Search NASA⌕ Search

SEARCH · Search NASA

Results for “clustering methods”

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 721 records · Page 40

NWChem: Past, present, and future

Specialized computational chemistry packages have permanently reshaped the landscape of chemical sciences by providing tools to support and guide the experimental effort and for prediction of chemical and materials properties. In this regard, a special role has been played by electronic structure packages where complex chemical and materials processes can be modeled using first-principle-driven methodologies. Over the last few decades, the rapid development of computing technologies and a tremendous increase in computational power has offered a unique chance to study complex chemical transformations using sophisticated and predictive many-body techniques to describe correlated behavior of electrons in molecular and condensed phase systems at different levels of theory. In enabling these simulations, a critical role has been played by novel parallel algorithms capable of taking advantage of computational resources to address polynomial scaling of electronic structure methods. NWChem was among the first electronic structure codes that focused on delivering scalable performance for electronic structure simulations. Herein, we briefly review the NWChem suite of computational codes including its history, design principles, parallel tools, current capabilities, outreach and outlook.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Leveraging graph clustering techniques for cyber‐physical system analysis to enhance disturbance characterisation

Abstract Cyber‐physical systems have behaviour that crosses domain boundaries during events such as planned operational changes and malicious disturbances. Traditionally, the cyber and physical systems are monitored separately and use very different toolsets and analysis paradigms. The security and privacy of these cyber‐physical systems requires improved understanding of the combined cyber‐physical system behaviour and methods for holistic analysis. Therefore, the authors propose leveraging clustering techniques on cyber‐physical data from smart grid systems to analyse differences and similarities in behaviour during cyber‐, physical‐, and cyber‐physical disturbances. Since clustering methods are commonly used in data science to examine statistical similarities in order to sort large datasets, these algorithms can assist in identifying useful relationships in cyber‐physical systems. Through this analysis, deeper insights can be shared with decision‐makers on what cyber and physical components are strongly or weakly linked, what cyber‐physical pathways are most traversed, and the criticality of certain cyber‐physical nodes or edges. This paper presents several types of clustering methods for cyber‐physical graphs of smart grid systems and their application in assessing different types of disturbances for informing cyber‐physical situational awareness. The collection of these clustering techniques provide a foundational basis for cyber‐physical graph interdependency analysis.

97 MATHEMATICS AND COMPUTING↗

Novel Nanostructured Rare-Earth-Free Magnetic Materials with High Energy Products

Novel nanostructured Zr 2 Co 11 –based magnetic materials are fabricated in a single step process using cluster–deposition method. Here, the composition, atomic ordering, and spin structure are precisely controlled to achieve a substantial magnetic remanence and coercivity, as well as the highest energy product for non–rare–earth and Pt–free permanent–magnet alloys.

36 MATERIALS SCIENCE↗

Two‐Dimensional Boron Monolayers Mediated by Metal Substrates

Abstract Two‐dimensional (2D) materials, such as graphene and boron nitride, have specific lattice structures independent of external conditions. In contrast, the structure of 2D boron sensitively depends on metal substrate, as we show herein using the cluster expansion method and a newly developed surface structure‐search method, both based on first‐principles calculations. The preferred 2D boron on weaker interacting Au is nonplanar with significant buckling and numerous polymorphs as in vacuum, whereas on more reactive Ag, Cu, and Ni, the polymorphic energy degeneracy is lifted and a particular planar structure is found to be most stable. We also show that a layer composed of icosahedral B 12 is unfavorable on Cu and Ni but unexpectedly becomes a possible minimum on Au and Ag. The substrate‐dependent 2D boron choices originate from a competition between the strain energy of buckling and chemical energy of electronic hybridization between boron and metal.

Zhang, Zhuhua↗

Gate‐Voltage Control of Borophene Structure Formation

Abstract Boron nanostructures are easily charged but how charge carriers affect their structural stability is unknown. We combined cluster expansion methods with first‐principles calculations to analyze the dependence of the preferred structure of two‐dimensional (2D) boron, or “borophene”, on charge doping controlled by a gate voltage. At a reasonable doping level of 3.12×10 14 cm −2 , the hollow hexagon concentration in the ground state of 2D boron increases to 1/7 from 1/8 in its charge‐neutral state. The numerical result for the dependence of hollow hexagon concentration on the doping level is well described by an analytical method based on an electron‐counting rule. Aside from in‐plane electronic bonding, the hybridization among out‐of‐plane boron orbitals is crucial for determining the relative stability of different sheets at a given doping level. Our results offer new insight into the stability mechanism of 2D boron and open new ways for the control of the lattice structure during formation.

