Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Predictive analytics”

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 1,477 records · Page 82

An Analytical Methodology to Evaluate Thermal-Hydraulic Performance of Compact Heat Exchangers, Accounting for Heat Loss

High efficiency and compactness, offered by the Printed Circuit Heat Exchanger (PCHE) makes it a strong candidate for use as an Intermediate Heat Exchanger (IHX) and/or Secondary Heat Exchanger (SHX) for Advanced High Temperature Reactor (AHTR) systems and also the Very High Temperature Reactor (VHTR). Idaho National Laboratory (INL) has been developing a computer code for the PCHE design and analysis [1], and will construct an experimental loop to evaluate the thermal hydraulic performance of PCHEs [2]. In order to correctly compare experimental results with the simulations, the extraneous heat losses from the PCHEs must be determined and incorporated. These heat exchangers operate at high temperature and consequently, despite the use of external thermal insulation, experience non-negligible heat losses to the surroundings. The objective of this work is to develop an analytical methodology for quantification of these heat losses based on experimental data to evaluate and predict the performance of the heat exchanger accurately.

99 GENERAL AND MISCELLANEOUS↗

Multisensor Agile Adaptive Sampling of Convective Storms Driven by Real-time Analytics

Convective storms vertically transport water vapor and condensate from Earth’s surface to the upper troposphere. Life on Earth is fundamentally linked to this transport which determines the hydrological cycle, and the intensity of severe weather responsible for the destruction of life and property. Despite advances in high-resolution modeling and better observational capabilities, the scientific community continues to be confronted with knowledge gaps about convective storms that limit our predictive capabilities. The ongoing developments in the high-resolution Energy Exascale Earth System Model (E3SM), large eddy simulations, and AI-based analytics to evaluate uncertainties are expected to provide a comprehensive framework for new scientific discovery. The model-experiment (MODEX) approach suggests that the aforementioned advancements in model development and AI-based inference techniques should be complemented by similar advancements in the experimental (observational) side so that the former does not outstrip the ability of the latter to provide meaningful constraints. What are the recent advancements in observations that will provide the necessary leap forward in improving our predictive capabilities? To address this question, we propose a new experimental paradigm called Multisensor Agile Adaptive Sampling (MAAS) that capitalizes on advancements in communications (5G), computational resources (edge/fog computing), sensor capabilities, and machine learning (ML) and AI techniques (Kollias et al., 2020). The MAAS framework allows for the collection of higher spatiotemporal resolution and quality observations of convective storms than is traditionally possible. The MAAS framework is scalable and applicable to atmospheric observatories such as those operated by the Department of Energy (DoE) Atmospheric Radiation Measurement (ARM) facility.

54 ENVIRONMENTAL SCIENCES↗

Predicting battery capacity from impedance at varying temperature and state of charge using machine learning

Prediction of battery health from electrochemical impedance spectroscopy (EIS) data can enable rapid measurement of battery state in real-world applications without using additional sensors or time-consuming performance measurements. However, deconvoluting the effect of capacity, state of charge, and temperature on EIS response is complicated analytically. Here, various machine-learning models, such as linear, Gaussian process, random forest, and artificial neural network regression, are utilized to predict capacity from EIS using hundreds of capacity, direct current (DC) resistance, and EIS measurements recorded under varying conditions of health, temperature, and state of charge (SOC). Several feature extraction and selection methods from traditional electrochemical analysis and statistical modeling are explored using machine-learning pipelines. EIS data from just two frequencies can accurately predict capacity, and interrogation shows that the optimal set of frequencies is not usually intuitive. Best results are achieved with an ensemble model, which predicts battery capacity with a mean absolute error of 1.9% on data from unobserved cells.

