Ocean Compressed Air Energy Storage (OCAES)
Explore the source record for details and available documents.
SEARCH · Search NASA
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.
Explore the source record for details and available documents.
The farnesoid X receptor (FXR) is a ligand-activated transcription factor that plays important roles in regulating bile acid homeostasis. The aim of the present study was to investigate the effects of obeticholic acid (OCA), a novel synthetic FXR agonist, carbon tetrachloride (CCl{sub 4})-induced acute liver injury. Mice were intraperitoneally injected with CCl{sub 4} (0.15 ml/kg). In CCl{sub 4} + OCA group, mice were orally with OCA (5 mg/kg) 48, 24 and 1 h before CCl{sub 4}. As expected, hepatic FXR was activated by OCA. Interestingly, OCA pretreatment alleviated CCl{sub 4}-induced elevation of serum ALT and hepatic necrosis. Moreover, OCA pretreatment inhibited CCl{sub 4}-induced hepatocyte apoptosis. Additional experiment showed that OCA inhibits CCl{sub 4}-induced hepatic chemokine gene Mcp-1, Mip-2 and Kc. Moreover, OCA inhibits CCl{sub 4}-induced hepatic pro-inflammatory gene Tnf-α and Il-1β. By contrast, OCA pretreatment elevated hepatic anti-inflammatory gene Il-4. Further analysis showed that OCA pretreatment inhibited hepatic IκB phosphorylation and blocked nuclear translocation of NF-κB p65 and p50 subunits during CCl{sub 4}-induced acute liver injury. In addition, OCA pretreatment inhibited hepatic Akt, ERK and p38 phosphorylation in CCl{sub 4}-induced acute liver injury. These results suggest that OCA protects against CCl{sub 4}-induced acute liver injury and inflammation. Synthetic FXR agonists may be effective antidotes for hepatic inflammation during acute liver injury. - Highlights: • OCA pretreatment activates hepatic FXR. • FXR activation protects against CCl{sub 4}-induced acute liver injury. • FXR activation inhibits hepatocyte apoptosis during CCl{sub 4}-induced liver injury. • FXR activation differentially regulates hepatic inflammatory genes. • Synthetic FXR agonists are effective antidotes for acute liver injury.
Hybrid halide double perovskites are a class of compounds attracting growing interest due to their richness of structure and property. Two-dimensional (2D) derivatives of hybrid double perovskites are formed by the incorporation of organic spacer cations into three-dimensional (3D) double perovskites. Here, we report a series of seven new layered double perovskite halides with propylammonium (PA), octylammonium (OCA), and 1,4-butyldiammonium (BDA) cations. The general formulas of the compounds are A m M I M III X 8 (single-layered Ruddlesden–Popper type with m = 4 and A = PA or OCA, and single-layered Dion–Jacobson type with m = 2 and A = BDA, M I = Ag, M III = In or Bi, X = Cl or Br) and PA 2 Cs M I M III Br 7 (bilayered, with M I = Ag, M III = In or Bi). These families of compounds demonstrate significant versatility, with tunable layer thickness, the ability to vary the interlayer spacing, and the ability to selectively tune the band gap by varying the M I and M III cations along with the halide anions. The band gap of the single-layered materials varies from 2.41 eV for PA 4 AgBiBr 8 to 3.96 eV for PA 4 AgInCl 8 . Photoluminescent emission spectra of the layered double perovskites at low-temperature (100 K) are reported, and density functional theory electronic structure calculations are presented to understand the nature of the band gap evolution. The development of new structural and compositions in layered double perovskite halides enhances the understanding of structure–property relations in this important family.
