Search NASA⌕ Search

SEARCH · Search NASA

Results for “robust estimation”

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 289 records · Page 16

BAYESIAN ESTIMATION OF THERMONUCLEAR REACTION RATES

The problem of estimating non-resonant astrophysical S -factors and thermonuclear reaction rates, based on measured nuclear cross sections, is of major interest for nuclear energy generation, neutrino physics, and element synthesis. Many different methods have been applied to this problem in the past, almost all of them based on traditional statistics. Bayesian methods, on the other hand, are now in widespread use in the physical sciences. In astronomy, for example, Bayesian statistics is applied to the observation of extrasolar planets, gravitational waves, and Type Ia supernovae. However, nuclear physics, in particular, has been slow to adopt Bayesian methods. We present astrophysical S -factors and reaction rates based on Bayesian statistics. We develop a framework that incorporates robust parameter estimation, systematic effects, and non-Gaussian uncertainties in a consistent manner. The method is applied to the reactions d(p, γ ){sup 3}He, {sup 3}He({sup 3}He,2p){sup 4}He, and {sup 3}He( α , γ ){sup 7}Be, important for deuterium burning, solar neutrinos, and Big Bang nucleosynthesis.

79 ASTRONOMY AND ASTROPHYSICS↗

Estimating heterotrophic respiration at large scales: challenges, approaches, and next steps

Heterotrophic respiration (HR), the aerobic and anaerobic processes mineralizing organic matter, is a key carbon flux but one impossible to measure at scales significantly larger than small experimental plots. This impedes our ability to understand carbon and nutrient cycles, benchmark models, or reliably upscale point measurements. Given that a new generation of highly mechanistic, genomic-specific global models is not imminent, we suggest that a useful step to improve this situation would be the development of "Decomposition Functional Types" (DFTs). Analogous to plant functional types (PFTs), DFTs would abstract and capture important differences in HR metabolism and flux dynamics, allowing models to efficiently group and vary these characteristics across space and time. We argue that DFTs should be initially informed by top-down expert opinion, but ultimately developed using bottom-up, data-driven analyses, and provide specific examples of potential dependent and independent variables that could be used. We present and discuss an example clustering analysis to show how model-produced annual HR can be broken into distinct groups associated with global variability in biotic and abiotic factors, and demonstrate that these groups are distinct from already-existing PFTs. A similar analysis, incorporating observational data, could form a basis for future DFTs. Finally, we suggest next steps and critical priorities: collection and synthesis of existing data; more in-depth analyses combining open data with high-performance computing; rigorous testing of analytical results; and planning by the global modeling community for decoupling decomposition from fixed site data. These are all critical steps to build a foundation for DFTs in global models, thus providing the ecological and climate change communities with robust, scalable estimates of HR at large scales.

carbon cycle↗

Timing gamma-ray pulsars with the Fermi large area telescope: Timing noise and astrometry

We have constructed timing solutions for 81 γ-ray pulsars covering more than five years of Fermi data. The sample includes 37 radio-quiet or radio-faint pulsars which cannot be timed with other telescopes. These timing solutions and the corresponding pulse times of arrival are prerequisites for further study, e.g., phase-resolved spectroscopy or searches for mode switches. Many γ-ray pulsars are strongly affected by timing noise (TN), and we present a new method for characterizing the noise process and mitigating its effects on other facets of the timing model. We present an analysis of TN over the population using a new metric for characterizing its strength and spectral shape, namely, its time-domain correlation. The dependence of the strength on ν and ν-dot is in good agreement with previous studies. We find that noise process power spectra S(f) for unrecycled pulsars are steep, with strong correlations over our entire data set and spectral indices S(f)∝f{sup −α} of α ∼ 5–9. One possible explanation for these results is the occurrence of unmodeled, episodic “microglitches.” Finally, we show that our treatment of TN results in robust parameter estimation, and in particular we measure a precise timing position for each pulsar. We extensively validate our results with multi-wavelength astrometry, and using our updated position, we firmly identify the X-ray counterpart of PSR J1418−6058.

