Search NASA⌕ Search

SEARCH · Search NASA

Results for “Gaussian Process Models”

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 1,117 records · Page 62

A robust estimator of mutual information for deep learning interpretability

Abstract We develop the use of mutual information (MI), a well-established metric in information theory, to interpret the inner workings of deep learning (DL) models. To accurately estimate MI from a finite number of samples, we present GMM-MI (pronounced ‘Jimmie’), an algorithm based on Gaussian mixture models that can be applied to both discrete and continuous settings. GMM-MI is computationally efficient, robust to the choice of hyperparameters and provides the uncertainty on the MI estimate due to the finite sample size. We extensively validate GMM-MI on toy data for which the ground truth MI is known, comparing its performance against established MI estimators. We then demonstrate the use of our MI estimator in the context of representation learning, working with synthetic data and physical datasets describing highly non-linear processes. We train DL models to encode high-dimensional data within a meaningful compressed (latent) representation, and use GMM-MI to quantify both the level of disentanglement between the latent variables, and their association with relevant physical quantities, thus unlocking the interpretability of the latent representation. We make GMM-MI publicly available in this GitHub repository.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Modeling and optimization of single-pass laser amplifiers for high-repetition-rate laser pulses

We propose a model for a continuously pumped single-pass amplifier for continuous and pulsed laser beams. The model takes into account Gaussian shape and focusing geometry of pump and seed beam. As the full-wave simulation is complex we have developed a largely simplified numerical method that can be applied to rotationally symmetric geometries. With the tapered-shell model we treat (focused) propagation and amplification of an initially Gaussian beam in a gain crystal. The implementation can be done with a few lines of code that are given in this paper. With this code, a numerical parameter optimization is straightforward and example results are shown. We compare the results of our simple model with those of a full-wave simulation and show that they agree well. A comparison of model and experimental data also shows good agreement. We investigate in detail different regimes of amplification, namely the unsaturated, the fully saturated, and the intermediate regime. Because the amplification process is affected by spatially varying saturation and exhibits a nonlinear response against pump and seed power, no analytical expression for the expected output is available. For modeling of the amplification we employ a four-level system and show that if the fluorescence lifetime of the gain medium is larger than the inverse repetition rate of the seed beam, continuous-wave amplification can be employed to describe the amplification process of ultrashort pulse trains. We limit ourselves to this regime, which implies that if titanium:sapphire is chosen as gain medium the laser repetition rate has to be larger than a few megahertz. We show detailed simulation results for titanium:sapphire for a large parameter set.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Reconstruction of broad features in the primordial spectrum and inflaton potential from Planck

With the recently published Cosmic Microwave Background data from Planck we address the optimized binning of the primordial power spectrum. As an important modification to the usual binning of the primordial spectrum, along with the spectral amplitude of the bins, we allow the position of the bins also to vary. This technique enables us to address the location of the possible broad physical features in the primordial spectrum with relatively smaller number of bins compared to the analysis performed earlier. This approach is in fact a reconstruction method looking for broad features in the primordial spectrum and avoiding fitting noise in the data. Performing Markov Chain Monte Carlo analysis we present samples of the allowed primordial spectra with broad features consistent with Planck data. To test how realistic it is to have step-like features in primordial spectrum we revisit an inflationary model, proposed by A. A. Starobinsky which can address the similar features obtained from the binning of the spectrum. Using the publicly available code BINGO, we numerically calculate the local f{sub NL} for this model in equilateral and arbitrary triangular configurations of wavevectors and show that the obtained non-Gaussianity for this model is consistent with Planck results. In this paper we have also considered different spectral tilts at different bins to identify the cosmological scale that the spectral index needs to have a red tilt and it is interesting to report that spectral index cannot be well constrained up to k ≈ 0.01Mpc{sup −1}.

