Search NASA⌕ Search

SEARCH · Search NASA

Results for “Data reduction methods”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 271 records · Page 15

Learning to Solve Large-Scale Security-Constrained Unit Commitment Problems

Security-constrained unit commitment (SCUC) is a fundamental problem in power systems and electricity markets. In practical settings, SCUC is repeatedly solved via mixed-integer linear programming (MIP), sometimes multiple times per day, with only minor changes in input data. In this work, we propose a number of machine learning techniques to effectively extract information from previously solved instances in order to significantly improve the computational performance of MIP solvers when solving similar instances in the future. Based on statistical data, we predict redundant constraints in the formulation, good initial feasible solutions, and affine subspaces where the optimal solution is likely to lie, leading to a significant reduction in problem size. Computational results on a diverse set of realistic and large-scale instances show that using the proposed techniques, SCUC can be solved on average 4.3 times faster with optimality guarantees and 10.2 times faster without optimality guarantees, with no observed reduction in solution quality. Out-of-distribution experiments provide evidence that the method is somewhat robust against data-set shift. Summary of Contribution. The paper describes a novel computational method, based on a combination of mixed-integer linear programming (MILP) and machine learning (ML), to solve a challenging and fundamental optimization problem in the energy sector. The method advances the state-of-the-art, not only for this particular problem, but also, more generally, in solving discrete optimization problems via ML. We expect that the techniques presented can be readily used by practitioners in the energy sector and adapted, by researchers in other fields, to other challenging operations research problems that are solved routinely.

Machine Learning↗

Predicting Reduction Rates of Energetic Nitroaromatic Compounds Using Calculated One-Electron Reduction Potentials

The evaluation of new energetic nitroaromatic compounds (NACs) for use in green munitions formulations requires models that can predict their environmental fate. The susceptibility of energetic NACs to nitro reduction might be predicted from correlations between rate constants (k) for this reaction and one-electron reduction potentials (E1NAC) / 0.059 V, but the mechanistic implications of such correlations are inconsistent with evidence from other methods. To address this inconsistency, we have reevaluated existing kinetic data using a (non-linear) free-energy relationship (FER) based on the Marcus theory of outer-sphere electron transfer. For most reductants, the results are inconsistent with rate limitation by an initial, outer-sphere electron transfer, suggesting that the strong correlation between k and E1NAC is justified only as an empirical model. This empirical correlation was used to calibrate a new quantitative structure-activity relationship (QSAR) using previously reported values of k for non-energetic NAC reduction by Fe(II) porphyrin and newly reported values of E1NAC determined using density functional theory at the B3LYP/6-311++G(2d,2p) level with the COSMO solvation model. The QSAR was then validated for energetic NACs using newly measured kinetic data for 2,4,6-trinitrotoluene (TNT), 2,4-dinitrotoluene (2,4-DNT), and 2,4-dinitroanisole (DNAN). The data show close agreement with the QSAR, supporting its applicability to energetic NACs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Sampling Low-Dimensional Markovian Dynamics for Preasymptotically Recovering Reduced Models from Data with Operator Inference

This work introduces a method for learning low-dimensional models from data of high-dimensional black-box dynamical systems. The novelty is that the learned models are exactly the reduced models that are traditionally constructed with classical projection-based model reduction techniques. Thus, the proposed approach learns models that are guaranteed to have the well-studied properties of reduced models known from model reduction, without requiring full knowledge of the governing equations and without requiring the operators of the high-dimensional systems. The key ingredient is a new data sampling scheme to obtain re-projected trajectories of high-dimensional systems that correspond to Markovian dynamics in low-dimensional subspaces. The exact recovery of reduced models from these re-projected trajectories is guaranteed pre-asymptotically under certain conditions for finite amounts of data and for a large class of systems with polynomial nonlinear terms. Numerical results demonstrate that the low-dimensional models learned with the proposed approach match reduced models from traditional model reduction up to numerical errors in practice. In conclusion, the numerical results further indicate that low-dimensional models fitted to re-projected trajectories are predictive even in situations where models fitted to trajectories without re-projection are inaccurate and unstable.

