Search NASA⌕ Search

SEARCH · Search NASA

Results for “Level set method”

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 1,225 records · Page 68
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

An air quality index-based multistage type-2-fuzzy interval-stochastic programming model for energy and environmental systems management under multiple uncertainties

Highlights: • A multistage type-2-fuzzy interval-stochastic programming (MTIP) method is developed. • Air quality index (AQI) is introduced to manage energy and environmental systems. • AQI-MTIP model is formulated and results for Tianjin's EES are generated. • Impacts of environmental requirements on energy systems is analyzed. • Results can provide support for making effective pollutant mitigation policy. In this study, a multistage type-2-fuzzy interval-stochastic programming (MTIP) method is developed, which extends upon the existing multistage stochastic programming (MSP) by allowing uncertainties expressed as probabilistic distributions, interval values and type-2 fuzzy sets to be effectively incorporated within the optimization framework. Through coupling air quality index (AQI) with MTIP, an AQI-MTIP model is formulated for energy and environmental systems (EES) management of Tianjin. A number of scenarios based on changed AQIs are examined to analyze the impacts of environmental requirements on the city's energy system. Results indicate that (i) with the improvement of environmental requirement, utilization of clean energies (especially natural gas) is provoked markedly; (ii) PM{sub 2.5} is the primary pollutant, 64.50% of which should be reduced each period to maintain the city's air quality at a health-safe level. These findings can help decision makers adjust energy structure, make effective mitigation strategy, and gain deep insight into the relationship between energy consumption and environmental requirement.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2015

Realistic respiratory motion margins for external beam partial breast irradiation

Purpose: Respiratory margins for partial breast irradiation (PBI) have been largely based on geometric observations, which may overestimate the margin required for dosimetric coverage. In this study, dosimetric population-based respiratory margins and margin formulas for external beam partial breast irradiation are determined. Methods: Volunteer respiratory data and anterior–posterior (AP) dose profiles from clinical treatment plans of 28 3D conformal radiotherapy (3DCRT) PBI patient plans were used to determine population-based respiratory margins. The peak-to-peak amplitudes (A) of realistic respiratory motion data from healthy volunteers were scaled from A = 1 to 10 mm to create respiratory motion probability density functions. Dose profiles were convolved with the respiratory probability density functions to produce blurred dose profiles accounting for respiratory motion. The required margins were found by measuring the distance between the simulated treatment and original dose profiles at the 95% isodose level. Results: The symmetric dosimetric respiratory margins to cover 90%, 95%, and 100% of the simulated treatment population were 1.5, 2, and 4 mm, respectively. With patient set up at end exhale, the required margins were larger in the anterior direction than the posterior. For respiratory amplitudes less than 5 mm, the population-based margins can be expressed as a fraction of the extent of respiratory motion. The derived formulas in the anterior/posterior directions for 90%, 95%, and 100% simulated population coverage were 0.45A/0.25A, 0.50A/0.30A, and 0.70A/0.40A. The differences in formulas for different population coverage criteria demonstrate that respiratory trace shape and baseline drift characteristics affect individual respiratory margins even for the same average peak-to-peak amplitude. Conclusions: A methodology for determining population-based respiratory margins using real respiratory motion patterns and dose profiles in the AP direction was described. It was found that the currently used respiratory margin of 5 mm in partial breast irradiation may be overly conservative for many 3DCRT PBI patients. Amplitude alone was found to be insufficient to determine patient-specific margins: individual respiratory trace shape and baseline drift both contributed to the dosimetric target coverage. With respiratory coaching, individualized respiratory margins smaller than the full extent of motion could reduce planning target volumes while ensuring adequate coverage under respiratory motion.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2026