25 ENERGY STORAGE↗

Simultaneously improving accuracy and computational cost under parametric constraints in materials property prediction tasks

Abstract Modern data mining techniques using machine learning (ML) and deep learning (DL) algorithms have been shown to excel in the regression-based task of materials property prediction using various materials representations. In an attempt to improve the predictive performance of the deep neural network model, researchers have tried to add more layers as well as develop new architectural components to create sophisticated and deep neural network models that can aid in the training process and improve the predictive ability of the final model. However, usually, these modifications require a lot of computational resources, thereby further increasing the already large model training time, which is often not feasible, thereby limiting usage for most researchers. In this paper, we study and propose a deep neural network framework for regression-based problems comprising of fully connected layers that can work with any numerical vector-based materials representations as model input. We present a novel deep regression neural network, iBRNet, with branched skip connections and multiple schedulers, which can reduce the number of parameters used to construct the model, improve the accuracy, and decrease the training time of the predictive model. We perform the model training using composition-based numerical vectors representing the elemental fractions of the respective materials and compare their performance against other traditional ML and several known DL architectures. Using multiple datasets with varying data sizes for training and testing, We show that the proposed iBRNet models outperform the state-of-the-art ML and DL models for all data sizes. We also show that the branched structure and usage of multiple schedulers lead to fewer parameters and faster model training time with better convergence than other neural networks. Scientific contribution: The combination of multiple callback functions in deep neural networks minimizes training time and maximizes accuracy in a controlled computational environment with parametric constraints for the task of materials property prediction.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

An analytical coarse-graining method which preserves the free energy, structural correlations, and thermodynamic state of polymer melts from the atomistic to the mesoscale

Structural and thermodynamic consistency of coarse-graining models across multiple length scales is essential for the predictive role of multi-scale modeling and molecular dynamic simulations that use mesoscale descriptions. Our approach is a coarse-grained model based on integral equation theory, which can represent polymer chains at variable levels of chemical details. The model is analytical and depends on molecular and thermodynamic parameters of the system under study, as well as on the direct correlation function in the k → 0 limit, c{sub 0}. A numerical solution to the PRISM integral equations is used to determine c{sub 0}, by adjusting the value of the effective hard sphere diameter, d{sub HS}, to agree with the predicted equation of state. This single quantity parameterizes the coarse-grained potential, which is used to perform mesoscale simulations that are directly compared with atomistic-level simulations of the same system. We test our coarse-graining formalism by comparing structural correlations, isothermal compressibility, equation of state, Helmholtz and Gibbs free energies, and potential energy and entropy using both united atom and coarse-grained descriptions. We find quantitative agreement between the analytical formalism for the thermodynamic properties, and the results of Molecular Dynamics simulations, independent of the chosen level of representation. In the mesoscale description, the potential energy of the soft-particle interaction becomes a free energy in the coarse-grained coordinates which preserves the excess free energy from an ideal gas across all levels of description. The structural consistency between the united-atom and mesoscale descriptions means the relative entropy between descriptions has been minimized without any variational optimization parameters. The approach is general and applicable to any polymeric system in different thermodynamic conditions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A suite of benchmark and challenge problems for enhanced geothermal systems

A diverse suite of numerical simulators is currently being applied to predict or understand the performance of enhanced geothermal systems (EGS). To build confidence and identify critical development needs for these analytical tools, the United States Department of Energy, Geothermal Technologies Office sponsored a Code Comparison Study (GTO-CCS), with participants from universities, industry, and national laboratories. A principal objective for the study was to create a community forum for improvement and verification of numerical simulators for EGS modeling. Teams participating in the study were those representing U.S. national laboratories, universities, and industries, and each team brought unique numerical simulation capabilities to bear on the problems. Two classes of problems were developed during the study, benchmark problems and challenge problems. The benchmark problems were structured to test the ability of the collection of numerical simulators to solve various combinations of coupled thermal, hydrologic, geomechanical, and geochemical processes. This class of problems was strictly defined in terms of properties, driving forces, initial conditions, and boundary conditions. The challenge problems were based on the enhanced geothermal systems research conducted at Fenton Hill, near Los Alamos, New Mexico, between 1974 and 1995. The problems involved two phases of research, stimulation, development, and circulation in two separate reservoirs. The challenge problems had specific questions to be answered via numerical simulation in three topical areas: (1) reservoir creation/stimulation, (2) reactive and passive transport, and (3) thermal recovery. Whereas the benchmark class of problems were designed to test capabilities for modeling coupled processes under strictly specified conditions, the stated objective for the challenge class of problems was to demonstrate what new understanding of the Fenton Hill experiments could be realized via the application of modern numerical simulation tools by recognized expert practitioners. Lastly, we present the suite of benchmark and challenge problems developed for the GTO-CCS, providing problem descriptions and sample solutions.