Zhang, Zhuhua↗

Two‐Dimensional Boron Monolayers Mediated by Metal Substrates

Abstract Two‐dimensional (2D) materials, such as graphene and boron nitride, have specific lattice structures independent of external conditions. In contrast, the structure of 2D boron sensitively depends on metal substrate, as we show herein using the cluster expansion method and a newly developed surface structure‐search method, both based on first‐principles calculations. The preferred 2D boron on weaker interacting Au is nonplanar with significant buckling and numerous polymorphs as in vacuum, whereas on more reactive Ag, Cu, and Ni, the polymorphic energy degeneracy is lifted and a particular planar structure is found to be most stable. We also show that a layer composed of icosahedral B 12 is unfavorable on Cu and Ni but unexpectedly becomes a possible minimum on Au and Ag. The substrate‐dependent 2D boron choices originate from a competition between the strain energy of buckling and chemical energy of electronic hybridization between boron and metal.

Zhang, Zhuhua↗

Evaluating the benefits of kinetic Monte Carlo and microkinetic modeling for catalyst design studies in the presence of lateral interactions

Popular computational catalyst design strategies rely on the identification of reactivity descriptors, which can be used along with Brønsted–Evans–Polanyi (BEP) and scaling relations as input to a microkinetic model (MKM) to make predictions for activity or selectivity trends. The main benefit of this approach is related to the inherent dimensionality reduction of the large material space to just a few catalyst descriptors. Conversely, it is well documented that a small set of descriptors is insufficient to capture the intricacies and complexities of a real catalytic system. The inclusion of coverage effects through lateral adsorbate-adsorbate interactions can narrow the gap between simplified descriptor predictions and real systems, but mean-field MKMs cannot properly account for local coverage effects. This shortcoming of the mean-field approximation can be rectified by switching to a lattice-based kinetic Monte Carlo (kMC) method using cluster expansion representation of adsorbate–adsorbate lateral interactions. Using the prototypical CO oxidation reaction as an example, we critically evaluate the benefits of kMC over MKM in terms of trend predictions and computational cost when using only a small set of input parameters. After confirming that in the absence of lateral interactions the kMC and MKM approaches yield identical trends and mechanistic information, we observed substantial differences between the two kinetic models when lateral interactions were introduced. Here, the mean-field implementation applies coverage corrections directly to the descriptors, causing an artificial overprediction of the activity of strongly binding metals. In contrast, the cluster expansion in kMC implementation can differentiate among the highly active metals but it is very sensitive to the set of included interaction parameters. Considering that computational screening relies on a minimal set of descriptors, for which MKM makes reasonable trend predictions at a ca. three orders of magnitude lower computational cost than kMC, the MKM approach does provide a better entry point for computational catalyst design.

36 MATERIALS SCIENCE↗

Exploratory analysis and performance prediction of big data transfer in High-performance Networks

Big data transfer in large-scale scientific and business applications is increasingly carried out over connections with guaranteed bandwidth provisioned in High-performance Networks (HPNs) via advance bandwidth reservation. Provisioning agents need to carefully schedule data transfer requests, compute network paths, and allocate appropriate bandwidths. Such reserved bandwidths, if not fully utilized, could be simply wasted due to the exclusive access during the approved time window, and cause extra overhead and complexity for resource management. This calls for accurate performance prediction to reserve bandwidths that match actual needs and avoid over-provisioning. We employ machine learning algorithms to predict big data transfer performance based on extensive performance measurements collected in the past several years from data transfer tests using different protocols and toolkits between various end sites on several real-life physical or emulated testbeds. We first analyze the performance patterns in response to a comprehensive list of parameters in end-host systems, network connections, and data transfer applications, which motivate the use of machine learning and also help us identify the effects of latent factors. We then propose threshold- and clustering-based methods to eliminate negative effects of latent factors in data preprocessing and build a robust performance predictor based on customized domain-oriented loss functions. The performance of the proposed methods is verified by extensive experiments using SVR and RFR as well as theoretical analysis of the general performance bound.

97 MATHEMATICS AND COMPUTING↗

Characterization of Sodium Ion Electrochemical Reaction with Tin Anodes: Experiment and Theory