97 MATHEMATICS AND COMPUTING↗

Enhancement of measurement sensitivity in the formation of shear interferograms of transparent plates with small residual wedging

We have proposed a method for increasing the sensitivity of measurements of the wedge angle in transparent plates. The method is based on formation of the holographic shear interferograms using a combination of 180 Degree-Sign rotation of the plate with byturn adjustment of interferograms in its images to an infinitely wide fringe. The sensitivity enhancement is due to the increased number of interference fringes in the observed images of the wedged plate, which favours the reduction of the measurement error during optical processing of the obtained interferograms. Data on the experimental validation of the proposed method are presented.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Workplace Charge Management with Aggregated Building Loads

This paper describes a workplace charge management system developed to control plug-in electric vehicle charging stations based on aggregated building loads. A system to collect information from drivers was also developed for better charge management performance since the present AC charging station standard does not provide battery state of charge information. First, simulations with uncontrolled charging data were conducted to investigate several scenarios and control methods, and then one method with the most power curtailment during peak load was selected for verification tests. This paper illustrates load reduction test results for 36 charging stations and real-time campus net load data.

25 ENERGY STORAGE↗

A probabilistic graphical model based stochastic input model construction

Model reduction techniques have been widely used in modeling of high-dimensional stochastic input in uncertainty quantification tasks. However, the probabilistic modeling of random variables projected into reduced-order spaces presents a number of computational challenges. Due to the curse of dimensionality, the underlying dependence relationships between these random variables are difficult to capture. In this work, a probabilistic graphical model based approach is employed to learn the dependence by running a number of conditional independence tests using observation data. Thus a probabilistic model of the joint PDF is obtained and the PDF is factorized into a set of conditional distributions based on the dependence structure of the variables. The estimation of the joint PDF from data is then transformed to estimating conditional distributions under reduced dimensions. To improve the computational efficiency, a polynomial chaos expansion is further applied to represent the random field in terms of a set of standard random variables. This technique is combined with both linear and nonlinear model reduction methods. Numerical examples are presented to demonstrate the accuracy and efficiency of the probabilistic graphical model based stochastic input models. - Highlights: • Data-driven stochastic input models without the assumption of independence of the reduced random variables. • The problem is transformed to a Bayesian network structure learning problem. • Examples are given in flows in random media.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Sky Residual Correction

The near-infrared is populated by numerous emission lines radiated from the Earth’s atmosphere, and these emission lines are often several orders of magnitude brighter in intensity than the typical astrophysical science target. The subtraction of these emission lines, sky subtraction, can create large systematic errors in ground-based astronomical spectra effectively limiting the number of usable resolution elements. A more effective sky subtraction, and the reduction of the systematic errors due to the sky subtraction process, is a major hurdle which ground-based astronomy must overcome to increase the amount of usable spectrum, and to be able to observe fainter scientific targets. Large high-quality data sets such as the Sloan Digital Sky Survey (SDSS) in themselves present opportunities for a reduction of the systematic sky subtraction errors through self-calibration. The sky residual correction method of Wild and Hewett is one such self-calibration technique which uses principal component analysis to reduce the systematic sky residual errors present in SDSS spectra. We apply sky residual corrections to the SDSS, Baryon Oscillation Spectroscopic Survey, and Apache Point Observatory Galactic Evolution Experiment data sets to optimize the number of subtraction components, and to quantify the reduction of the systematic errors in the science spectra. Finally as a proof of concept we use the sky-residual-corrected SDSS luminous red galaxy spectra to search for gravitationally lensed emission line Galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Industrial battery operation and utilization in the presence of electrical load uncertainty using Bayesian decision theory