15 GEOTHERMAL ENERGY↗

Analysis of steady compaction waves in polyurea aerogel

In this work, steady compaction waves in an inert porous material are investigated using a p - α model. In a steady traveling wave reference frame, the one-dimensional Euler equations are reduced to a set of ordinary differential equations. A Mie-Grüneisen equation of state (EOS) is used with parameters calibrated for polyurea aerogel (PUA). Analytic solutions for non-equilibrium compaction are developed which compliment numerical simulations and are able to predict the complete wave structure, including the compaction wave speed, zone length, and final compacted solid volume fraction. The dynamic compaction of PUA is studied for a range of piston velocities. Three regions of behavior are identified: supersonic, subsonic-complete, and subsonic-partial compaction. Below a critical piston velocity, a subsonic compaction wave is produced without a leading shock. Finally, at even lower piston velocities, there is partial compaction and a greater dependence on the dynamic compaction relation. Some features and limitations of the current model are discussed.

36 MATERIALS SCIENCE↗

Social Sensor Analytics: Making Sense of Network Models in Social Media

Social networks can be thought of as noisy sensor networks mapping real world information to the web. Owing to the extensive body of literature in sensor network analysis, this work sought to apply several novel and traditional methods in sensor network analysis for the purposes of efficiently interrogating social media data streams from raw data. We carefully revisit our definition of a social media signal from previous work both in terms of time-varying features within the data and the networked nature of the medium. Further, we detail our analysis of global patterns in Twitter over the months of November 2013 and June 2014, detect and categorize events, and illustrate how these analyses can be used to inform graph-based models of Twitter, namely using a recent network influence model called PhySense: similar to PageRank but tuned to behavioral analysis by leveraging a sociologically inspired probabilistic model. We ultimately identify forms of information dissemination via analysis of time series and dynamic graph spectra and corroborate these findings through manual investigation of the data as a requisite step in modeling the diffusion process with PhySense. We hope to sufficiently characterize global behavior in a medium such as Twitter as a means of learning global model parameters one may use to predict or simulate behavior on a large scale. We have made our time series and dynamic graph analytical code available via a GitHub repository https://github.com/cpatdowling/salsa and our data are available upon request.

event detection↗

K-ε Turbulence Model Parameter Estimates Using an Approximate Self-similar Jet-in-Crossflow Solution

The k-ε turbulence model has been described as perhaps “the most widely used complete turbulence model.” This family of heuristic Reynolds Averaged Navier-Stokes (RANS) turbulence closures is supported by a suite of model parameters that have been estimated by demanding the satisfaction of well-established canonical flows such as homogeneous shear flow, log-law behavior, etc. While this procedure does yield a set of so-called nominal parameters, it is abundantly clear that they do not provide a universally satisfactory turbulence model that is capable of simulating complex flows. Recent work on the Bayesian calibration of the k-ε model using jet-in-crossflow wind tunnel data has yielded parameter estimates that are far more predictive than nominal parameter values. In this paper, we develop a self-similar asymptotic solution for axisymmetric jet-in-crossflow interactions and derive analytical estimates of the parameters that were inferred using Bayesian calibration. The self-similar method utilizes a near field approach to estimate the turbulence model parameters while retaining the classical far-field scaling to model flow field quantities. Our parameter values are seen to be far more predictive than the nominal values, as checked using RANS simulations and experimental measurements. They are also closer to the Bayesian estimates than the nominal parameters. A traditional simplified jet trajectory model is explicitly related to the turbulence model parameters and is shown to yield good agreement with measurement when utilizing the analytical derived turbulence model coefficients. Finally, the close agreement between the turbulence model coefficients obtained via Bayesian calibration and the analytically estimated coefficients derived in this paper is consistent with the contention that the Bayesian calibration approach is firmly rooted in the underlying physical description.