79 ASTRONOMY AND ASTROPHYSICS↗

Evaluation of the Repeatability of the Delta Q Duct Leakage Testing TechniqueIncluding Investigation of Robust Analysis Techniques and Estimates of Weather Induced Uncertainty

The DeltaQ test is a method of estimating the air leakage from forced air duct systems. Developed primarily for residential and small commercial applications it uses the changes in blower door test results due to forced air system operation. Previous studies established the principles behind DeltaQ testing, but raised issues of precision of the test, particularly for leaky homes on windy days. Details of the measurement technique are available in an ASTM Standard (ASTM E1554-2007). In order to ease adoption of the test method, this study answers questions regarding the uncertainty due to changing weather during the test (particularly changes in wind speed) and the applicability to low leakage systems. The first question arises because the building envelope air flows and pressures used in the DeltaQ test are influenced by weather induced pressures. Variability in wind induced pressures rather than temperature difference induced pressures dominates this effect because the wind pressures change rapidly over the time period of a test. The second question needs to answered so that DeltaQ testing can be used in programs requiring or giving credit for tight ducts (e.g., California's Building Energy Code (CEC 2005)). DeltaQ modeling biases have been previously investigated in laboratory studies where there was no weather induced changes in envelope flows and pressures. Laboratory work by Andrews (2002) and Walker et al. (2004) found biases of about 0.5% of forced air system blower flow and individual test uncertainty of about 2% of forced air system blower flow. The laboratory tests were repeated by Walker and Dickerhoff (2006 and 2008) using a new ramping technique that continuously varied envelope pressures and air flows rather than taking data at pre-selected pressure stations (as used in ASTM E1554-2003 and other previous studies). The biases and individual test uncertainties for ramping were found to be very close (less than 0.5% of air handler flow) to those found in for the pressure station approach. Walker and Dickerhoff also included estimates of DeltaQ test repeatability based on the results of field tests where two houses were tested multiple times. The two houses were quite leaky (20-25 Air Changes per Hour at 50Pa (0.2 in. water) (ACH50)) and were located in the San Francisco Bay area. One house was tested on a calm day and the other on a very windy day. Results were also presented for two additional houses that were tested by other researchers in Minneapolis, MN and Madison, WI, that had very tight envelopes (1.8 and 2.5 ACH50). These tight houses had internal duct systems and were tested without operating the central blower--sometimes referred to as control tests. The standard deviations between the multiple tests for all four houses were found to be about 1% of the envelope air flow at 50 Pa (0.2 in. water) (Q50) that led to the suggestion of this as a rule of thumb for estimating DeltaQ uncertainty. Because DeltaQ is based on measuring envelope air flows it makes sense for uncertainty to scale with envelope leakage. However, these tests were on a limited data set and one of the objectives of the current study is to increase the number of tested houses. This study focuses on answering two questions: (1) What is the uncertainty associated with changes in weather (primarily wind) conditions during DeltaQ testing? (2) How can these uncertainties be reduced? The first question is addressing issues of repeatability. To study this five houses were tested as many times as possible over a day. Weather data was recorded on-site--including the local windspeed. The result from these five houses were combined with the two Bay Area homes from the previous studies. The variability of the tests (represented by the standard deviation) is the repeatability of the test method for that house under the prevailing weather conditions. Because the testing was performed over a day a wide range of wind speeds was achieved following typical diurnal variations of low wind in the early morning and greatest winds in the late afternoon/early evening. Typically about ten tests were performed in each house. To answer the second question, different data analysis techniques were investigated that looked at averaging techniques, elimination of outliers, limiting leak pressures, etc. in order to minimize the influence of changing wind conditions during the test. The objective was to find a reasonable compromise between test precision and robustness--because many of the changes to the analysis to make the test more robust limit its ability to examine wide ranges of pressures and leakage flows. A secondary goal of this study is to show that DeltaQ uncertainties are acceptable for testing low leakage systems. Therefore houses with low duct leakage were deliberately chosen to be tested.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Robust Real-Time Modeling of Distribution Systems with Data-Driven Grid-Wise Observability (Final Technical Report)

The overall objective of this project is to leverage existing and emerging sensor measurements to develop data-driven observability enhancement algorithms as well as robust state estimation and parameter identification techniques to enable real-time grid-wise monitoring and modeling of loads and distributed energy resources (DERs). The project resulted in a holistic framework with the following key components: 1) data-driven and machine learning-based grid-edge visibility enhancement, 2) robust branch-current-based state estimation (BCSE), and 3) robust real-time steady-state and dynamic-state modeling of loads and DERs. The research outcomes have provided utility companies better network visibility, higher-fidelity load/DER models, and more accurate assessment of DERs’ impacts, thus, facilitating the integration of renewable energy sources. The synergistic collaborative project among Iowa State University (ISU), Argonne National Laboratory (ANL), Electric Power Research Center (EPRC), SIEMENS Industry, Alliant Energy, Cedar Falls Utilities (CFU), and Maquoketa Valley Electric Cooperative (MVEC) to leverage the team’s extensive expertise and experience in power distribution systems, state estimation, online modeling and identification, and DER integrations. The proposed frameworks have been verified with industry adopted software and attempted integrated with existing tool wherever possible