Behind the meter battery storage is becoming increasing popular in all sectors, though enthusiasm has recently lagged in the industrial sector. Even though there may be many factors contributing to this including lack of innovation, prohibitive costs, and undesirable rate structures, a difficulty arises in accounting for uncertainty of electrical load in industrial facilities while still attempting to utilize battery storage as much as possible all while trying to achieve fiscal profitability. Here this study utilizes Gaussian process regression and Bayesian decision theory to organize load data and quantify electrical load uncertainty to properly and effectively discharge industrial battery storage. The study employs a simulation model to set battery load setpoints for the span of the utility billing period according to the degree of risk aversion. This combination of economic analysis according to utility billing period and utilization of degree of risk aversion to make decisions on the uncertainty of the data has not before been applied to battery storage. The method resulted in an annual average reduction of peak demand by 3.8 % at the lowest amount of savings and lowest risk aversion. The highest risk aversion resulted in an annual average reduction of peak demand of 7.5 %. The maximum reduction of peak load in any month was 13.8 % in the month of December with a relatively high risk aversion. With a the highest amount risk aversion tested, the model reduced demand ten of the twelve months of the year.

25 ENERGY STORAGE↗

Comparison of implosion core metrics: A 10 ps dilation X-ray imager vs a 100 ps gated microchannel plate [Comparison of implosion core shape observations, 10 ps dilation X-ray imager vs 100 ps gated microchannel plate]

The dilation x-ray imager (DIXI) is a high-speed x-ray framing camera that uses the pulse-dilation technique to achieve a temporal resolution of less than 10 ps. This is a 10× improvement over conventional framing cameras currently employed on the National Ignition Facility (NIF) (100 ps resolution), and otherwise only achievable with 1D streaked imaging. A side effect of the dramatically reduced gate width is the comparatively lower detected signal level. Therefore we implement a Poisson noise reduction with non-local principal component analysis method to improve the robustness of the DIXI data analysis. Furthermore, we present results on ignition-relevant experiments at the NIF using DIXI. In particular we focus on establishing that/when DIXI gives reliable shape metrics (P 0 , P 2 and P 4 Legendre modes, and their temporal evolution/swings).

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Using Machine Learning to Discern Eruption in Noisy Environments: A Case Study Using CO2-Driven Cold-Water Geyser in Chimayó, New Mexico

In this paper, we present an approach based on machine learning (ML) to distinguish eruption and precursory signals of Chimayó geyser (New Mexico, U.S.A.) under noisy environmental conditions. This geyser can be considered a natural analog of CO 2 intrusion into shallow water aquifers. By studying this geyser, we can understand upwelling of CO 2 -rich fluids from depth, which has relevance to leak monitoring in a CO 2 sequestration project. ML methods such as random forests (RFs) are known to be robust multiclass classifiers and perform well under unfavorable, noisy conditions. However, the extent of the RF method’s accuracy is poorly understood for this CO 2 -driven geysering application. The current study aims to quantify the performance of RF classifiers to discern the geyser state. Toward this goal, we first present the data collected from the seismometer that is installed near the Chimayó geyser. The seismic signals collected at this site contain different types of noises such as daily temperature variations, animal movement near the geyser, and human activity. First, we filter the signals from these noises by combining the Butterworth high-pass (BH) filter and an autoregressive (AR) method in a multilevel fashion. We show that by combining these filtering techniques in a hierarchical fashion leads to a reduction in noise in the seismic data without removing the precursors and eruption event signals. We then use RF on the filtered data to classify the state of geyser into three classes: remnant noise, precursor, and eruption states. RF classifier is constructed based on the comprehensive features extracted using the Tsfresh Python package. We show that the classification accuracy using RF on the filtered data is greater than 90%. We also evaluate the accuracy of other classical time-series methods such as dynamic time warping (DTW) on filtered data along with RF on partially filtered data in which we remove the daily trends. Classification accuracy shows that DTW performs poorly (44%) and RF on partially filtered data performs decently (87%). Denoising seismic signals from both daily trends and human activity enhances RF classifier performance by 7%. Finally, these aspects make the proposed ML framework attractive for event discrimination and signal enhancement under noisy conditions, with strong potential for application to monitor leaks in CO 2 sequestration.

