Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Partitioning”

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.

At least 91 records · Page 5

Texture Development and Stress–Strain Partitioning in Periclase + Halite Aggregates

Multiphase materials are widely applied in engineering due to desirable mechanical properties and are of interest to geoscience as rocks are multiphase. High-pressure mechanical behavior is important for understanding the deep Earth where rocks deform at extreme pressure and temperature. In order to systematically study the underlying physics of multiphase deformation at high pressure, we perform diamond anvil cell deformation experiments on MgO + NaCl aggregates with varying phase proportions. Lattice strain and texture evolution are recorded using in-situ synchrotron x-ray diffraction and are modeled using two-phase elasto-viscoplastic self-consistent (EVPSC) simulations to deduce stress, strain, and deformation mechanisms in individual phases and the aggregate. Texture development of MgO and NaCl are affected by phase proportions. In NaCl, a (100) compression texture is observed when small amounts of MgO are present. In contrast, when deformed as a single phase or when large amounts of MgO are present, NaCl develops a (110) texture. Stress and strain evolution in MgO and NaCl also show different trends with varying phase proportions. Based on the results from this study, we construct a general scheme of stress evolution as a function of phase proportion for individual phases and the aggregate.

Lin, Feng↗

Multilevel Graph Partitioning for Three-Dimensional Discrete Fracture Network Flow Simulations

We present a topology-based method for mesh-partitioning in three-dimensional discrete fracture network (DFN) simulations that takes advantage of the intrinsic multi-level nature of a DFN. DFN models are used to simulate flow and transport through low-permeability fractured media in the subsurface by explicitly representing fractures as discrete entities. The governing equations for flow and transport are numerically integrated on computational meshes generated on the interconnected fracture networks. Modern high-fidelity DFN simulations require high-performance computing on multiple processors where performance and scalability depends partially on obtaining a high-quality partition of the mesh to balance work-loads and minimize communication across all processors. The discrete structure of a DFN naturally lends itself to various graph representations, which can be thought of as coarse-scale representations of the computational mesh. Using this concept, we develop two applications of the multilevel graph partitioning algorithm to partition the mesh of a DFN. In the first, we project a partition of the graph based on the DFN topology onto the mesh of the DFN and in the second, this DFN-based projection is used as the initial condition for further partitioning refinement of the mesh. We compare the performance of these methods with standard multi-level graph partitioning using graph-based metrics (cut, imbalance, partitioning time), computational-based metrics (FLOPS, iterations, solver time), and total run time. The DFN-based and the mesh-based partitioning methods are comparable in terms of the graph-based metrics, but the time required to obtain the partition is several orders of magnitude faster using the DFN-based partitions. The computation-based metrics show comparable performance between both methods so, in combination, the DFN-based partitions are several orders of magnitude faster than the mesh-based partition. Furthermore, the method which uses the DFN-partition solution as the initial condition of the mesh partition provided cut and imbalance values that were close to the mesh-based partition but in a fraction of the time. In turn, this hybrid method outperformed both of the other methods in terms of the total run time.

58 GEOSCIENCES↗

Variable length adjacent partitioning for PTS based PAPR reduction of OFDM signal

Peak-to-Average power ratio (PAPR) is a major drawback in OFDM communication. It leads the power amplifier into nonlinear region operation resulting into loss of data integrity. As such, there is a strong motivation to find techniques to reduce PAPR. Partial Transmit Sequence (PTS) is an attractive scheme for this purpose. Judicious partitioning the OFDM data frame into disjoint subsets is a pivotal component of any PTS scheme. Out of the existing partitioning techniques, adjacent partitioning is characterized by an attractive trade-off between cost and performance. With an aim of determining effects of length variability of adjacent partitions, we performed an investigation into the performances of a variable length adjacent partitioning (VL-AP) and fixed length adjacent partitioning in comparison with other partitioning schemes such as pseudorandom partitioning. Simulation results with different modulation and partitioning scenarios showed that fixed length adjacent partition had better performance compared to variable length adjacent partitioning. As expected, simulation results showed a slightly better performance of pseudorandom partitioning technique compared to fixed and variable adjacent partitioning schemes. However, as the pseudorandom technique incurs high computational complexities, adjacent partitioning schemes were still seen as favorable candidates for PAPR reduction.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Partitioning of Nanoparticles into Organic Phases and Model Cells