24 POWER TRANSMISSION AND DISTRIBUTION↗

Robust a posteriori error estimation for finite element approximation to H(curl) problem

In this work, we introduce a novel a posteriori error estimator for the conforming finite element approximation to the H (curl) problem with inhomogeneous media and with the right-hand side only in L 2 . The estimator is of the recovery type. Independent with the current approximation to the primary variable (the electric field), an auxiliary variable (the magnetizing field) is recovered in parallel by solving a similar H (curl) problem. An alternate way of recovery is presented as well by localizing of the error flux. The estimator is then defined as the sum of the modified element residual and the residual of the constitutive equation defining the auxiliary variable. It is proved that the estimator is approximately equal to the true error in the energy norm without the quasi-monotonicity assumption. Lastly, we present numerical results for several H (curl) interface problems.

97 MATHEMATICS AND COMPUTING↗

Robust and scalable uncertainty estimation with conformal prediction for machine-learned interatomic potentials

Abstract Uncertainty quantification (UQ) is important to machine learning (ML) force fields to assess the level of confidence during prediction, as ML models are not inherently physical and can therefore yield catastrophically incorrect predictions. Established a-posteriori UQ methods, including ensemble methods, the dropout method, the delta method, and various heuristic distance metrics, have limitations such as being computationally challenging for large models due to model re-training. In addition, the uncertainty estimates are often not rigorously calibrated. In this work, we propose combining the distribution-free UQ method, known as conformal prediction (CP), with the distances in the neural network’s latent space to estimate the uncertainty of energies predicted by neural network force fields. We evaluate this method (CP+latent) along with other UQ methods on two essential aspects, calibration, and sharpness, and find this method to be both calibrated and sharp under the assumption of independent and identically-distributed (i.i.d.) data. We show that the method is relatively insensitive to hyperparameters selected, and test the limitations of the method when the i.i.d. assumption is violated. Finally, we demonstrate that this method can be readily applied to trained neural network force fields with traditional and graph neural network architectures to obtain estimates of uncertainty with low computational costs on a training dataset of 1 million images to showcase its scalability and portability. Incorporating the CP method with latent distances offers a calibrated, sharp and efficient strategy to estimate the uncertainty of neural network force fields. In addition, the CP approach can also function as a promising strategy for calibrating uncertainty estimated by other approaches.

74 ATOMIC AND MOLECULAR PHYSICS↗

Response to “how robust are reductions in modeled estimates from GTAP-BIO of the indirect land use change induced by conventional biofuels?”

Malins et al. (2020) recently published "How robust are reductions in modeled estimates from GTAP-BIO of the indirect land use change induced by conventional biofuels?", provided their narrative from the model improvements in GTAP-BIO over time, made several critical points regarding this model, and argued that the implemented improvements in this model tended to decrease Induce Land Use Changes (ILUC) emissions. Furthermore, they also provided several critical points regarding the Carbon Calculator for Land Use Change from Biofuels Production (CCLUB) emissions model. In this response to Malins et al. we address these critiques point-by-point by providing our detailed responses to key issues: 1) the GTAP-BIO model and its improvements over time; 2) the inclusion of cropland pasture in the model; 3) the observed land use changes in the US that have been the bases for improvements in GTAP-BIO model, 4) the time trends in corn price and yield to challenge the concept of the yield to price response; 5) some sources of land intensification in crop production; 6) the FAO notifications with respect to the comparison between harvested area and arable land; and 7) the GTAP-BIO results for multiple cropping. We also provided responses to Malins et al. critical points regarding the CCLUB emissions model. We hope that this response letter will open more constructive discussion among the LUC modeling community to remain focused on the big picture regarding agriculture's role as a very effective GHG mitigation tool that can shape the new policies to govern the production and consumption of biofuels.

09 BIOMASS FUELS↗

Efficient and Optimal Attitude Determination Using Recursive Global Positioning System Signal Operations

In this paper, a new and efficient algorithm is developed for attitude determination from Global Positioning System signals. The new algorithm is derived from a generalized nonlinear predictive filter for nonlinear systems. This uses a one time-step ahead approach to propagate a simple kinematics model for attitude determination. The advantages of the new algorithm over previously developed methods include: it provides optimal attitudes even for coplanar baseline configurations; it guarantees convergence even for poor initial conditions; it is a non-iterative algorithm; and it is computationally efficient. These advantages clearly make the new algorithm well suited to on-board applications. The performance of the new algorithm is tested on a dynamic hardware simulator. Results indicate that the new algorithm accurately estimates the attitude of a moving vehicle, and provides robust attitude estimates even when other methods, such as a linearized least-squares approach, fail due to poor initial starting conditions.