We present the results of a comparative analysis of optical immersion clearing of skin in laboratory animals in vivo with and without preliminary ablation of epidermis. Laser ablation is implemented using a setup based on a pulsed erbium laser ( λ = 2940 nm). The size of the damaged region amounted to 6 × 6 mm, the depth being smaller than 50 μm. As an optical clearing agent (OCA), use is made of polyethylene glycol (PEG-300). Based on optical coherence tomography, we use the single scattering model to estimate the scattering coefficient in the process of optical clearing in 2 regions at depths of 50−170 μm and 150−400 μm. The results show that skin surface ablation leads to the local oedema of the affected region that increases the scattering coefficient. However, the intense evaporation of water from the ablation zone facilitates the optical clearing at the expense of tissue dehydration, particularly in the upper layers. The assessment of the optical clearing efficiency shows that the efficiency exceeding 30% can be achieved at a depth from 50 to 170 μm in 120 min after ablation, as well as after the same ablation with subsequent application of PEG-300, which increases the efficiency of the immersion method by almost 1.8 times. At a depth from 150 to 400 μm, dehydration of upper layers cannot completely compensate for an increase in light scattering by dermis after epidermis ablation. The additional effect of OCA enhances the optical clearing of skin at the expense of improving the refractive index matching between dermis components, but the maximal efficiency of optical clearing in 120 min does not exceed 6%. (biophotonics)
In this study, we investigate the normal state of the superconducting compound PuCoGa 5 using the combination of density functional theory (DFT) and dynamical mean-field theory (DMFT), with the continuous time quantum Monte Carlo (CTQMC) and the vertex-corrected one-crossing approximation (OCA) as the impurity solvers. Our DFT+DMFT (CTQMC) calculations suggest a strong tendency of Pu-5f orbitals to differentiate at low temperatures. The renormalized 5f 5/2 states exhibit a Fermi-liquid behavior whereas one electron in the 5f 7/2 states is at the edge of a Mott localization. We find that the orbital differentiation is manifested as the removing of 5f 7/2 spectral weight from the Fermi level relative to DFT. Finally, we corroborate these conclusions with DFT+DMFT (OCA) calculations which demonstrate that 5f 5/2 electrons have a much larger Kondo scale than the 5f 7/2 .
Over the past decade, techno-economical advancements in solar energy systems, particularly the improvement in the conversion efficiency of PV modules made with mono-crystalline silicon solar cells from 12% to 20%, as well as the cost reduction in manufacturing by a factor of about 10%, have made it possible to achieve a levelized cost of electricity (LCOE) in PV plants that is comparable to, or less than, the cost of deriving electricity from fossil fuels. Operating PV plants in mid-latitude sun-belt regions, where the solar irradiance level is highest, provides a high annual energy-yield (kWh/kWp) due to two factors: (1) the availability of predictable high solar irradiance throughout the year with the fewest interruptions in solar flux from clouds and rain, and (2) the increased conversion efficiency of crystalline solar cells, as recombination loss has decreased with increased intensity of the sunlight that illuminates the silicon solar cells. Semi-arid and desert regions, however, are plagued by high atmospheric dust concentrations and frequent sand storms. The deposition of a layer of dust on the optical surfaces of solar collectors such as PV modules and concentrating mirrors reduces the transmission efficiency of sunlight that actually reaches the solar cells or receivers, resulting in high energy-yield soiling loss. There are two major cost components to operating a solar plant: (1) installation costs, and (2) operation-and-maintenance (O&M) costs. There is no fuel cost; hence operating a solar plant in a semi-arid or desert region provides high returns on investment if soiling losses are mitigated via efficient cleaning methods and optimized cleaning frequency. If solar collectors are not cleaned, the accumulation of dust layers on solar collectors may cause the operation of such plants in arid regions to become economically unviable. Washing solar collectors with water and detergent, as is most commonly done now, is an efficient method for cleaning. The conventional approach in utility-scale solar plants is to use a large truck with a water tank and pump system for spraying deionized water on the surface of the solar collectors. Robotic cleaning with brushes, used for many solar plants, requires lesser water for cleaning. The water consumed using semi-automated cleaning of PV modules in utility-scale solar plants is approximately 2 liters/m 2 per cleaning cycle. The total optical surface area of the solar collectors in 1 TW-scale solar installation will be more than 3 × 109 m 2 ; hence an enormous amount of water be needed for cleaning. There simply is not enough fresh water in the sun-belt