58 GEOSCIENCES↗

LARGE-SCALE EXTRAGALACTIC JETS IN THE CHANDRA ERA. I. DATA REDUCTION AND ANALYSIS

In this paper, we report the first stages of an investigation into the X-ray properties of extragalactic jets. Our approach is to subject all sources for which X-ray emission has been detected by Chandra to uniform reduction procedures. Using Chandra archival data for 106 such sources, we measure X-ray fluxes in three bands and compare these to radio fluxes. We discuss the sample, the reduction methods, and present first results for the ratio of X-ray to radio flux for jet knots and hotspots. In particular, we apply statistical tests to various distributions of key observational parameters to evaluate differences between the different classes of sources. Subsequent papers will deal with various ramifications such as considerations of how well the observational data fulfill expectations of the different radiation processes proposed for the knots of FRI radio galaxies and quasars.

79 ASTRONOMY AND ASTROPHYSICS↗

Monoenergetic photon-induced fission cross-section ratio measurements for U 235 , U 238 , and Pu 239 from 9.0 to 17.0 MeV

Background: High-accuracy data are needed to advance microscopic descriptions of the nuclear fission process, to improve the predictive precision of phenomenological models, and for applications in nuclear energy and homeland security. Purpose: In this work, the main goal is to provide high-accuracy cross-section data for photofission of 235 U, 238 U, and 239 Pu in the energy range from the fission threshold to the high-energy tail of the giant dipole resonance. These new data should contribute significantly to the reduction of the systematic uncertainty in evaluated photofission cross-section databases. Method: Cross-section ratios for photofission of the “big three” actinide nuclei were measured using a quasimonoenergetic photon beam and dual-fission chambers. The measurements were performed at the HIγS facility using a photon beam produced from Compton backscattering of free-electron laser light. The dual-fission chamber enabled simultaneous counting of fission events from two targets. This method allows for cross-section ratio measurements with very small systematic errors. The largest source of systematic error in the determination of the cross-section ratios is the uncertainty in the ratio of the target thicknesses. Results: We report photon-induced fission cross-section ratios for σ[ 235 U(γ,f) / 238 U(γ,f)] , σ[235U(γ,f) / 239 Pu(γ,f)] , and σ[ 238 U(γ,f) / 239 Pu(γ,f)] at photon energies from 9.0–17.0 MeV. More than 20 data points were measured for each ratio, and the systematic uncertainties are less than 2%. Conclusion: The present photofission cross-section ratio data sets are compared to ratios computed from previous measurements, to the corresponding neutron-induced fission cross-section ratios, and to ratios computed from evaluated databases. Finally, the data obtained in this work for σ[ 235 U(γ,f) / 238 U(γ,f)] are consistent with the existing data, while for σ[ 235 U(γ,f) / 239 Pu(γ,f)] and σ[ 238 U(γ,f) / 239 Pu(γ,f)] the present data are systematically lower.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

SonicPy: a suite of programs for ultrasound pulse-echo data acquisition and analysis

Sound speed and elastic constants measurements in solids and liquids are commonly performed using the ultrasound pulse-echo technique. Recent advances have expanded the use of this technique at numerous high pressure synchrotron beamlines and offline laboratories. However, the increased experimental throughput has revealed many limitations in existing software for handling the rapid measurement and the subsequent data-reduction. Here, we report the development of a collection of computer programs for sound speed measurements using the ultrasound pulse-echo technique, compatible with stepped multi-frequency, as well as broadband-pulse, couplant-corrected methods. The programs provide a highly interactive graphical interface, enable efficient measurement, exploration and near real-time analysis of the ultrasound data, and contain features useful for working with samples under high pressure and/or high temperature. The included analysis programs can alleviate the time required for data reduction from hours to less than a minute, allowing users to make timely and informed decisions regarding the appropriate experimental parameters.