42 ENGINEERING↗

Using Artificial Neural Networks to Predict Physical Properties of Membrane Polymers

Membrane polymers are a promising technology for use in many challenging gas separation applications. Here, the techniques of computer-aided molecular design can be used to search through the massive molecular space of heteropolymers and develop a set of likely candidate repeat units matching specific physical property targets. However, reasonably accurate property prediction algorithms are needed, but these algorithms must be very fast in order to be combined with an optimization framework. Artificial neural networks (ANNs), a branch of machine learning, are applied in this work to predict the physical properties of polymers. All of the physical properties investigated were found to be predicted by ANNs with R 2 scores exceeding 0.82.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Defining Determinants and Dynamics and Cellulose Microfibril Biosynthesis, Assembly and Degredation OSP Number: 63079/A001

The central paradigm for converting plant biomass into soluble sugars for subsequent conversion to transportation fuels involves the enzymatic depolymerization of lignocellulosic plant cell walls by microbial enzymes. Despite decades of intensive research, this is still a relatively inefficient process, due largely to the recalcitrance and enormous complexity of the substrate. A major obstacle is still insufficient understanding of the detailed structure and biosynthesis of major wall components, including cellulose. For example, although cellulose is generally depicted as rigid, insoluble, uniformly crystalline microfibrils that are resistant to enzymatic degradation, the in vivo structures of plant cellulose microfibrils are surprisingly complex. Crystallinity is frequently disrupted, for example by dislocations and areas containing chain ends, resulting in “amorphous” disordered regions. Importantly, microfibril structure and the relative proportions of crystalline and non-crystalline disordered surface regions vary substantially and yet the molecular mechanisms by which plants regulate microfibril crystallinity, and other aspects of microfibril architecture, are still entirely unknown. This obviously has a profound effect on susceptibility to enzymatic hydrolysis and so this is a critical area of research in order to characterize and optimize cellulosic biomass degradation. The entire field of cell wall assembly, as distinct from polysaccharide biosynthesis, and the degree to which they are coupled, are relatively unexplored, despite the great potential for major advances in addressing the hurdle of biomass recalcitrance. Our overarching hypothesis was that identification of the molecular machinery that determine microfibril polymerization, deposition and structure will allow the design of more effective degradative systems, and the generation of cellulosic materials with enhanced and predictable bioconversion characteristics. Our experimental framework had been based on the idea that the most effective way to address this long standing and highly complex question is to adopt a broad ‘systems approach’. Accordingly, we assembled a multi-disciplinary collaborative team with collective expertise in plant biology and molecular genetics, polymer structure and chemistry, enzyme biochemistry and biochemical engineering. We used a spectrum of cutting edge technologies, including plant functional genomics, chemical genetics, live cell imaging, advanced microscopy, high energy X-ray spectroscopy and nanotechnology, to study the molecular determinants of cellulose microfibril structure. Importantly, this research effort was closely coupled with an analytical pipeline to characterize the effects of altering microfibril architecture on bioconversion potential, with the goal of generating predictive models to help guide the identification, development and implementation of new feedstocks. This project therefore spanned core basic science and applied research, in line with the goals of the program. Over the course of the project, accomplishments included: - Establishing platforms through reverse and forward genetics to identify and manipulate candidate genes that influence cellulose microfibril synthesis and structure in a model C3 grass, Brachypodium distachyon and a model C4 grass Setaria viridis; Identifying and characterizing the effects of a number of cellulose biosynthesis inhibitors (CBIs), and particularly those that target monocots with the aim of generating resistance loci; Developing protocols for the use of high energy X-ray diffraction (XRD) to study the structure and organization of cellulose microfibrils in plant walls, notably those in Arabidopsis and Brachypodium; Using the chemical and genetic based inhibition strategies to develop new mechanistic models of cellulose microfibril crystallization, and of how altering microfibril architecture influences digestibility.