Tin anodes show a rich structure and reaction chemistry which we have investigated in detail. Upon discharge five plateaus are observed corresponding to -Sn, an unidentified phase (Na/Sn = 0.6), an amorphous phase (Na/Sn = 1.2), a hexagonal R-3m Na5Sn2, and fully sodiated Na15Sn 4. With charging there are six plateaus, which are related to the formation of Na5Sn2 followed by the formation of amorphous phases and -Sn. Upon cycling the formation of metastable Na5Sn2 seems to be suppressed, leading to a single charge plateau at 0.2 V. Theoretical voltages calculated from existing crystal structures using DFT provide a good match with constant current measurements, however, the voltage is more negative compared to quasi-equilibrium measurements (GITT). Search for additional (meta)stable phases using cluster-expansion method predicts many phases lower in energy than the convex hull, including the R-3m Na5Sn2 phase characterized experimentally. The presence of multiple phases in varying lattices with very similar formation energy suggests why the reaction mechanism is non-reversible. Interpretation of M ssbauer spectroscopy data is not yet elucidated due to the very low recoil-free fraction of the materials. The electrode surface is terminated with a SEI layer rich in carbonates such as Na2CO3 and Na alkyl carbonates as evidenced by XPS. After a full charge at 2V, strong evidence for the formation of oxidized Sn4+ is obtained. Subjecting the electrode to a rest after a full charge at 2 V reveals that aging in the electrolyte reduces the oxidized Sn4+ into Sn2+ and Sn0, and concomitantly suppresses the electrolyte decomposition represented by an anomalous discharge plateau at 1.2 V. Thereby, the catalytic decomposition of the electrolyte during discharge is caused by nanosized Sn particles covered by oxidized Sn4+ and not by pure metallic Sn.

Na5Sn2 metastable phase (XRD)↗

Low-lying states and total internal partition sums of CH

The electronic structure and spin-orbit states of the CH radical have been systematically investigated using multi-reference configuration interaction (MRCI) and single-reference coupled-cluster (CC) methods. These calculations were performed in conjunction with large correlation-consistent basis sets of quadruple-, quintuple-, and sextuple-ζ quality. To achieve high accuracy, electronic energies for all states were extrapolated to the complete basis set (CBS) limit, enabling the detailed construction of potential energy curves and determination of reliable spectroscopic constants. Spin-orbit coupling effects were explicitly incorporated, and vibrational energy levels were computed via Numerov analysis. Furthermore, the resulting values exhibit good to excellent agreement with available experimental data. Dipole moment and transition dipole moment curves were evaluated to assess the opacity characteristics of CH, revealing that transitions such as Χ 2 Π (u′′ = 0) → Α 2 Δ (u′ = 0), Χ 2 Π (u′′ = 0) → Β 2 Σ − (u′ = 0), Χ 2 Π (u′′ = 0) → C 2 Σ + (u′ = 0), and Χ 2 Π (u′′ = 0) → D 2 Σ + (u′ = 3) are particularly probable. Finally, the total internal partition function sum (TIPS) of CH was computed over a broad temperature range (10–30,000 K) based on our high-accuracy ab initio results.

74 ATOMIC AND MOLECULAR PHYSICS↗

Nanoparticles of La{sub 0.8}Ca{sub 0.2}Fe{sub 0.8}Ni{sub 0.2}O{sub 3-{delta}} perovskite for solid oxide fuel cell application

Polycrystalline samples of La{sub 0.8}Ca{sub 0.2}Fe{sub 0.8}Ni{sub 0.2}O{sub 3-{delta}} (LCFN) with perovskite type structure have been prepared by combustion, freeze drying, citrate-gel process and liquid mix method. The analysis of X-ray powder diffraction indicated that the samples were single phase and crystallized in an orthorhombic (space group, Pnma no. 62) structure. Transmission electron microscopy (TEM) analysis on the synthesized powder at 600 {sup o}C by liquid mix method showed clusters of 150 nm formed by nanoparticles of 20 nm. Electrochemical performance of LCFN cathodes, which are used for intermediate temperature solid oxide fuel cells, were investigated. The polarization resistance was studied using two different electrolytes: Y-doped zirconia (YSZ) and Sm-doped ceria (SDC). The dc four-probe measurement exhibits a total electrical conductivity, over 100 S cm{sup -1} at T {>=} 600 {sup o}C, pointing out that strontium can be substituted for the cheaper calcium cation without destroying the electrochemical properties. Experimental results indicate that nanoparticles have more advantages in terms of smaller particle size and better electrochemical performance.

36 MATERIALS SCIENCE↗

Reducing computation in an i-vector speaker recognition system using a tree-structured universal background model

The majority of state-of-the-art speaker recognition systems (SR) utilize speaker models that are derived from an adapted universal background model (UBM) in the form of a Gaussian mixture model (GMM). This is true for GMM supervector systems, joint factor analysis systems, and most recently i-vector systems. In all of the identified systems, the posterior probabilities and sufficient statistics calculations represent a computational bottleneck in both enrollment and testing. We propose a multi-layered hash system, employing a tree-structured GMM–UBM which uses Runnalls’ Gaussian mixture reduction technique, in order to reduce the number of these calculations. Moreover, with this tree-structured hash, we can trade-off reduction in computation with a corresponding degradation of equal error rate (EER). As an example, we also reduce this computation by a factor of 15× while incurring less than 10% relative degradation of EER (or 0.3% absolute EER) when evaluated with NIST 2010 speaker recognition evaluation (SRE) telephone data.