areas of the world for predicted cleaning needs. In solar power plants, the estimated cost of cleaning solar collectors includes expenses related to the equipment used, labor, cost of transportation of water, the energy required for cleaning, plus ancillary costs whereas the cost of water is not considered. The water used is obtained from sources located close to plant sites, unmindful of the environmental and societal impact as the power plants are oftentimes located in regions that face severe drought. This practice is very similar to the cost calculations in deriving the levelized cost of electricity (LCOE) in conventional power plants based on burning fossil fuel such as coal and gas, while disregarding the cost of climate change and health effects. Unless a water-free or low-water cleaning method is established, the expansion of solar plants may lose public support in areas suffering long intervals of drought. The goal of this research project has been the development and application of the Electrodynamic Screen (EDS) as a means for a water-free, scalable cleaning process applicable to solar-power installations, including rooftop applications. We describe here the development of an EDS film-based cleaning process as an emerging method for use on PV modules, parabolic troughs, and heliostats. This report aims to show the feasibility of integrating or retrofitting EDS films onto the optical surfaces of solar collectors (both PV and CSP) while maintaining high transmission or reflection efficiency. The cleaning action provided by the EDS film is an active method to remove dust deposits by electrodynamic force. Current lab-scale prototype EDS films, retrofitted onto solar panels and mirrors, have shown to be capable of maintaining optical transmission or specular-reflection efficiencies higher than 90% of initial values under clean conditions. The optical surfaces of solar collectors laminated with EDS films can remove more than 90% of deposited dust when the EDS is activated for less than two minutes. As an electrodynamic dust removal process, the EDS film-based method is designed primarily for the removal of dust in solar installations located in semi-arid and desert areas, where the atmosphere is often dry and dusty and rainfall is infrequent. While EDS film application minimizes water consumption and facilitates cleaning as frequently as needed, it has limitations in removing contaminants such as soot, organic pollutants deposited as fine films on the surface, and bird droppings. The dust removal efficiency of the EDS is maximum at relative humidity RH is 40 to 50% and decreases at when RH > 65%. Many solar plant sites undergo diurnal and seasonal cycles of high ambient, early afternoon temperatures and an RH that reaches the dew point early in the morning. These variations in atmospheric conditions do not limit the operation of the EDS film. This report presents a brief review of the progress and the potential of EDS film technology for mitigating the impact of dust on solar collectors via water-free cleaning, as well as current technical challenges regarding efficiency and durability. Our experimental data on the performance of EDS films show that: (1) the dust-removal efficiency (DRE) can reach levels higher than 90%, (2) the specular reflectivity (SR) of EDS film-laminated second-surface mirrors reach levels in excess of 90%, (3) the specular reflectivity restoration (SRR) can exceed 90%, (4) the output-power restoration (OPR) of PV modules can exceed 95%, and (5) the optical transmission efficiency (TE) of the EDS films can be greater than 90%. Working with Sandia National Laboratories (NM), Corning Research and Development Corporation (NY), Eastman Kodak (NY), Tomark-Worthen Industries (NH), and EDS Chile SPA (Chile), we have produced EDS film-laminated PV modules and demonstrated their self-cleaning functions without requiring water. We have demonstrated that the operational range of EDS films will cover the expected ambient temperature of solar fields at RH cycling varying from 20 to 95% as long as the EDS films are activated in the RH range 20 to 50%. (Typical solar-field climates in deserts and semi-arid lands often reach near dew point in coastal areas.) Our experiments on the application of hydrophobic-fluorinated nanoparticle coatings on EDS film surfaces show that the EDS operational range can be extended to higher RH levels that approach the dew point. At Eastman Kodak, as one of our industrial partners, we were able to establish a process for manufacturing EDS films using flexographic printing of the electrodes onto transparent polymer films. This process utilizes an existing manufacturing line at Eastman Kodak that allows fabrication of medium-scale EDS films (26 cm × 30 cm). The manufacturing process has the capacity to produce EDS films at high production speeds. The medium-scale EDS films that have been printed at Kodak were evaluated in the lab at Boston University. The EDS films produced at Kodak were laminated at Tomark-Worthen using an industrial scale vacuum laminator to produce EDS film stacks, which can be affixed onto the optical surfaces of PV modules or