09 BIOMASS FUELS↗

Comprehensive End-to-End Design of Novel High Energy Density Materials: II. Computational Modeling and Predictions

In this work, we have proposed a holistic approach to design novel energetic materials by bridging synthesis, experimental characterization, computational modeling, and validation. Multiscale computational modeling that combines first-principles calculations, analytical theory, and empirical statistical analysis served to further advance the proposed methodology. The established materials design guiding principles led to development of a set of new energetic molecules, PHE-1, PHE-2, and PHE-3, that represent improved variations of the heterocyclic energetics and are predicted to be superior to the existing conventional energetic materials. Molecular mechanisms of the enhanced performance and sensitivity of the proposed energetic materials as a function of their chemical composition and structure are discussed.

36 MATERIALS SCIENCE↗

A molecularly based theory for electron transfer reorganization energy

Using field-theoretic techniques, we develop a molecularly based dipolar self-consistent-field theory (DSCFT) for charge solvation in pure solvents under equilibrium and nonequilibrium conditions and apply it to the reorganization energy of electron transfer reactions. The DSCFT uses a set of molecular parameters, such as the solvent molecule’s permanent dipole moment and polarizability, thus avoiding approximations that are inherent in treating the solvent as a linear dielectric medium. A simple, analytical expression for the free energy is obtained in terms of the equilibrium and nonequilibrium electrostatic potential profiles and electric susceptibilities, which are obtained by solving a set of self-consistent equations. With no adjustable parameters, the DSCFT predicts activation energies and reorganization energies in good agreement with previous experiments and calculations for the electron transfer between metallic ions. Because the DSCFT is able to describe the properties of the solvent in the immediate vicinity of the charges, it is unnecessary to distinguish between the inner-sphere and outer-sphere solvent molecules in the calculation of the reorganization energy as in previous work. Furthermore, examining the nonequilibrium free energy surfaces of electron transfer, we find that the nonequilibrium free energy is well approximated by a double parabola for self-exchange reactions, but the curvature of the nonequilibrium free energy surface depends on the charges of the electron-transferring species, contrary to the prediction by the linear dielectric theory.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Differences in S/G ratio in natural poplar variants do not predict catalytic depolymerization monomer yields

The ratio of syringyl (S) and guaiacyl (G) units in lignin has been noticed as a major factor in determining the maximum monomer yield from lignin depolymerization. This limit arises from the notion that G units are prone to C-C bond formation during lignin biosynthesis, resulting in less ether linkages that generate monomers. This study uses reductive catalytic fractionation (RCF) in flow-through reactors as an analytical tool to depolymerize lignin in poplar with naturally varying S/G ratios, and directly challenges the common conception that the S/G ratio predicts monomer yields. Rather, this work suggests that the plant controls C-O and C-C bond content by regulating monomer transport during lignin biosynthesis. In general, our results indicate that additional factors beyond the monomeric composition of native lignin are important in developing a fundamental understanding of lignin biosynthesis.

09 BIOMASS FUELS↗

Tunable Spectral Ordering of Magnetic Plasmon Resonances in Noble Metal Nanoclusters

Experimental characterization of the optical and magnetic properties of noble metal nanoclusters has exposed a size regime in between single-ring and extended two-dimensional nanocluster networks where the behavior of magnetic plasmon resonances is not well understood. In this intermediate size regime, individual electric dipole plasmons on each nanoparticle within the cluster can hybridize into delocalized magnetic modes that couple to either the electric or magnetic field of light with seemingly arbitrary energy ordering. In this work, using a coupled-dipole model that includes fully retarded interactions between electric dipole plasmons, we show that magnetic plasmon resonance energies can be controllably tuned and even made to cross as a function of nanocluster size. Experimental confirmation of this prediction is challenging because optical selection rules dictate the simultaneous excitation of many spectrally overlapping magnetic plasmon modes. However, based on analytic modeling and numerical simulation, we show that angle-resolved cathodoluminescence offers an approach to cleanly isolate these magnetic modes and observe the size-dependence of their spectrum. Not only does our work clarify the rich optical and magnetic behavior of noble metal nanoclusters in the intermediate-size regime, but it also suggests strategies to design negative-index plasmonic metamaterials with multiple tunable resonances in the visible spectrum.