Reference floating wind array designs for three representative regions

This work presents the systematic development of three open-source reference floating wind array designs. The designs are tailored to representative site conditions for three regions of the United States: Humboldt Bay off the coast of California, the Gulf of Maine, and the Gulf of America. We adopted existing reference designs for the individual 15 MW turbines, semisubmersible floating platforms, substations, mooring systems, and power cables – integrating and adapting them as needed for each location. We adapted existing dynamic cable designs to use larger conductor sizes to meet the arrays' power transmission requirements, and we set up redundant mooring systems for each substation. The layout of each array is a uniform-grid design optimized to approximately minimize the levelized cost of energy (LCOE) within a square lease area while satisfying spatial constraints. These constraints ensure adequate clearances between adjacent turbines and between underwater components during the layout optimization to prevent clashing and ensure that all components reside within the lease boundaries. Substations are included to allow accounting for intra-array cable costs. They are placed within the uniform grid to maintain the navigability of the arrays. For each feasible layout considered, annual energy production and cable routing costs are calculated and updated in the LCOE objective function. After the optimization, we refined the cable routing with a mix of algorithmic and manual methods to ensure that the cables avoid mooring system components and approach the substation with adequate clearances. We confirmed the suitability of each reference array's layout by comparing the wake losses at each wind heading angle to the wind rose, observing that the optimized layouts largely avoid wake losses in the predominant wind directions. These reference arrays provide open-source baseline designs to enable future research and innovation of floating wind technology at the array scale.

17 WIND ENERGY↗
DOE OSTI2024

Glass Design Using Machine Learning Property Models with Prediction Uncertainties: Nuclear Waste Glass Formulation

The United States Department of Energy is responsible for managing the legacy nuclear waste stored in underground tanks at the Hanford Site. The waste will be separately vitrified as low-activity waste and high-level waste fractions. Waste glass formulation algorithms have been traditionally developed using partial quadratic mixture property-composition models. Recently, machine learning (ML) techniques have been used to predict glass properties and discover new glass materials for nuclear waste vitrification, and these advancements can be utilized to improve waste glass composition design. In this proof-of-principle study, ML algorithms such as Gaussian process regression (GPR) were used to interpolate glass properties (e.g., viscosity, electrical conductivity, chemical durability). After selecting appropriate sets of GPR hyper-parameters for each property, an optimization program was developed to formulate glass compositions to maximize waste loading while simultaneously satisfying property within constraints. The results of the ML-based waste loadings and glass compositions were compared to those obtained using the traditional methods. Comparing to the previous glass design framework, the ML-based optimization methods offer improved glass designs and a streamlined approach to generation of optimally designed data and near real-time updates.

glass formulation, machine learning, constraints, ↗
DOE OSTI2020

Holographic wavefront sensors

A brief historical review of the first designs of holographic wavefront sensors (WFS’s) and the concepts lying in their basis is presented. The main directions in the development of these sensors are highlighted and considered. One of these directions implies a two-stage transition from the use of several separate measuring channels with holograms filtering only one Zernike mode: first to the use of one channel with a multiplexed hologram filtering several first Zernike modes, and then to filtering the entire set of Zernike modes with the aid of one combined multiplexed hologram. Another line of research in this field (related to the first one) is the optimisation of the filter hologram structure in order to reduce cross-modulation interferences, increase multiplexing level, etc. One more line of research implements principles of dynamic holography by introducing spatial light modulators into the WFS composition. Hence, the advantages of time multiplexing of holograms can be used. The approach developed by G. Andersen’s team, aimed at adapting the holographic WFS design for operation as an element of adaptive optical system with a zonal corrector, as well as an approach implying development of hybrid holographic sensors, are considered separately. The results of the authors’ studies of holographic WFS’s with application of the methods of Fourier holography (i.e., holography of focused beams, in particular, using diffuse scatterers in a hologram recording channel) are also reported. (paper)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

Improved Decarbonization Planning through Climate Resiliency Modeling

California has set ambitious decarbonization goals concerning both emissions and the portion of energy generated by renewable resources. However, climate change poses considerable uncertainties. Variation in load is driven both by the level of electrification as well as the impact of climate change on weather. Further, climate change stands to make weather more variable, impacting not just load but generation from renewable resources. In this paper, we approach the issue of the impacts of climate change on decarbonization planning from two perspectives. First, we look at the range of decarbonization pathways through 8 pathways that account for differences in socioeconomic development, global emissions, and warming. Second, we develop a more robust way of ensuring the reliability of energy resources planning than the commonly used planning reserve margin. We show that the proposed method can save between 6 and 14 billion dollars in investment and maintenance costs and outline critical policy implications concerning the reliance of power plants for satisfying planning reliability requirements, including the potential retirement of dozens of peaker power plants.