concentrating mirrors. The EDS film stack consists of the EDS films that have Willow® Glass (WG) which has a thickness of 100 μm as the front surface. The WG sheets obtained from Corning Research and Development Corporation are customized in size and shape to cover the active area of the EDS films. The back surface of the EDS film stack is integrated onto the optical surface of the solar panel or mirror using optically clear adhesive (OCA) films or silicone adhesives. The OCA films (thickness 25 μm) are produced by 3M. The EDS film stacks have the architecture: WG/OCA/EDS Film/OCA/ over PV module or solar mirror. The power-supply units needed for activating the electrodes of the EDS film were designed and produced at Boston University. These power supply units provide three-phase, 1.2 kV voltage pulses at a very low current (micro-ampere) level and at a low frequency (≈ 5 Hz). The voltage pulses are applied to the electrodes in a sequence such that the train of pulses resembles a unidirectional traveling wave of electrical field on the surface of the EDS film. The dust particles on the surface become charged electrostatically and are levitated by the Coulomb force. The lateral sweeping action of the traveling electric field created by the three-phase voltages pulses then sweeps the dust off the surface. The energy consumed by the EDS electrodes is less than 0.2 Wh/m 2 /cleaning cycle, enabling energy-efficient restoration of output power (OPR) of PV module or specular reflectivity restoration (SRR) for solar mirrors. The EDS system consists of (1) an EDS film stack laminated onto the solar collectors, (2) connection of the EDS film to its power supply unit and (3) Interconnection of the power supply to PV modules or solar mirrors. Design, construction and assessment of field-testing units that have EDS stack laminated PV modules for evaluating the performance of the EDS films in solar fields is being carried out at BU. Our progress under this project, aimed at the advancement of EDS film technology, has reached DOE Technology Readiness Level (TRL) 6. Based on the extensive laboratory evaluations and limited field trials, as well as contacts with potential users, we believe that the technology has reached its commercial stage. We are conducting a cost analysis using the National Renewable Energy Laboratory (NREL) System Advisory Model (SAM) and are preparing for field trials of EDS films in different solar fields in the US, Chile, India and in the Middle East. A brief description of EDS film performance, construction and testing of the field-test unit, autonomous operation of the field-test unit for evaluating EDS performance in increasing energy yield, associated revenue savings, and water conservation is presented.
This report includes a compilation of several slide presentations: 1) Interatomic Potentials for Materials Science and Beyond–Advances in Machine Learned Spectral Neighborhood Analysis Potentials (Wood); 2) Agile Materials Science and Advanced Manufacturing through AI/ML (de Oca Zapiain); 3) Machine Learning for DFT Calculations (Rajamanickam); 4) Structure-preserving ML discovery of a quantum-to-continuum codesign stack (Trask); and 5) IBM Overview of Accelerated Discovery Technology (Pitera)
CoCrMo alloys are believed to be a kind of potential material for artificial cervical disc. However, the tribological properties of CoCrMo alloys against different metals and ceramics are not systematically studied. In this study, the tribological behaviors of CoCrMo alloys against metals (316L, Ti6Al4V) and ceramics (Si{sub 3}N{sub 4}, ZrO{sub 2}) were focused under dry friction and 25 wt.% newborn calf serum (NCS)-lubricated conditions using a ball-on-disc apparatus under reciprocating motion. The microstructure, composition and hardness of CoCrMo alloys were characterized using x-ray diffraction, scanning electron microscopy (SEM) and hardness testers, respectively. The contact angles of the CoCrMo alloys with deionized water and 25 wt.% NCS were measured by the OCA contact angle measuring instrument. The maximum wear width, wear depth and wear volume were measured by three-dimensional white light interference. The morphology and the EDX analysis of the wear marks on CoCrMo alloys were examined by SEM to determine the basic mechanism of friction and wear. The dominant wear mechanism in dry friction for CoCrMo alloys against all pairings was severe abrasive wear, accompanied with a lot of material transfer. Under 25 wt.% NCS-lubricated condition, the wear mechanism for CoCrMo alloys against ceramics (Si{sub 3}N{sub 4}, ZrO{sub 2}) was also mainly severe abrasive wear. However, severe abrasive wear and electrochemical corrosion occurred for the CoCrMo-316L pairing under lubrication. Severe abrasive wear, adhesive wear and electrochemical corrosion occurred for the CoCrMo-Ti6Al4V pairing under lubrication. According to the results, the tribological properties of CoCrMo alloys against ceramics were better than those against metals. The CoCrMo-ZrO{sub 2} pairing displayed the best tribological behaviors and could be taken as a potential candidate bearing material for artificial cervical disc.