97 MATHEMATICS AND COMPUTING↗

High resolution dual detector volume-of-interest cone beam breast CT - Demonstration with a bench top system

Purpose: In this study, we used a small field high resolution detector in conjunction with a full field flat panel detector to implement and investigate the dual detector volume-of-interest (VOI) cone beam breast computed tomography (CBCT) technique on a bench-top system. The potential of using this technique to image small calcifications without increasing the overall dose to the breast was demonstrated. Significant reduction of scatter components in the high resolution projection image data of the VOI was also shown. Methods: With the regular flat panel based CBCT technique, exposures were made at 80 kVp to generate an air kerma of 6 mGys at the isocenter. With the dual detector VOI CBCT technique, a high resolution small field CMOS detector was used to scan a cylindrical VOI (2.5 cm in diameter and height, 4.5 cm off-center) with collimated x-rays at four times of regular exposure level. A flat panel detector was used for full field scan with low x-ray exposures at half of the regular exposure level. The low exposure full field image data were used to fill in the truncated space in the VOI scan data and generate a complete projection image set. The Feldkamp-Davis-Kress (FDK) filtered backprojection algorithm was used to reconstruct high resolution images for the VOI. Two scanning techniques, one breast centered and the other VOI centered, were implemented and investigated. Paraffin cylinders with embedded thin aluminum (Al) wires were imaged and used in conjunction with optically stimulated luminescence (OSL) dose measurements to demonstrate the ability of this technique to image small calcifications without increasing the mean glandular dose (MGD). Results: Using exposures that produce an air kerma of 6 mGys at the isocenter, the regular CBCT technique was able to resolve the cross-sections of Al wires as thin as 254 {mu}m in diameter in the phantom. For the specific VOI studied, by increasing the exposure level by a factor of 4 for the VOI scan and reducing the exposure level by a factor of 2 for the full filed scan, the dual-detector CBCT technique was able to resolve the cross-sections of Al wires as thin as 152 {mu}m in diameter. The CNR evaluated for the entire Al wire cross-section was found to be improved from 5.5 in regular CBCT to 14.4 and 16.8 with the breast centered and VOI centered scanning techniques, respectively. Even inside VOI center, the VOI scan resulted in significant dose saving with the dose reduced by a factor of 1.6 at the VOI center. Dose saving outside the VOI was substantial with the dose reduced by a factor of 7.3 and 7.8 at the breast center for the breast centered and VOI centered scans, respectively, when compared to full field scan at the same exposure level. The differences between the two dual detector techniques in terms of dose saving and scatter reduction were small with VOI scan at 4x exposure level and full field scan at 0.5x exposure level. The MGDs were only 94% of that from the regular CBCT scan. Conclusions: For the specific VOI studied, the dual detector VOI CBCT technique has the potential to provide high quality images inside the VOI with MGD similar to or even lower than that of full field breast CBCT. It was also found that our results were compromised by the use of inadequate detectors for the VOI scan. An appropriately selected detector would better optimize the image quality improvement that can be achieved with the VOI CBCT technique.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Tabulated data for Lignin-Derived Phenolic Compounds and Water Are Effective Cosolvents for Reductive Catalytic Fractionation

Reductive catalytic fractionation (RCF) is an effective lignin-first biorefining method to extract lignin as a stabilized oil from lignocellulosic biomass. To realize RCF at scale, process modeling has shown that minimizing the use of exogenous organic solvents is critical. To this end, here we investigate the ability of lignin-derived monomers to act as either solvents or cosolvents for RCF. We begin by examining the influence of lignin-derived aromatic compounds (4-propylguaiacol, 4-propylphenol, and propylbenzene) on RCF monomer yields and subsequently extend our analysis to mixtures of 4-propylguaiacol and either methanol or water. We demonstrate that 4-propylguaiacol is an effective solvent for lignin extraction and depolymerization during RCF, especially when used in combination with water as a cosolvent. Cosolvent mixtures of 4-propylguaiacol and water enable up to 81% lignin extraction, monomer yields up to 25 wt %, and postreaction phase separation. However, unlike methanol, water as a cosolvent fails to inhibit aromatic ring hydrogenation when conducted over Ru/C as a catalyst, potentially leading to excess hydrogen consumption in a process utilizing this approach. Nonetheless, these results suggest a promising strategy for eliminating external organic solvents from RCF by utilizing mixtures of lignin-derived compounds and water as alternative extraction solvents. This is the tabulated data for the paper