Anderson, Osten P.↗
DOE OSTI2016

SU-F-R-32: Evaluation of MRI Acquisition Parameter Variations On Texture Feature Extraction Using ACR Phantom

Purpose: To investigate the sensitivity of classic texture features to variations of MRI acquisition parameters. Methods: This study was performed on American College of Radiology (ACR) MRI Accreditation Program Phantom. MR imaging was acquired on a GE 750 3T scanner with XRM explain gradient, employing a T1-weighted images (TR/TE=500/20ms) with the following parameters as the reference standard: number of signal average (NEX) = 1, matrix size = 256×256, flip angle = 90°, slice thickness = 5mm. The effect of the acquisition parameters on texture features with and without non-uniformity correction were investigated respectively, while all the other parameters were kept as reference standard. Protocol parameters were set as follows: (a). NEX = 0.5, 2 and 4; (b).Phase encoding steps = 128, 160 and 192; (c). Matrix size = 128×128, 192×192 and 512×512. 32 classic texture features were generated using the classic gray level run length matrix (GLRLM) and gray level co-occurrence matrix (GLCOM) from each image data set. Normalized range ((maximum-minimum)/mean) was calculated to determine variation among the scans with different protocol parameters. Results: For different NEX, 31 out of 32 texture features’ range are within 10%. For different phase encoding steps, 31 out of 32 texture features’ range are within 10%. For different acquisition matrix size without non-uniformity correction, 14 out of 32 texture features’ range are within 10%; for different acquisition matrix size with non-uniformity correction, 16 out of 32 texture features’ range are within 10%. Conclusion: Initial results indicated that those texture features that range within 10% are less sensitive to variations in T1-weighted MRI acquisition parameters. This might suggest that certain texture features might be more reliable to be used as potential biomarkers in MR quantitative image analysis.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-F-J-93: Automated Segmentation of High-Resolution 3D WholeBrain Spectroscopic MRI for Glioblastoma Treatment Planning

Purpose: We report on an automated segmentation algorithm for defining radiation therapy target volumes using spectroscopic MR images (sMRI) acquired at nominal voxel resolution of 100 microliters. Methods: Wholebrain sMRI combining 3D echo-planar spectroscopic imaging, generalized auto-calibrating partially-parallel acquisitions, and elliptical k-space encoding were conducted on 3T MRI scanner with 32-channel head coil array creating images. Metabolite maps generated include choline (Cho), creatine (Cr), and N-acetylaspartate (NAA), as well as Cho/NAA, Cho/Cr, and NAA/Cr ratio maps. Automated segmentation was achieved by concomitantly considering sMRI metabolite maps with standard contrast enhancing (CE) imaging in a pipeline that first uses the water signal for skull stripping. Subsequently, an initial blob of tumor region is identified by searching for regions of FLAIR abnormalities that also display reduced NAA activity using a mean ratio correlation and morphological filters. These regions are used as starting point for a geodesic level-set refinement that adapts the initial blob to the fine details specific to each metabolite. Results: Accuracy of the segmentation model was tested on a cohort of 12 patients that had sMRI datasets acquired pre, mid and post-treatment, providing a broad range of enhancement patterns. Compared to classical imaging, where heterogeneity in the tumor appearance and shape across posed a greater challenge to the algorithm, sMRI’s regions of abnormal activity were easily detected in the sMRI metabolite maps when combining the detail available in the standard imaging with the local enhancement produced by the metabolites. Results can be imported in the treatment planning, leading in general increase in the target volumes (GTV60) when using sMRI+CE MRI compared to the standard CE MRI alone. Conclusion: Integration of automated segmentation of sMRI metabolite maps into planning is feasible and will likely streamline acceptance of this new acquisition modality in clinical practice.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Energy Distribution of Pickup Ions at the Solar Wind Termination Shock