Simple aqueous-solid solution models of C-S-H (calcium silicate hydrate) are widely used in studies of cement hydration and waste-cement interactions. Even without a clear structural/mechanistic basis, such thermodynamic models yield a good description of solubility data in [Ca]-C/S space, while only satisfactory in [Si]-C/S, H{sub 2}O-C/S, [Ca]-[Si] spaces and in the 0.1 < C/S < 0.8 range. Here, using a multi-site (sublattice) concept, the ideal solid solution model of C-S-H is revised to make it consistent with the Richardson-Groves structural model of C-S-H and with the modern interpretation of spectroscopic ({sup 29}Si MAS NMR) and solubility data. Consideration of two site substitutions, (1) coupled H{sub 2}OCa{sup 2+} for SiO{sub 2}H{sub 2}{sup 2+} replacement in bridging tetrahedral and adjacent interlayer sites, and (2) substitution of interstitial Ca(OH){sub 2} for a vacancy, leads to a new CSHQ model of (A,B) (C,D)X type composed of two tobermorite-like and two jennite-like end members. Because this ideal sublattice SS model cannot fit solubility data well at 0.8 < C/S < 1.1, a simpler CSH3T model is constructed from a polymeric TobH (CaO){sub 2}(SiO{sub 2}){sub 3}(H{sub 2}O){sub 5}, a dimeric T2C (CaO){sub 3}(SiO{sub 2}){sub 2}(H{sub 2}O){sub 5}, and an ordered pentameric T5C (CaO){sub 2.5}(SiO{sub 2}){sub 2.5}(H{sub 2}O){sub 5} tobermorite-like end members. This solid solution model, limited to the range 0.67 < C/S < 1.5, has a correct built-in dependence of the mean silicate chain length on C/S, yields quite realistic fits to the solubility data, and provides a basis for extensions with foreign cations whose sites in the defect-tobermorite structure of C-S-H are known. To account for C-S-H compositions with C/S > 1.5, CSHQ end members were downscaled to one tetrahedral site and used within the simple mixing model. Despite some loss of structural consistency, the solubility and mean silicate chain length data can be reproduced well with this downscaled CSHQ model, capable of temperature corrections and dependencies of density and water content in fully-hydrated C-S-H on C/S ratio. Most literature solubility data sets can be modeled at the cost of moderate adjustments of CSHQ end-member solubility products within 0.2 to 0.6 pK units.
The Lawrence Berkeley National Laboratory (LBNL) Environment, Safety, and Health (ES&H) Self-Assessment Program was established to ensure that Integrated Safety Management (ISM) is implemented institutionally and by all divisions. The ES&H Self-Assessment Program, managed by the Office of Contractor Assurance (OCA), provides for an internal evaluation of all ES&H programs and systems at LBNL. The primary objective of the program is to ensure that work is conducted safely and with minimal negative impact to workers, the public, and the environment. Self-assessment follows the five core functions and guiding principles of ISM. Self-assessment is the mechanism used to promote the continuous improvement of the Laboratory's ES&H programs. The process is described in the Environment, Safety, and Health Assurance Plan (PUB-5344) and is composed of three types of self-assessments: Division ES&H Self-Assessment, ES&H Technical Assurance Program Assessment, and Division ES&H Peer Review. The Division ES&H Self-Assessment Manual (PUB-3105) provides the framework by which divisions conduct formal ES&H self-assessments to systematically identify program deficiencies. Issue-specific assessments are designed and implemented by the divisions and focus on areas of interest to division management. They may be conducted by teams and involve advance planning to ensure that appropriate resources are available. The ES&H Technical Assurance Program Manual (PUB-913E) provides the framework for systematic reviews of ES&H programs and processes. The ES&H Technical Assurance Program Assessment is designed to evaluate whether ES&H programs and processes are compliant with guiding regulations, are effective, and are properly implemented by LBNL divisions. The Division ES&H Peer Review Manual provides the framework by which division ISM systems are evaluated and improved. Peer Reviews are conducted by teams under the direction of senior division management and focus on higher-level management issues. Peer Review teams are selected on the basis of members knowledge and experience in the issues of interest to the division director. LBNL periodically requests in-depth independent assessments of selected ES&H programs. Such assessments augment LBNL's established assessment processes and provide an objective view of ES&H program effectiveness. Institutional Findings, Observations, and Noteworthy Practices identified during independent assessments are specifically intended to help LBNL identify opportunities for program improvement. This report includes the results of the Division ES&H Self-Assessment, ES&H Technical Assurance Program Assessment, and Division ES&H Peer Review, respectively.