biorefining↗

The Effects of Compounded Model Size Reductions on Adversarial Robustness

Recent advances in Edge AI and Tiny Machine Learning (TinyML) have enabled the deployment of machine learning models on resource-constrained environments. However, deploying these models on edge devices, such as micro-controllers, requires significant model footprint reduction through a variety of techniques such as quantization, pruning, and clustering. While these optimization methods offer considerable advantages, they potentially introduce AI-related security vulnerabilities, particularly concerning model robustness with respect to adversarial AI attacks. Prior research has extensively examined the impact of quantization on adversarial robustness; however, the effects of alternative reduction techniques and their combinations remain understudied. This paper investigates the impact of model size reduction techniques on adversarial robustness, when applied individually and combined. We utilized Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) attacks to generate adversarial perturbations for both training and testing data, and then evaluated the models' accuracy under adversarial training conditions. Our findings revealed that reduction techniques generally diminished robustness; although, combining techniques was not found to make robustness any worse than when applied individually. Moreover, specific techniques can potentially enhance resistance to small size perturbations. This research provides insights into the trade-offs between model size reduction and security, establishing a foundation for future investigations into improving adversarial training techniques and methodologies for maintaining robustness while preserving memory footprint benefits.

Austria, Phillipe [ORNL] (ORCID:0000000236223973)↗

Uncovering interpretable relationships in high-dimensional scientific data through function preserving projections

Abstract In many fields of science and engineering, we frequently encounter experiments or simulations datasets that describe the behavior of complex systems and uncovering human interpretable patterns between their inputs and outputs via exploratory data analysis is essential for building intuition and facilitating discovery. Often, we resort to 2D embeddings for examining these high-dimensional relationships (e.g. dimensionality reduction). However, most existing embedding methods treat the dimensions as coordinates for samples in a high-dimensional space, which fail to capture the potential functional relationships, and the few methods that do take function into consideration either only focus on linear patterns or produce non-linear embeddings that are hard to interpret. To address these challenges, we proposed function preserving projections (FPP), which construct 2D linear embeddings optimized to reveal interpretable yet potentially non-linear patterns between the domain and the range of a high-dimensional function. The intuition here is that humans are good at understanding potentially non-linear patterns in 2D but unable to interpret non-linear mapping from high-dimensional space to 2D. Therefore, we should restrict the projection to linear but not the pattern we are seeking. Using FPP on real-world datasets, one can obtain fundamentally new insights about high-dimensional relationships in extremely large datasets that could not be processed with existing dimension reduction methods.

97 MATHEMATICS AND COMPUTING↗

Measuring the β-decay Properties of Neutron-rich Exotic Pm, Sm, Eu, and Gd Isotopes to Constrain the Nucleosynthesis Yields in the Rare-earth Region

Abstract The β -delayed neutron-emission probabilities of 28 exotic neutron-rich isotopes of Pm, Sm, Eu, and Gd were measured for the first time at RIKEN Nishina Center using the Advanced Implantation Detector Array (AIDA) and the BRIKEN neutron detector array. The existing β -decay half-life ( T 1/2 ) database was significantly increased toward more neutron-rich isotopes, and uncertainties for previously measured values were decreased. The new data not only constrain the theoretical predictions of half-lives and β -delayed neutron-emission probabilities, but also allow for probing the mechanisms of formation of the high-mass wing of the rare-earth peak located at A ≈ 160 in the r -process abundance distribution through astrophysical reaction network calculations. An uncertainty quantification of the calculated abundance patterns with the new data shows a reduction of the uncertainty in the rare-earth peak region. The newly introduced variance-based sensitivity analysis method offers valuable insight into the influence of important nuclear physics inputs on the calculated abundance patterns. The analysis has identified the half-lives of 168 Sm and of several gadolinium isotopes as some of the key variables among the current experimental data to understand the remaining abundance uncertainty at A = 167–172.

79 ASTRONOMY AND ASTROPHYSICS↗