In-situ measurements taken by the Voyager 2 spacecraft suggest that the solar wind termination shock is significantly affected by the presence of pickup ions that are produced in the inner heliosphere due to charge exchange between interstellar neutrals and the solar wind ions. We use a fully kinetic particle-in-cell method to self-consistently simulate the shock with all physical properties available from Voyager 2 . We have performed a set of simulations with varying velocity distribution functions for the pickup ions, since it was not determined by Voyager's measurements. We show that the measurements suggest that the pickup ions upstream of the shock are more energetic than generally believed. If their velocity distribution function assumes a filled-shell shape in the wind frame, the maximum cutoff speed for the pickup ions should be $≳$650 km s -1 in order to reproduce the measurements, which is almost twice the local wind speed. We suggest that pickup ions upstream of the shock are energized by adiabatic compression of the solar wind plasma as well as due to an enhanced level of turbulence in a broad foreshock region.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Structure Perception in 3D Point Clouds: Preprint

Understanding human perception is critical to the design of ef- fective visualizations. The relative benefits of using 2D versus 3D techniques for data visualization is a complex decision space, with varying levels of uncertainty and disagreement in both the liter- ature and in practice. This study aims to add easily reproducible, empirical evidence on the role of depth cues in perceiving structures or patterns in 3D point clouds. We describe a method to synthesize a 3D point cloud that contains a 3D structure, where 2D projec- tions of the data strongly resemble a Gaussian distribution. We performed a within-subjects structure identification study with 128 participants that compared scatterplot matrices (canonical 2D projections) and 3D scatterplots under three types of motion: rota- tion, xy-translation, and z-translation. We found that users could consistently identify three separate hidden structures under ro- tation, while those structures remained hidden in the scatterplot matrices and under translation. This work contributes a set of 3D point clouds that provide definitive examples of 3D patterns per- ceptible in 3D scatterplots under rotation but imperceptible in 2D scatterplots.

data analysis↗
DOE OSTI2024

Exploration of signal processing methods for superconducting magnet and quench data

Quenching is the phenomenon of a superconducting magnetic material carrying current transitioning into a regular conducting material. This may cause severe and irreparable damage to the superconductor due to Joule heating. The Magnet Department at Fermi National Accelerator Laboratory (FNAL) has acquired experimental data through quench antenna arrays that are recorded when the quench is detected. These data are in terms of voltage signals that are sampled at 100kHz for several minutes. There are multiple channels and each channel provides a data set of more than 20 million observations, while there is one channel, called the trigger channel which shows the time when quench is detected. Despite some advancements that were made including machine learning, data complexity still shadows the progress. In this work, we studied a multi-resolution analysis of the quench antenna data through the Haar wavelet transform. In particular, we applied the maximally overlapped discrete w avelet transform (MODWT) of a suitable level L to the given data and then projected it onto the wavelet basis. This decomposes a given signal (Original data) $x ϵ \mathbb{R}^N$ into $L + 1$ subspaces of $\mathbb{R}^N$. One of the subspaces called the approximation, captures the trend of the signal, and the others, called the details, capture the fluctuations at different frequency bands. This decomposition provides a clear trend of the data at a suitable level and also various activities (spikes) are seen in the details of the decomposition at every level. These spikes might reveal some information about the quench under investigation but in any case, give information about magnet behavior. Also, this decomposition is seen to be very useful in removing noise present in the data due to the source or mechanism of the experiment.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2016

SU-D-BRC-02: Application of Six Sigma Approach to Improve the Efficiency of Patient-Specific QA in Proton Therapy