Organoclays, often referred to as nanoclays (NCs), are organically modified phyllosilicates. They are derived from naturally occurring clay minerals and possess unique characteristics in improving the physical and mechanical properties of polymers and rubbers. Recently, NCs have drawn interest in the modification of asphalt binders as micro-scale fillers. A few recent studies have revealed that the addition of NCs in asphalt binders improved binders’ stiffness, aging characteristics, and fatigue resistance. This study aims to evaluate the changes in moisture resistance of different NC-modified asphalt binders through the surface free energy (SFE) technique. This technique estimates SFE properties of binders and aggregates from static contact angle (SCA) data measured from an optical contact angle analyzer (OCA). Further, this study examines the changes in chemical compositions (functional groups) of binders due to the addition of selected NCs by using the Fourier Transformation Infrared (FTIR) technique. To this end, a commonly used performance grade (PG) binder (PG 64-22) modified with four different types (shape and size) of NCs, namely, Cloisite® 15 (C-15), Cloisite® 20 (C-20), Cloisite® Na+ (C-Na), and Cloisite® Ca++ (C-Ca) were evaluated in this study. The SFE data of five different types of aggregates from Oklahoma, namely, Davis Limestone (DL), Snyder Granite (SG), Dolese-Cooperton Limestone (DCL), Hanson-Davis Rhyolite (HDR), and Martin-Marietta-Mill-Creek Granite (MMMG), were used to perform their compatibility with the aforementioned NC-modified binders. The FTIR spectra revealed the presence of various alcohols, ethers, and esters in NC-modified asphalt binders. The moisture susceptibility analysis shows that the addition of NC results in an increase of SFE and cohesive energy of an asphalt binder, which are desired for improved moisture resistance. The addition of NC also shows improved compatibility in cases of all aggregates. Among the five aggregates, MMMCG showed the highest compatibility with all binders, followed by DCL, HDR, DL, and SG. In regard to NCs, the C-15 sample showed the highest compatibility followed by C-20, C-Na, and C-Ca.
Abstract Solvent‐based electrochemical CO 2 reduction (CO 2 R) enables the production of chemicals or fuels using CO 2 from a preceding absorption process. Employing previously tested CO 2 capture solvents does not ensure their suitability for either CO 2 R or integrated CO 2 absorption‐reduction. We propose solvent selection criteria that include the CO 2 solubility, kinetic constant, ionic conductivity, concentration of the bicarbonate, carbamate, and solvent cation in the CO 2 ‐loaded solution, and sustainability indicators. They are implemented for solvent selection (a) from novel, aqueous mixtures of N ‐methylcyclohexylamine (MCA) with piperazine (PZ), 2‐amino‐2‐methyl‐1‐propanol (AMP), potassium hydroxide (KOH), and potassium chloride (KCl) and (b) from aqueous monoethanolamine (MEA), AMP, KOH, MCA, and PZ solutions. Versions of a modified Kent‐Eisenberg model for strong bases, carbamate, and non‐carbamate‐forming amine solutions are developed and parameterized through experimental equilibrium measurements. CO 2 R experimental results are presented for solutions of KOH and MCA + KOH, as these indicate desired trade‐offs for CO 2 absorption and reduction.