There is a recognized need to understand and predict the fate, transport and bioavailability of engineered nanoparticles (ENPs) in aquatic and soil ecosystems. Recent research focuses on either collection of empirical data (e.g., removal of a specific NP through water or soil matrices under variable experimental conditions) or precise NP characterization (e.g. size, degree of aggregation, morphology, zeta potential, purity, surface chemistry, and stability). However, it is almost impossible to transition from these precise measurements to models suitable to assess the NP behavior in the environment with complex and heterogeneous matrices. For decades, the USEPA has developed and applies basic partitioning parameters (e.g., octanol-water partition coefficients) and models (e.g., EPI Suite, ECOSAR) to predict the environmental fate, bioavailability, and toxicity of organic pollutants (e.g., pesticides, hydrocarbons, etc.). In this project we have investigated the hypothesis that NP partition coefficients between water and organic phases (octanol or lipid bilayer) is highly dependent on their physiochemical properties, aggregation, and presence of natural constituents in aquatic environments (salts, natural organic matter), which may impact their partitioning into biological matrices (bioaccumulation) and human exposure (bioavailability) as well as the eventual usage in modeling the fate and bioavailability of ENPs. In this report, we use the terminology "partitioning" to operationally define the fraction of ENPs distributed among different phases. The mechanisms leading to this partitioning probably involve both chemical force interactions (hydrophobic association, hydrogen bonding, ligand exchange, etc.) and physical forces that bring the ENPs in close contact with the phase interfaces (diffusion, electrostatic interactions, mixing turbulence, etc.). Our work focuses on partitioning, but also provides insight into the relative behavior of ENPs as either "more like dissolved substances" or "more like colloids" as the division between behaviors of macromolecules versus colloids remains ill-defined. Below we detail our work on two broadly defined objectives: (i) Partitioning of ENP into octanol, lipid bilayer, and water, and (ii) disruption of lipid bilayers by ENPs. We have found that the partitioning of NP reaches pseudo-equilibrium distributions between water and organic phases. The equilibrium partitioning most strongly depends on the particle surface charge, which leads us to the conclusion that electrostatic interactions are critical to understanding the fate of NP in the environment. We also show that the kinetic rate at which particle partition is a function of their size (small particles partition faster by number) as can be predicted from simple DLVO models. We have found that particle number density is the most effective dosimetry to present our results and provide quantitative comparison across experiments and experimental platforms. Cumulatively, our work shows that lipid bilayers are a more effective organic phase than octanol because of the definable surface area and ease of interpretation of the results. Our early comparison of NP partitioning between water and lipids suggest that this measurement can be predictive of bioaccumulation in aquatic organisms. We have shown that nanoparticle disrupt lipid bilayer membranes and detail how NP-bilayer interaction leads to the malfunction of lipid bilayers in regulating the fluxes of ionic charges and molecules. Our results show that the disruption of the lipid membranes is similar to that of toxin melittin, except single particles can disrupt a bilayer. We show that only a single particle is required to disrupt a 150 nm DOPC liposome. The equilibrium leakage of membranes is a function of the particle number density and particle surface charge, consistent with results from our partitioning experiments. Our disruption experiments with varying surface functionality show that positively charged particles (poly amine) are most disruptive, consistent with in in vitro toxicity panels using cell cultures. Overall, this project has resulted in 8 published or submitted archival papers and has been presented 12 times. We have trained five students and provided growth opportunities for a postdoc.