Purpose: To show how the Six Sigma DMAIC (Define-Measure-Analyze-Improve-Control) can be used for improving and optimizing the efficiency of patient-specific QA process by designing site-specific range tolerances. Methods: The Six Sigma tools (process flow diagram, cause and effect, capability analysis, Pareto chart, and control chart) were utilized to determine the steps that need focus for improving the patient-specific QA process. The patient-specific range QA plans were selected according to 7 treatment site groups, a total of 1437 cases. The process capability index, Cpm was used to guide the tolerance design of patient site-specific range. We also analyzed the financial impact of this project. Results: Our results suggested that the patient range measurements were non-capable at the current tolerance level of ±1 mm in clinical proton plans. The optimized tolerances were calculated for treatment sites. Control charts for the patient QA time were constructed to compare QA time before and after the new tolerances were implemented. It is found that overall processing time was decreased by 24.3% after establishing new site-specific range tolerances. The QA failure for whole process in proton therapy would lead up to a 46% increase in total cost. This result can also predict how costs are affected by changes in adopting the tolerance design. Conclusion: We often believe that the quality and performance of proton therapy can easily be improved by merely tightening some or all of its tolerance requirements. This can become costly, however, and it is not necessarily a guarantee of better performance. The tolerance design is not a task to be undertaken without careful thought. The Six Sigma DMAIC can be used to improve the QA process by setting optimized tolerances. When tolerance design is optimized, the quality is reasonably balanced with time and cost demands.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

SU-F-R-54: CT-Texture Based Early Tumor Treatment Response Assessment During Radiation Therapy Delivery: Small Cell Versus Non-Small Cell Lung Cancers

Purpose: Tumor treatment response may potentially be assessed during radiation therapy (RT) by analyzing changes in CT-textures. We investigated the different early RT-responses between small cell (SCLC) and non-small cell lung cancer (NSCLC) as assessed by CT-texture. Methods: Daily diagnostic-quality CT acquired during routine CT-guided RT using a CT-on-Rails for 13-NSCLC and 5-SCLC patients were analyzed. These patient had ages ranging from 45–78 and 38–63 years, respectively, for NSCLC and SCLC groups, and tumor-stages ranging from T2-T4, and were treated with either RT or chemotherapy and RT with 45–66Gy/ 20–34 fractions. Gross-tumor volume (GTV) contour was generated on each daily CT by populating GTV contour from simulation to daily CTs with manual editing if necessary. CT-texture parameters, such as Hounsfield Unit (HU) histogram, mean HU, skewness, kurtosis, entropy, and short-run high-gray level emphasis (SRHGLE), were calculated in GTV from each daily CT-set using an in house software tool. Difference in changes of these texture parameters during RT between NSCLC and SCLC was analyzed and compared with GTV volume changes. Results: Radiation-induced changes in CT-texture were different between SCLC and NSCLC. Average changes from first to the last fractions for NSCLC and SCLC in GTV were 28±10(12–44) and 30±15(11–47) HU (mean HU reduction), 12.7% and 18.3% (entropy), 50% and 55% (SRHGLE), 19% and 22% (kurtosis), and 5.2% and 22% (skewness), respectively. Good correlation in kurtosis changes and GTV was seen (R{sup 2}=0.8923) for SCLC, but not for NSCLC (R{sup 2}=0.4748). SCLC had better correlations between GTV volume reduction and entropy (SCLC R{sup 2}=0.847; NSCLC R{sup 2}=0.6485), skewness (SCLC R{sup 2}=0.935; NSCLC R{sup 2}=0.7666), or SRHGLE (SCLC R{sup 2}=0.9619; NSCLC R{sup 2}=0.787). Conclusion: NSCLC and SCLC exhibited different early RT-responses as assessed by CT-texture changes during RT-delivery. The observed larger changes in various CT-texture parameters for SCLC indicate that SCLC may respond to RT more rapid than NSCLC.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