42 ENGINEERING↗

First-Principles Simulation of the (Li–Ni–Vacancy)O Phase Diagram and Its Relevance for the Surface Phases in Ni-Rich Li-Ion Cathode Materials

Despite several reports on the surface phase transformations from a layered to a disordered spinel and a rock-salt structure at the surface of the Ni-rich cathodes, the precise structures and compositions of these surface phases are unknown. The phenomenon, in itself, is complex and involves the participation of several contributing factors. Of these factors, transition metal (TM) ion migration toward the interior of the particle and hence formation of TM-densified surface layers, triggered by oxygen loss, is thermodynamically probable. Here in this study, we simulate the thermodynamic phase equilibria as a function of TM ion content in the cathode material in the context of lithium nickel oxides, using a combined approach of first-principles density functional calculations, the cluster expansion method, and grand canonical Monte Carlo simulations. We developed a unified lattice Hamiltonian that accommodates not only rock-salt like structures but also topologically different spinel-like structures. Also, our model provides a foundation to investigate metastable cation compositions and kinetics of the phase transformations. Our investigations predict the existence of several Ni-rich phases that were, to date, unknown in the scientific literature. Our simulated phase diagrams at finite temperature show a very low solubility range of the prototype spinel phase. Finally, we find a partially disordered spinel-like phase with far greater solubility that is expected to show very different Li diffusivity compared to that of the prototype spinel structure.

36 MATERIALS SCIENCE↗

Ab Initio Kinetics of Hydrogen Abstraction from Methyl Acetate by Hydrogen, Methyl, Oxygen, Hydroxyl, and Hydroperoxy Radicals

The kinetics of hydrogen abstraction by five radicals (H, O( 3 P), OH, CH 3 , and HO 2 ) from methyl acetate (MA) is investigated theoretically in order to gain further understanding of certain aspects of the combustion chemistry of biodiesels, such as the effect of the ester moiety. Here, we employ ab initio quantum chemistry methods, coupled cluster singles and doubles with perturbative triples correction (CCSD(T)) and multireference averaged coupled pair functional theory (MRACPF2), to predict chemically accurate reaction energetics. Overall, MRACPF2 predicts slightly higher barrier heights than CCSD(T) for MA + H/CH 3 /O/OH, but slightly lower barrier heights for hydrogen abstraction by HO 2 . Based on the obtained reaction energies, we also report high-pressure-limit rate constants using transition state theory (TST) in conjunction with the separable-hindered-rotor approximation, the variable reaction coordinate TST, and the multi-structure all-structure approach. The fitted modified Arrhenius expressions are provided over a temperature range of 250 to 2000 K. The predictions are in good agreement with available experimental results. Abstractions from both of the methyl groups in MA are expected to contribute to consumption of the fuel as they exhibit similar rate coefficients. Moreover, the reactions involving the OH radical are predicted to have the highest rates among the five abstracting radicals, while those initiated by HO 2 are expected to be the lowest.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Electronic Structure and Dynamics of Copper-Doped Indium Phosphide Nanocrystals Studied with Time-Resolved X-ray Absorption and Large-Scale DFT Calculations

Here, the recent development of time-resolved X-ray absorption spectroscopy (TR-XAS) has opened avenues for fundamental research on doped semiconductors. Reported herein are results on the successful doping of indium phosphide nanocrystals (NCs) with copper and subsequent charactenzation with time-resolved X-ray studies and largescale density functional theory (DFT) calculations. The synthetic protocol that was based on the cluster seed method afforded control of both the doping level and the quantity of NCs produced in a batch. The stability of the colloidal dispersion allowed us to monitor the photophysical properties of copper guest ions with static and TR-XAS at the copper K-edge. These data reveal that copper captures a hole from the photoexcited InP NC, and relaxation to the ground state occurs via multiple timescales likely due to the presence of interior and surface-bound dopant ions as well as recombination with surfaceand defect-trapped electrons. Large-scale DFT calculations reveal a striking dependence of the electronic structure on the ligand coating and other effects such as self-compensation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Automated discovery and construction of surface phase diagrams using machine learning

Surface phase diagrams are necessary for understanding surface chemistry in electrochemical catalysis, where a range of adsorbates and coverages exist at varying applied potentials. These diagrams are typically constructed using intuition, which risks missing complex coverages and configurations at potentials of interest. More accurate cluster expansion methods are often difficult to implement quickly for new surfaces. We adopt a machine learning approach to rectify both issues. Using a Gaussian process regression model, the free energy of all possible adsorbate coverages for surfaces is predicted for a finite number of adsorption sites. Our result demonstrates a rational, simple, and systematic approach for generating accurate free-energy diagrams with reduced computational resources. Finally, the Pourbaix diagram for the IrO 2 (110) surface (with nine coverages from fully hydrogenated to fully oxygenated surfaces) is reconstructed using just 20 electronic structure relaxations, compared to approximately 90 using typical search methods. Similar efficiency is demonstrated for the MoS 2 surface.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