54 ENVIRONMENTAL SCIENCES↗

Choosing the best partition of the output from a large-scale simulation

Data partitioning becomes necessary when a large-scale simulation produces more data than can be feasibly stored. The goal is to partition the data, typically so that every element belongs to one and only one partition, and store summary information about the partition, either a representative value plus an estimate of the error or a distribution. Once the partitions are determined and the summary information stored, the raw data is discarded. This process can be performed in-situ; meaning while the simulation is running. When creating the partitions there are many decisions that researchers must make. For instance, how to determine once an adequate number of partitions have been created, how are the partitions created with respect to dividing the data, or how many variables should be considered simultaneously. In addition, decisions must be made for how to summarize the information within each partition. Because of the combinatorial number of possible ways to partition and summarize the data, a method of comparing the different possibilities will help guide researchers into choosing a good partitioning and summarization scheme for their application.

97 MATHEMATICS AND COMPUTING↗

High-order partitioned spectral deferred correction solvers for multiphysics problems

We present an arbitrarily high-order, conditionally stable, partitioned spectral deferred correction (SDC) method for solving multiphysics problems using a sequence of pre-existing single-physics solvers. This method extends the work in [1], [2], which used implicit-explicit Runge-Kutta methods (IMEX) to build high-order, partitioned multiphysics solvers. We consider a generic multiphysics problem modeled as a system of coupled ordinary differential equations (ODEs), coupled through coupling terms that can depend on the state of each subsystem; therefore the method applies to both a semi-discretized system of partial differential equations (PDEs) or problems naturally modeled as coupled systems of ODEs. The sufficient conditions to build arbitrarily high-order partitioned SDC schemes are derived. Based on these conditions, various of partitioned SDC schemes are designed. The stability of the first-order partitioned SDC scheme is analyzed in detail on a coupled, linear model problem. We show that the scheme is conditionally stable, and under conditions on the coupling strength, the scheme can be unconditionally stable. We demonstrate the performance of the proposed partitioned solvers on several classes of multiphysics problems with moderate coupling strength. They include a stiff linear system of ODEs, advection-diffusion-reaction systems, and fluid-structure interaction problems with both incompressible and compressible flows, where we verify the design order of the SDC schemes and study various stability properties. We also directly compare the accuracy, stability, and cost of the proposed partitioned SDC solver with the partitioned IMEX method in [1], [2] on this suite of test problems. The results suggest that the high-order partitioned SDC solvers are more robust than the partitioned IMEX solvers for the numerical examples considered in this work, while the IMEX methods require fewer implicit solves.

97 MATHEMATICS AND COMPUTING↗

AutoAtlas: Neural Network for 3D Unsupervised Partitioning and Representation Learning

Here we present a novel neural network architecture called AutoAtlas for fully unsupervised partitioning and representation learning of 3D brain Magnetic Resonance Imaging (MRI) volumes. AutoAtlas consists of two neural network components: one neural network to perform multi-label partitioning based on local texture in the volume, and a second neural network to compress the information contained within each partition. We train both of these components simultaneously by optimizing a loss function that is designed to promote accurate reconstruction of each partition, while encouraging spatially smooth and contiguous partitioning, and discouraging relatively small partitions. We show that the partitions adapt to the subject specific structural variations of brain tissue while consistently appearing at similar spatial locations across subjects. AutoAtlas also produces very low dimensional features that represent local texture of each partition. We demonstrate prediction of metadata associated with each subject using the derived feature representations and compare the results to prediction using features derived from FreeSurfer anatomical parcellation. Since our features are intrinsically linked to distinct partitions, we can then map values of interest, such as partition-specific feature importance scores onto the brain for visualization.

42 ENGINEERING↗

Machine and Application Aware Partitioning for Adaptive Mesh Refinement Applications. In: HPDC '17 Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing

Load balancing and partitioning are critical when it comes to parallel computations. Popular partitioning strategies based on space filling curves focus on equally dividing work. The partitions produced are independent of the architecture or the application. Given the ever-increasing relative cost of data movement and increasing heterogeneity of our architectures, it is no longer sufficient to only consider an equal partitioning of work. Minimizing communication costs are equally if not more important. Our hypothesis is that an unequal partitioning that minimizes communication costs significantly can scale and perform better than conventional equal-work partitioning schemes. This tradeoff is dependent on the architecture as well as the application. We validate our hypothesis in the context of a finite-element computation utilizing adaptive mesh-refinement. Our central contribution is a new partitioning scheme that minimizes the overall runtime of subsequent computations by performing architecture and application-aware non-uniform work assignment in order to decrease time to solution, primarily by minimizing data-movement. We evaluate our algorithm by comparing it against standard space-filling curve based partitioning algorithms and observing time-to-solution as well as energy-to-solution for solving Finite Element computations on adaptively refined meshes. We demonstrate excellent scalability of our new partition algorithm up to $262,144$ cores on ORNL's Titan and demonstrate that the proposed partitioning scheme reduces overall energy as well as time-to-solution for application codes by up to 22.0%.

Fernando, Milinda↗

PARTITIONING OF GADOLINIUM IN THE CHEMICAL PROCESSING CELL

A combination of short-term beaker tests and longer-duration Sludge Receipt and Adjustment Tank (SRAT) simulations were performed to investigate the relative partitioning behaviors of gadolinium and iron under conditions applicable to the Chemical Processing Cell (CPC). The testing was performed utilizing non-radioactive simple Fe-Gd slurries, non-radioactive Sludge Batch 6 simulant slurries, and a radioactive real-waste slurry representative of Sludge Batch 7 material. The testing focused on the following range of conditions: (a) Fe:Gd ratios of 25-100; (b) pH values of 2-6; (c) acidification via addition of nitric, formic, and glycolic acids; (d) temperatures of {approx}93 C and {approx}22 C; and (e) oxalate concentrations of <100 mg/kg and {approx}10,000 mg/kg. The purpose of the testing was to provide data for assessing the potential use of gadolinium as a supplemental neutron poison when dispositioning excess plutonium. Understanding of the partitioning behavior of gadolinium in the CPC was the first step in assessing gadolinium's potential applicability. Significant fractions of gadolinium partitioned to the liquid-phase at pH values of 4.0 and below, regardless of the Fe:Gd ratio. In SRAT simulations targeting nitric and formic acid additions of 150% acid stoichiometry, the pH dropped to a minimum of 3.5-4.0, and the maximum fractions of gadolinium and iron partitioning to solution were both {approx}20%. In contrast, in a SRAT simulation utilizing a nitric and formic acid addition under atypical conditions (due to an anomalously low insoluble solids content), the pH dropped to a minimum of 3.7, and the maximum fractions of gadolinium and iron partitioning to solution were {approx}60% and {approx}70%, respectively. When glycolic acid was used in combination with nitric and formic acids at 100% acid stoichiometry, the pH dropped to a minimum of 3.6-4.0, and the maximum fractions of gadolinium and iron partitioning to solution were 60-80% and 3-5%, respectively. Thus, the presence of glycolic acid increased dissolution of gadolinium, but decreased dissolution of iron. In beaker tests, the fractions of gadolinium partitioning to solution were all less than the minimum detection limits at pH 6, on the order of a few percent at pH 4, and ranging from 70-90% at pH 2. In contrast, the fractions of iron partitioning to solution were all less than the minimum detection limits at pH 6, {le} 0.04% at pH 4, and {le} 0.9% at pH 2. A possible explanation for the small magnitude of these fractions (as compared to the fractions observed in the SRAT simulations) was incomplete equilibrium, due to the short duration (30 minutes) of the beaker tests. As demonstrated by the SRAT simulations, the typical partitioning equilibration time was on the order of hours. The Fe:Gd ratio appeared to impact the extent of liquid-phase conditions under certain conditions, although the exact relationship was not clear. Temperature impacts on the liquid-phase gadolinium concentrations were modest, with liquid phase concentrations typically increasing about 25% as temperatures rose from {approx}22 C to {approx}93 C. The presence of high concentrations of oxalate did not appear to change the liquid-phase gadolinium concentrations - however, it did increase the liquid-phase iron concentrations (from being undetectable to being detectable but still minor). Additional gadolinium partitioning testing is recommended. Of greatest usefulness will be SRAT simulations focusing on a wider range of acid addition scenarios and alternate sludge compositions, particularly those specific to future sludge batches where addition of excess plutonium is being considered.