Crassidis, John L.↗

Estimating uncertainty in ambient and saturation nutrient uptake metrics from nutrient pulse releases in stream ecosystems

Nutrient spiraling is an important ecosystem process characterizing nutrient transport and uptake in streams. Various nutrient addition methods are used to estimate uptake metrics; however, uncertainty in the metrics is not often evaluated. A method was developed to quantify uncertainty in ambient and saturation nutrient uptake metrics estimated from saturating pulse nutrient additions (Tracer Additions for Spiraling Curve Characterization; TASCC). Using a Monte Carlo (MC) approach, the 95% confidence interval (CI) was estimated for ambient uptake lengths (S w-amb ) and maximum areal uptake rates (U max ) based on 100,000 datasets generated from each of four nitrogen and five phosphorous TASCC experiments conducted seasonally in a forest stream in eastern Tennessee, U.S.A. Uncertainty estimates from the MC approach were compared to the CIs estimated from ordinary least squares (OLS) and non-linear least squares (NLS) models used to calculate S w-amb and U max , respectively, from the TASCC method. The CIs for Sw-amb and Umax were large, but were not consistently larger using the MC method. Despite the large CIs, significant differences (based on nonoverlapping CIs) in nutrient metrics among seasons were found with more significant differences using the OLS/NLS vs. the MC method. Lastly, we suggest that the MC approach is a robust way to estimate uncertainty, as the calculation of S w-amb and U max violates assumptions of OLS/NLS while the MC approach is free of these assumptions. The MC approach can be applied to other ecosystem metrics that are calculated from multiple parameters, providing a more robust estimate of these metrics and their associated uncertainties.

54 ENVIRONMENTAL SCIENCES↗

Towards Efficient Uncertainty estimation in deep learning for robust energy prediction in crystal materials

In material science, recent studies have started to explore the potential of using deep learning to improve property prediction from high-fidelity simulations, e.g, density functional theory (DFT). However, the design spaces are sometimes too large and intractable to sample completely. This results in a critical question that is how to evaluate the confidence and robustness of the prediction. In this paper, we propose an efficient approach to estimate uncertainty in deep learning using a single forward pass and then apply it for robust prediction of the total energy in crystal lattice structures. Our approach is built upon the deep kernel learning (DKL) that originally introduces to leverage the expressiveness of deep neural networks as input with a probabilistic prediction of Gaussian processes (GPs) as output. Existing DKL methods have difficulties in the accuracy of predictive uncertainty, training stability, and scaling to large datasets, which lead to significant barriers in real-world applications. We propose to address these challenges by using an inducing point approximate GP in feature space combined with spectral normalization as a regularization. We finally demonstrate our robust performance on an artificial example and a real-world application from materials chemistry.

Bi, Sirui↗

Emulators for Scarce and Noisy Data: Application to Auxiliary-Field Diffusion Monte Carlo for Neutron Matter

Understanding the equation of state (EOS) of pure neutron matter is necessary for interpreting multimessenger observations of neutron stars. Reliable data analyses of these observations require well-quantified uncertainties for the EOS input, ideally propagating uncertainties from nuclear interactions directly to the EOS. This, however, requires calculations of the EOS for a prohibitively larger number of nuclear Hamiltonians, solving the nuclear many-body problem for each one. Quantum Monte Carlo methods, such as auxiliary-field diffusion Monte Carlo (AFDMC), provide precise and accurate results for the neutron matter EOS, but they are very computationally expensive, making them unsuitable for the fast evaluations necessary for uncertainty propagation. Here, we employ parametric matrix models to develop fast emulators for AFDMC calculations of neutron matter and use them to directly propagate uncertainties of coupling constants in the Hamiltonian to the EOS. As these uncertainties include estimates of the effective field theory truncation uncertainty, this approach provides robust uncertainty estimates for use in astrophysical data analyses. In conclusion, this Letter will enable novel applications such as using astrophysical observations to put constraints on coupling constants for nuclear interactions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

SU-E-T-144: Bayesian Inference of Local Relapse Data Using a Poisson-Based Tumour Control Probability Model