flowVS: channel-specific variance stabilization in flow cytometry

Abstract Background Comparing phenotypes of heterogeneous cell populations from multiple biological conditions is at the heart of scientific discovery based on flow cytometry (FC). When the biological signal is measured by the average expression of a biomarker, standard statistical methods require that variance be approximately stabilized in populations to be compared. Since the mean and variance of a cell population are often correlated in fluorescence-based FC measurements, a preprocessing step is needed to stabilize the within-population variances. Results We present a variance-stabilization algorithm, called flowVS, that removes the mean-variance correlations from cell populations identified in each fluorescence channel. flowVS transforms each channel from all samples of a data set by the inverse hyperbolic sine (asinh) transformation. For each channel, the parameters of the transformation are optimally selected by Bartlett’s likelihood-ratio test so that the populations attain homogeneous variances. The optimum parameters are then used to transform the corresponding channels in every sample. flowVS is therefore an explicit variance-stabilization method that stabilizes within-population variances in each channel by evaluating the homoskedasticity of clusters with a likelihood-ratio test. With two publicly available datasets, we show that flowVS removes the mean-variance dependence from raw FC data and makes the within-population variance relatively homogeneous. We demonstrate that alternative transformation techniques such as flowTrans, flowScape, logicle, and FCSTrans might not stabilize variance. Besides flow cytometry, flowVS can also be applied to stabilize variance in microarray data. With a publicly available data set we demonstrate that flowVS performs as well as the VSN software, a state-of-the-art approach developed for microarrays. Conclusions The homogeneity of variance in cell populations across FC samples is desirable when extracting features uniformly and comparing cell populations with different levels of marker expressions. The newly developed flowVS algorithm solves the variance-stabilization problem in FC and microarrays by optimally transforming data with the help of Bartlett’s likelihood-ratio test. On two publicly available FC datasets, flowVS stabilizes within-population variances more evenly than the available transformation and normalization techniques. flowVS-based variance stabilization can help in performing comparison and alignment of phenotypically identical cell populations across different samples. flowVS and the datasets used in this paper are publicly available in Bioconductor.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Streamflow measurements from four sites on the Tuolumne River in Yosemite National Park from Water Years 2002 to 2021

Regions with remote and complex terrain experience spatially varying streamflow patterns, but are often poorly sampled due to difficult access. This data package includes streamflow measurements collected using low-visibility and low-impact installations at four sites on the Tuolumne River in Yosemite National Park, for water years 2002 to 2021. The resulting data set offers a unique opportunity to explore hydrologic processes in complex terrain.This data package contains half-hourly recordings of unvented pressure, vented pressure, and water temperature are measured and used to estimate discharge and stage height. Discharge flags provide insight into data anomalies. This dataset is formatted in accordance with ESS-Dive's Hydrologic Monitoring and File Level Metadata Formats. It contains the following files:1) Folder containing four csv files of time series streamflow measurements (unvented pressure, vented pressure, estimated discharge, water temperature, stage height, and discharge flag) from four locations on the Tuolumne River2) Data dictionary (dd.csv) containing units, definitions, human readable column names, and data type for all column headers throughout the dataset3) File-level metadata (FLMD.csv) containing metadata for files contained in the dataset4) Installation methods (InstallationMethods.csv) containing metadata on sensor installation

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

The Mira–Titan Universe – IV. High-precision power spectrum emulation

Modern cosmological surveys are delivering data sets characterized by unprecedented quality and statistical completeness; this trend is expected to continue in the future as new ground- and space-based surveys come online. In order to maximally extract cosmological information from these observations, matching theoretical predictions are needed. At low redshifts, the surveys probe the non-linear regime of structure formation where cosmological simulations are the primary means of obtaining the required information. The computational cost of sufficiently resolved large-volume simulations makes it prohibitive to run very large ensembles. Nevertheless, precision emulators built on a tractable number of high-quality simulations can be used to build very fast prediction schemes to enable a variety of cosmological inference studies. We have recently introduced the Mira–Titan Universe simulation suite designed to construct emulators for a range of cosmological probes. This gravity-only set of simulations covers the standard six cosmological parameters {ω m , ω b , σ 8 , $h, n_s, w_0$} and, in addition, includes massive neutrinos and a dynamical dark energy equation of state {ω ν , $w_a$}. In this paper, we present the final emulator for the matter power spectrum based on 111 cosmological simulations, each covering a (2.1 Gpc) 3 volume and evolving 3200 3 particles. In this work, an additional set of 1776 lower resolution simulations and TimeRG perturbation theory results for the power spectrum are used to cover scales straddling the linear to mildly non-linear regimes (maximum wavenumber $\textit{k}$ = 5 Mpc –1 ). The emulator provides predictions at the 2–3 percent level of accuracy over a wide range of cosmological parameters and is publicly released as part of this paper.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Multilayer Divide-Expand-Consolidate Coupled-Cluster Method: Demonstrative Calculations of the Adsorption Energy of Carbon Dioxide in the Mg-MOF-74 Metal–Organic Framework