54 ENVIRONMENTAL SCIENCES↗

Dissecting the Genetic Architecture of Carbon Partitioning in Sorghum Using Multiscale Phenotypes

Carbon partitioning in plants may be viewed as a dynamic process composed of the many interactions between sources and sinks. The accumulation and distribution of fixed carbon is not dictated simply by the sink strength and number but is dependent upon the source, pathways, and interactions of the system. As such, the study of carbon partitioning through perturbations to the system or through focus on individual traits may fail to produce actionable developments or a comprehensive understanding of the mechanisms underlying this complex process. Using the recently published sorghum carbon-partitioning panel, we collected both macroscale phenotypic characteristics such as plant height, above-ground biomass, and dry weight along with microscale compositional traits to deconvolute the carbon-partitioning pathways in this multipurpose crop. Multivariate analyses of traits resulted in the identification of numerous loci associated with several distinct carbon-partitioning traits, which putatively regulate sugar content, manganese homeostasis, and nitrate transportation. Using a multivariate adaptive shrinkage approach, we identified several loci associated with multiple traits suggesting that pleiotropic and/or interactive effects may positively influence multiple carbon-partitioning traits, or these overlaps may represent molecular switches mediating basal carbon allocating or partitioning networks. Conversely, we also identify a carbon tradeoff where reduced lignin content is associated with increased sugar content. The results presented here support previous studies demonstrating the convoluted nature of carbon partitioning in sorghum and emphasize the importance of taking a holistic approach to the study of carbon partitioning by utilizing multiscale phenotypes.

59 BASIC BIOLOGICAL SCIENCES↗

Limited evidence for spatial resource partitioning across temperate grassland biodiversity experiments

Locally, plant species richness supports many ecosystem functions. Yet, the mechanisms driving these often-positive biodiversity–ecosystem functioning relationships are not well understood. Spatial resource partitioning across vertical resource gradients is one of the main hypothesized causes for enhanced ecosystem functioning in more biodiverse grasslands. Spatial resource partitioning occurs if species differ in where they acquire resources and can happen both above- and belowground. However, studies investigating spatial resource partitioning in grasslands provide inconsistent evidence. We present the results of a meta-analysis of 21 data sets from experimental species-richness gradients in grasslands. We test the hypothesis that increasing spatial resource partitioning along vertical resource gradients enhances ecosystem functioning in diverse grassland plant communities above- and belowground. To test this hypothesis, we asked three questions. (1) Does species richness enhance biomass production or community resource uptake across sites? (2) Is there evidence of spatial resource partitioning as indicated by resource tracer uptake and biomass allocation above- and belowground? (3) Is evidence of spatial resource partitioning correlated with increased biomass production or community resource uptake? Although plant species richness enhanced community nitrogen and potassium uptake and biomass production above- and belowground, we found that plant communities did not meet our criteria for spatial resource partitioning, though they did invest in significantly more aboveground biomass in higher canopy layers in mixture relative to monoculture. Furthermore, the extent of spatial resource partitioning across studies was not positively correlated with either biomass production or community resource uptake. Our results suggest that spatial resource partitioning across vertical resource gradients alone does not offer a general explanation for enhanced ecosystem functioning in more diverse temperate grasslands.

54 ENVIRONMENTAL SCIENCES↗

Confocal Raman Microscopy for In-situ Measurement of Phospholipid-Water Partitioning into Model Phospholipid Bilayers within Individual Chromatographic Particles