Occupant behavior (OB) in buildings is a leading factor influencing energy use in buildings. Quantifying this influence requires the integration of OB models with building performance simulation (BPS). This study reviews approaches to representing and implementing OB models in today’s popular BPS programs, and discusses weaknesses and strengths of these approaches and key issues in integrating of OB models with BPS programs. Two of the key findings are: (1) a common data model is needed to standardize the representation of OB models, enabling their flexibility and exchange among BPS programs and user applications; the data model can be implemented using a standard syntax (e.g., in the form of XML schema), and (2) a modular software implementation of OB models, such as functional mock-up units for co-simulation, adopting the common data model, has advantages in providing a robust and interoperable integration with multiple BPS programs. Such common OB model representation and implementation approaches help standardize the input structures of OB models, enable collaborative development of a shared library of OB models, and allow for rapid and widespread integration of OB models with BPS programs to improve the simulation of occupant behavior and quantification of their impact on building performance.
The influence of the internal structure at micrometer length scales on the deformation of polycrystalline materials can be effectively captured using crystal plasticity finite element methods (CPFEM). However, the complexity and nonlinearity of the deformation equations CPFEM solves demand significant computational power and resources to achieve accurate predictions, limiting its broader application. To address this challenge, we have identified a reduced-order representation of the complex data in order to establish a computationally efficient reduced-order models (ROM) and drastically reduce the computational expense of CPFEM. Specifically, in this work, we developed a parametric, data-driven, and non-intrusive ROM framework for CPFEM using proper orthogonal decomposition (POD) and sparse variational Gaussian process (SVGP) regression for single-crystal microstructures under tensile loading conditions. The developed protocol enables one to compress field into a latent/low-dimensional space described by principal component analysis (PCA) via the singular value decomposition (SVD) algorithm. As a result, the high-dimensional data are reduced to a significantly smaller amount of dimensions with POD bases and POD coefficients. Furthermore, we deployed an ensemble of SVGPs—extended from the classical Gaussian process (GP) regression for scalability and handling big data—in a massively parallel manner to train and predict latent POD coefficients using known POD bases from a set of previously obtained simulations results. Lastly, using the predicted POD coefficients, we reconstructed the full-field results and showed reasonable agreement compared with the true values obtained from running CPFEM. The developed framework is validated with a set of CPFEM simulations of a single embedded void in single-crystal aluminum alloy. While the framework is broadly applicable, this work specifically focuses on single-crystal microstructures, a single load case (e.g., tensile), and a specific void geometry (spherical).
Here, we have explored data-driven methods for material microstructure quantification that improve sensitivity to microstructural changes compared to traditional approaches. The methods integrate multiple microstructural properties, including grain morphology, crystallographic orientation, and material phase information. The simpler method employs maps of the Euclidean distance transformation metric to evaluate the morphology of grain boundary networks. The more intensive approach employs generalized spherical harmonic mapping for crystallographic orientations, per-pixel phase information, and a variational auto-encoder for dimensionality reduction and results in a multidimensional clustering of by microstructure similarity. Applied to an experimental dataset of additively manufactured steel, both methods detected slight variations in samples produced under nominally identical processing conditions. Both methods were able to distinguish between samples from multiple (nominally identical) builds, while the generalized spherical harmonics-based method could additionally cluster data samples rotated at two orientations on the build plate. The improved sensitivity of the methods, demonstrated through comparison with traditional microstructure characterization techniques, offers advantages for microstructure quantification and comparisons in advanced manufacturing applications.
Generative modeling has opened new avenues for solving previously intractable materials design problems. However, these new opportunities are accompanied by a drastic increase in the required amount of training data. This is in stark juxtaposition to the high expense and difficulty in curating such large materials datasets. In this work, we propose a novel framework for integrating generative models within an active learning loop. Further, this enables the training of generative models with datasets significantly smaller than what has previously been demonstrated, providing a direct route for their application in data constrained environments. The functionality of this framework is then demonstrated by addressing the challenge of designing polycrystalline textures associated with target anisotropic mechanical properties. The developed protocol exhibited a cost reduction between 14 to 18 times over a randomly sampled experimental design.
Not Available
Not Available