The implementation and evaluation of a multilayer extension of the divide-expand-consolidate (DEC) scheme within the LSDalton program is presented. The DEC scheme is a linear-scaling, fragmentation-based local coupled-cluster (CC) method that provides a means of overcoming the scaling wall associated with canonical CC electronic structure calculations on large molecular systems. Taking advantage of the local nature of correlation effects, the correlation energy for the full molecule is calculated from a set of independent fragments using localized molecular orbitals. However, when only a small subsystem of a larger system is of interest, for example, adsorption sites or catalytically active sites, the majority of the computational time may be spent evaluating the correlation energy of fragments which have little effect on the properties in the area of interest (AOI). The multilayer DEC (ML-DEC) scheme addresses this by taking advantage of the independent nature of the fragments in order to evaluate the correlation energy of various regions of the system at different levels of theory. Regions far from the AOI are evaluated at lower (cheaper) levels of theory such as Hartree–Fock (HF) or Møller–Plesset second-order perturbation theory (MP2), while the area immediately surrounding the AOI is treated with a higher level CC model. Through the ML-DEC scheme, the computational cost of CC calculations on these types of systems can be significantly reduced while maintaining the accuracy of higher-level calculations. Results from HF/RI-MP2 and RI-MP2/CCSD ML-DEC calculations of the binding energy of a fatty acid dimer are presented. We find that the ML-DEC scheme is capable of reproducing DEC energy differences at a target level of theory, provided that the region treated at the target level of theory is chosen to be sufficiently large. Time-to-solution is found to be significantly reduced, particularly in the RI-MP2/CCSD calculations. Finally, the ML-DEC scheme is applied to the calculation of CO 2 adsorption in a Mg-MOF-74 channel.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Resilient Control of Networked Microgrids Using Vertical Federated Reinforcement Learning: Designs and Real-Time Test-Bed Validations

Improving system-level resiliency of networked microgrids against adversarial cyber-attacks is an important aspect in the current regime of increased inverter-based resources (IBRs). To achieve that, this paper contributes in designing a hierarchical control layer, in conjunction with the existing control layers, resilient to adversarial attack signals. Considering model complexities, unknown dynamical behaviors of IBRs, and privacy issues regarding data sharing in multi-party-owned microgrids, designing such a control layer is non-trivial. Here, to tackle these issues, a novel federated reinforcement learning (Fed-RL) method is proposed. To grasp the interconnected dynamics of networked microgrids, the paper develops Federated Soft Actor-Critic (FedSAC) algorithm following the vertical structure of implementing Fed-RL. Next, utilizing the OpenAI Gym interface, we built a custom set-up in GridLAB-D/HELICS co-simulation platform, named Resilient RL Co-simulation (ResRLCoSIM), to train the RL agents with IEEE 123-bus benchmark comprising 3 interconnected microgrids. Finally, the learned policies in the simulation are transferred to the real-time hardware-in-the-loop (HIL) test-bed developed using the high-fidelity Hypersim platform. Finally, experiments show that the simulator-trained RL controllers achieve desirable performance with the test-bed platform, validating the minimization of the sim-to-real gap.

24 POWER TRANSMISSION AND DISTRIBUTION↗