36 MATERIALS SCIENCE↗

Theory of attosecond transient absorption spectroscopy of strong-field-generated ions

Strong-field ionization generally produces ions in a superposition of ionic eigenstates. This superposition is generally not fully coherent and must be described in terms of a density matrix. A recent experiment [E. Goulielmakis et al., Nature (London) 466, 739 (2010)] employed attosecond transient absorption spectroscopy to determine the density matrix of strong-field-generated Kr{sup +} ions. The experimentally observed degree of coherence of the strong-field-generated Kr{sup +} ions is well reproduced by a recently developed multichannel strong-field-ionization theory, but there is significant disagreement between experiment and theory with respect to the degree of alignment of the Kr{sup +} ions. In the present paper, the theory underlying attosecond transient absorption spectroscopy of strong-field-generated ions is developed. The theory is formulated in such a way that the nonperturbative nature of the strong-field-ionization process is systematically taken into account. The impact of attosecond pulse propagation effects on the interpretation of experimental data is investigated both analytically and numerically. It is shown that attosecond pulse propagation effects cannot explain why the experimentally determined degree of alignment of strong-field-generated Kr{sup +} ions is much smaller than predicted by existing theory.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A Prognostics and Health Management Framework for Wind

Operation and maintenance costs are a major driver for levelized cost of energy of wind power plants and can be reduced through optimized operation and maintenance practices accomplishable by various prognostics and health management (PHM) technologies. In recent years, the wind industry has become more open to adopting PHM solutions, especially those focusing on diagnostics. However, prognostics activities are, in general, still at the research and development stage. On the other hand, the industry has a request to estimate a component's remaining useful life (RUL) when it has faulted, and this is a key output of prognostics. Systematically presenting PHM technologies to the wind industry by highlighting the RUL prediction need potentially helps speed up its acceptance and provides more benefits from PHM to the industry. In this paper, we introduce a PHM for wind framework. It highlights specifics unique to wind turbines and features integration of data and physics domain information and models. The output of the framework focuses on RUL prediction. To demonstrate its application, a data domain method for wind turbine gearbox fault diagnostics is presented. It uses supervisory control and data acquisition system time series data, normalizes gearbox temperature measurements with reference to environmental temperature and turbine power, and leverages big data analytics and machine-learning techniques to make the model scalable and the diagnostics process automatic. Another physics-domain modeling method for RUL prediction of wind turbine gearbox high-speed-stage bearings failed by axial cracks is also discussed. Bearing axial cracking has been shown to be the prevalent wind turbine gearbox failure mode experienced in the field and is different from rolling contact fatigue, which is targeted during the bearing design stage. The method uses probability of failure as a component reliability assessment and RUL prediction metric, which can be expanded to other drivetrain components or failure modes. The presented PHM for wind framework is generic and applicable to both land-based and offshore wind turbines.

17 WIND ENERGY↗

NREL Develops Diagnostic Test Cases to Improve Building Energy Simulation Programs (Fact Sheet)

This technical highlight describes NREL research to develop a set of diagnostic test cases for building energy simulations in order to achieve more accurate energy use and savings predictions. The National Renewable Energy Laboratory (NREL) Residential and Commercial Buildings research groups developed a set of diagnostic test cases for building energy simulations. Eight test cases were developed to test surface conduction heat transfer algorithms of building envelopes in building energy simulation programs. These algorithms are used to predict energy flow through external opaque surfaces such as walls, ceilings, and floors. The test cases consist of analytical and vetted numerical heat transfer solutions that have been available for decades, which increases confidence in test results. NREL researchers adapted these solutions for comparisons with building energy simulation results. Testing the new cases with EnergyPlus identified issues with the conduction finite difference (CondFD) heat transfer algorithm in versions 5 and 6. NREL researchers resolved these issues for EnergyPlus version 7. The new test cases will help users and developers of EnergyPlus and other building energy tools to identify and fix problems associated with solid conduction heat transfer algorithms of building envelopes and their boundary conditions. In the long term, improvements to software algorithms will result in more accurate energy use and savings predictions. NREL researchers plan to document the set of test cases and make them available for future consideration by validation standards such as ASHRAE Standard 140: Standard Method of Test for the Evaluation of Building Energy Analysis Computer Programs. EnergyPlus users will also have access to the improved CondFD model in version 7 after its next scheduled release.

14 SOLAR ENERGY↗