular catalysts, correlations between the reaction rate and energy efficiency have recently been identified. Improved catalysis requires circumventing the rate vs. overpotential tradeoffs implied by such ‘scaling relationships.’ Described here is an ORR system—using a soluble iron-porphyrin and weak acids—with the best reported combination of rate and efficiency for a soluble ORR catalyst. This advance is achieved not by “breaking” scaling relationships but 15 rather by combining two of them. Key to this behavior is a polycationic ligand, which enhances anionic ligand binding and changes the catalyst E1/2. These results show how combining scaling relationships is powerful way towards improved electrocatalysis.
Large-Scale Materials Modeling at Quantum Accuracy: Ab Initio Simulations of Quasicrystals and Interacting Extended Defects in Metallic Alloys
Ab initio electronic-structure has remained dichotomous between achievable accuracy and length-scale. Quantum many-body (QMB) methods realize quantum accuracy but fail to scale. Density functional theory (DFT) scales favorably but remains far from quantum accuracy. We present a framework that breaks this dichotomy by use of three interconnected modules: (i) invDFT: a methodological advance in inverse DFT linking QMB methods to DFT; (ii) MLXC: a machine-learned density functional trained with invDFT data, commensurate with quantum accuracy; (iii) DFT-FE-MLXC: an adaptive higher-order spectral finite-element (FE) based DFT implementation that integrates MLXC with efficient solver strategies and HPC innovations in FE-specific dense linear algebra, mixed-precision algorithms, and asynchronous compute-communication. Furthermore, we demonstrate a paradigm shift in DFT that not only provides an accuracy commensurate with QMB methods in ground-state energies, but also attains an unprecedented performance of 659.7 PFLOPS (43.1% peak FP64 performance) on 619,124 electrons using 8,000 GPU nodes of Frontier supercomputer.
Structural Design of a 1/5th Scale Gravo-Aeroelastically Scaled Wind Turbine Demonstrator Blade for Field Testing
This paper presents the structural design of a 21-meter 1/5th scale SUMR demonstrator (SUMR-D) blade, which was designed to replicate the full-scale behavior of a 104-meter SUMR13 initial blade rated at 13.2 MW. This is a challenging structural design as two competing objectives were required including the traditional safety objectives and novel scaling objectives based on gravo-aeroelastic scaling (GAS) requirements. To satisfy the two competing requirements, the spar cap, shear webs and materials were optimized while also ensuring a manufacturable and cost-effective design solution. Based on the wind turbine operating conditions of the test site, which is the NWTC, the blade was analyzed under the extreme loading case and corresponding safety factors were applied. The final blade model had an acceptable maximum strain, buckling performance, and displayed a safe level of deflection under the extreme loading condition. The flutter speed and fatigue life of the blade were also vetted. The mode shape was analyzed for the installation of the blade sensors. This paper documents the structural design process of SUMR-D to achieve both the safety and scaling objectives. At the end, we provide an acceptable solution to control blade mass, limiting the blade deflection, blade max strain and buckling under extreme loading condition by optimizing spar cap, shear web, and core materials while satisfying scaling requirements.
A PORTRAIT OF COLD GAS IN GALAXIES AT 60 pc RESOLUTION AND A SIMPLE METHOD TO TEST HYPOTHESES THAT LINK SMALL-SCALE ISM STRUCTURE TO GALAXY-SCALE PROCESSES
The cloud-scale density, velocity dispersion, and gravitational boundedness of the interstellar medium (ISM) vary within and among galaxies. In turbulent models, these properties play key roles in the ability of gas to form stars. New high-fidelity, high-resolution surveys offer the prospect to measure these quantities across galaxies. We present a simple approach to make such measurements and to test hypotheses that link small-scale gas structure to star formation and galactic environment. Our calculations capture the key physics of the Larson scaling relations, and we show good correspondence between our approach and a traditional “cloud properties” treatment. However, we argue that our method is preferable in many cases because of its simple, reproducible characterization of all emission. Using, low- J {sup 12}CO data from recent surveys, we characterize the molecular ISM at 60 pc resolution in the Antennae, the Large Magellanic Cloud (LMC), M31, M33, M51, and M74. We report the distributions of surface density, velocity dispersion, and gravitational boundedness at 60 pc scales and show galaxy-to-galaxy and intragalaxy variations in each. The distribution of flux as a function of surface density appears roughly lognormal with a 1 σ width of ∼0.3 dex, though the center of this distribution varies from galaxy to galaxy. The 60 pc resolution line width and molecular gas surface density correlate well, which is a fundamental behavior expected for virialized or free-falling gas. Varying the measurement scale for the LMC and M31, we show that the molecular ISM has higher surface densities, lower line widths, and more self-gravity at smaller scales.