The phospholipid-water partition coefficient is a commonly measured parameter that correlates with drug efficacy, small-molecule toxicity, and accumulation of molecules in biological systems in the environment. Despite the utility of this parameter, methods for measuring phospholipid-water partition coefficients are limited. This is due to the difficulty of making quantitative measurements in vesicle membranes or supported phospholipid bilayers, both of which are small-volume phases that challenge the sensitivity of many analytical techniques. In this paper, we employ in-situ confocal Raman microscopy to probe the partitioning of a model membrane-active compound, 2-(4-isobutylphenyl) propionic acid or ibuprofen, into both hybrid- and supported-phospholipid bilayers deposited on the pore walls of individual chromatographic particles. The large surface-area-to-volume ratio of chromatographic silica allows interrogation of a significant lipid bilayer area within a very small volume. The local phospholipid concentration within a confocal probe volume inside the particle can be as high as 0.5 M, which overcomes the sensitivity limitations of making measurements in the limited membrane areas of single vesicles or planar supported bilayers. Quantitative determination of ibuprofen partitioning is achieved by using the phospholipid acyl-chains of the within-particle bilayer as an internal standard. This approach is tested for measurements of pH-dependent partitioning of ibuprofen into both hybrid-lipid and supported-lipid bilayers within silica particles, and the results are compared with octanol-water partitioning and with partitioning into individual optically-trapped phospholipid vesicle membranes. Finally and additionally, the impact of ibuprofen partitioning on bilayer structure is evaluated for both within-particle model membranes and compared with the structural impacts of partitioning into vesicle lipid bilayers.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Radionuclide Partitioning in an Underground Nuclear Test Cavity

In 2004, a borehole was drilled into the 1983 Chancellor underground nuclear test cavity to investigate the distribution of radionuclides within the cavity. Sidewall core samples were collected from a range of depths within the re-entry hole and two sidetrack holes. Upon completion of drilling, casing was installed and a submersible pump was used to collect groundwater samples. Test debris and groundwater samples were analyzed for a variety of radionuclides including the fission products {sup 99}Tc, {sup 125}Sb, {sup 129}I, {sup 137}Cs, and {sup 155}Eu, the activation products {sup 60}Co, {sup 152}Eu, and {sup 154}Eu, and the actinides U, Pu, and Am. In addition, the physical and bulk chemical properties of the test debris were characterized using Scanning Electron Microscopy (SEM) and Electron Microprobe measurements. Analytical results were used to evaluate the partitioning of radionuclides between the melt glass, rubble, and groundwater phases in the Chancellor test cavity. Three comparative approaches were used to calculate partitioning values, though each method could not be applied to every nuclide. These approaches are based on: (1) the average Area 19 inventory from Bowen et al. (2001); (2) melt glass, rubble, and groundwater mass estimates from Zhao et al. (2008); and (3) fission product mass yield data from England and Rider (1994). The U and Pu analyses of the test debris are classified and partitioning estimates for these elements were calculated directly from the classified Miller et al. (2002) inventory for the Chancellor test. The partitioning results from this study were compared to partitioning data that were previously published by the IAEA (1998). Predictions of radionuclide distributions from the two studies are in agreement for a majority of the nuclides under consideration. Substantial differences were noted in the partitioning values for {sup 99}Tc, {sup 125}Sb, {sup 129}I, and uranium. These differences are attributable to two factors: chemical volatility effects that occur during the initial plasma condensation, and groundwater remobilization that occurs over a much longer time frame. Fission product partitioning is very sensitive to the early cooling history of the test cavity because the decay of short-lived (t{sub 1/2} < 1 hour) fission-chain precursors occurs on the same time scale as melt glass condensation. Fission product chains that include both volatile and refractory elements, like the mass 99, 125, and 129 chains, can show large variations in partitioning behavior depending on the cooling history of the cavity. Uranium exhibits similar behavior, though the chemical processes are poorly understood. The water temperature within the Chancellor cavity remains elevated (75 C) more than two decades after the test. Under hydrothermal conditions, high solubility chemical species such as {sup 125}Sb and {sup 129}I are readily dissolved and transported in solution. SEM analyses of melt glass samples show clear evidence of glass dissolution and secondary hydrothermal mineral deposition. Remobilization of {sup 99}Tc is also expected during hydrothermal activity, but moderately reducing conditions within the Chancellor cavity appear to limit the transport of {sup 99}Tc. It is recommended that the results from this study should be used together with the IAEA data to update the range in partitioning values for contaminant transport models at the Nevada National Security Site (formerly known as the Nevada Test Site).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