Purpose: The purpose of this project is to develop a robust method of parameter estimation for a Poisson-based TCP model using Bayesian inference. Methods: Bayesian inference was performed using the PyMC3 probabilistic programming framework written in Python. A Poisson-based TCP regression model that accounts for clonogen proliferation was fit to observed rates of local relapse as a function of equivalent dose in 2 Gy fractions for a population of 623 stage-I non-small-cell lung cancer patients. The Slice Markov Chain Monte Carlo sampling algorithm was used to sample the posterior distributions, and was initiated using the maximum of the posterior distributions found by optimization. The calculation of TCP with each sample step required integration over the free parameter α, which was performed using an adaptive 24-point Gauss-Legendre quadrature. Convergence was verified via inspection of the trace plot and posterior distribution for each of the fit parameters, as well as with comparisons of the most probable parameter values with their respective maximum likelihood estimates. Results: Posterior distributions for α, the standard deviation of α (σ), the average tumour cell-doubling time (Td), and the repopulation delay time (Tk), were generated assuming α/β = 10 Gy, and a fixed clonogen density of 10{sup 7} cm−{sup 3}. Posterior predictive plots generated from samples from these posterior distributions are in excellent agreement with the observed rates of local relapse used in the Bayesian inference. The most probable values of the model parameters also agree well with maximum likelihood estimates. Conclusion: A robust method of performing Bayesian inference of TCP data using a complex TCP model has been established.

60 APPLIED LIFE SCIENCES↗

Estimating nonrigid motion from inconsistent intensity with robust shape features

Purpose: To develop a nonrigid motion estimation method that is robust to heterogeneous intensity inconsistencies amongst the image pairs or image sequence. Methods: Intensity and contrast variations, as in dynamic contrast enhanced magnetic resonance imaging, present a considerable challenge to registration methods based on general discrepancy metrics. In this study, the authors propose and validate a novel method that is robust to such variations by utilizing shape features. The geometry of interest (GOI) is represented with a flexible zero level set, segmented via well-behaved regularized optimization. The optimization energy drives the zero level set to high image gradient regions, and regularizes it with area and curvature priors. The resulting shape exhibits high consistency even in the presence of intensity or contrast variations. Subsequently, a multiscale nonrigid registration is performed to seek a regular deformation field that minimizes shape discrepancy in the vicinity of GOIs. Results: To establish the working principle, realistic 2D and 3D images were subject to simulated nonrigid motion and synthetic intensity variations, so as to enable quantitative evaluation of registration performance. The proposed method was benchmarked against three alternative registration approaches, specifically, optical flow, B-spline based mutual information, and multimodality demons. When intensity consistency was satisfied, all methods had comparable registration accuracy for the GOIs. When intensities among registration pairs were inconsistent, however, the proposed method yielded pronounced improvement in registration accuracy, with an approximate fivefold reduction in mean absolute error (MAE = 2.25 mm, SD = 0.98 mm), compared to optical flow (MAE = 9.23 mm, SD = 5.36 mm), B-spline based mutual information (MAE = 9.57 mm, SD = 8.74 mm) and mutimodality demons (MAE = 10.07 mm, SD = 4.03 mm). Applying the proposed method on a real MR image sequence also provided qualitatively appealing results, demonstrating good feasibility and applicability of the proposed method. Conclusions: The authors have developed a novel method to estimate the nonrigid motion of GOIs in the presence of spatial intensity and contrast variations, taking advantage of robust shape features. Quantitative analysis and qualitative evaluation demonstrated good promise of the proposed method. Further clinical assessment and validation is being performed.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Estimates of Lightning NOx Production Based on OMI NO2 Observations Over the Gulf of Mexico

We evaluate nitrogen oxide (NO(sub x) NO + NO2) production from lightning over the Gulf of Mexico region using data from the Ozone Monitoring Instrument (OMI) aboard NASAs Aura satellite along with detection efficiency-adjusted lightning data from the World Wide Lightning Location Network (WWLLN). A special algorithm was developed to retrieve the lightning NOx [(LNO(sub x)] signal from OMI. The algorithm in its general form takes the total slant column NO2 from OMI and removes the stratospheric contribution and tropospheric background and includes an air mass factor appropriate for the profile of lightning NO(sub x) to convert the slant column LNO2 to a vertical column of LNO(sub x). WWLLN flashes are totaled over a period of 3 h prior to OMI overpass, which is the time an air parcel is expected to remain in a 1 deg. x 1 deg. grid box. The analysis is conducted for grid cells containing flash counts greater than a threshold value of 3000 flashes that yields an expected LNO(sub x) signal greater than the background. Pixels with cloud radiance fraction greater than a criterion value (0.9) indicative of highly reflective clouds are used. Results for the summer seasons during 2007-2011 yield mean LNO(sub x) production of approximately 80 +/- 45 mol per flash over the region for the two analysis methods after accounting for biases and uncertainties in the estimation method. These results are consistent with literature estimates and more robust than many prior estimates due to the large number of storms considered but are sensitive to several substantial sources of uncertainty.

Lightning↗