79 ASTRONOMY AND ASTROPHYSICS↗

Learning reference governor for cycle-to-cycle combustion control with misfire avoidance in spark-ignition engines at high exhaust gas recirculation–diluted conditions

Cycle-to-cycle feedback control is employed to achieve optimal combustion phasing while maintaining high levels of exhaust gas recirculation by adjusting the spark advance and the exhaust gas recirculation valve position. The control development is based on a control-oriented model that captures the effects of throttle position, exhaust gas recirculation valve position, and spark timing on the combustion phasing. Under the assumption that in-cylinder pressure information is available, an adaptive extended Kalman filter approach is used to estimate the exhaust gas recirculation rate into the intake manifold based on combustion phasing measurements. The estimation algorithm is adaptive since the cycle-to-cycle combustion variability (output covariance) is not known a priori and changes with operating conditions. A linear quadratic regulator controller is designed to maintain optimal combustion phasing while maximizing exhaust gas recirculation levels during load transients coming from throttle tip-in and tip-out commands from the driver. During throttle tip-outs, however, a combination of a high exhaust gas recirculation rate and an overly advanced spark, product of the dynamic response of the system, generates a sequence of misfire events. In this work, an explicit reference governor is used as an add-on scheme to the closed-loop system in order to avoid the violation of the misfire limit. The reference governor is enhanced with model-free learning which enables it to avoid misfires after a learning phase. Experimental results are reported which illustrate the potential of the proposed control strategy for achieving an optimal combustion process during highly diluted conditions for improving fuel efficiency.

42 ENGINEERING↗

Quantum confined Stark effect in Gaussian quantum wells: A tight-binding study

The main characteristics of the quantum confined Stark effect (QCSE) are studied theoretically in quantum wells of Gaussian profile. The semi-empirical tight-binding model and the Green function formalism are applied in the numerical calculations. A comparison of the QCSE in quantum wells with different kinds of confining potential is presented.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Exploring scenarios for enhanced fuel compression and performance on the National Ignition Facility with machine-learning-aided design techniques

Recent fusion experiments on the National Ignition Facility (NIF) have achieved ignition, producing multi-MJ fusion yields for input laser energies of roughly 2 MJ [Abu-Shawareb et al., Phys. Rev. Lett. 132, 065102 (2024)]. Building on the success of the target designs that have achieved ignition, we explore new implosion scenarios predicted to generate significantly more compression of the dense DT ice layer and correspondingly higher yields while preserving many of the key physics characteristics of present-day ignition designs. Our main result is a novel 3-shock implosion scheme that effectively minimizes the shock-induced entropy in the dense, accelerating DT shell and maximizes the resulting fuel compression subject to a fixed leading shock strength consistent with present-day ignition experiments, which is necessary to melt the crystalline high-density carbon ablator. Compared to the first NIF experiment to fulfill Lawson's ignition criterion, shot N210808 [Abu-Shawareb et al., Phys. Rev. Lett. 129, 075001 (2022)], our design exhibits a 40% increase in simulated peak areal density (ρR) and a 5× increase in 1D fusion yield using a 4% lighter ablator and identical DT payloads. We also present a complete integrated 2D hohlraum design and laser pulse specifications capable of generating the desired 3-shock drive and maintaining control of the low-mode capsule implosion symmetry, where the increase in simulated 2D yield relative to N210808 is > 10×. This new implosion regime was discovered with help from a machine-learning-enabled capsule design optimization framework. We outline the workflow this automated tool uses to identify improved design candidates by running several rounds of capsule simulations, constructing a surrogate model mapping input variations to key physics output quantities, and querying the resulting statistical model to propose adjustments to the x-ray drive and capsule to reach a set of physics objectives prescribed by the designer.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Multifidelity Active Learning for Failure Estimation of TRISO Nuclear Fuel

The Tristructural isotropic (TRISO)-coated particle fuel is a robust nuclear fuel proposed to be used for multiple modern nuclear technologies. Therefore, characterizing its safety is vital for the reliable operation of nuclear technologies. However, the TRISO fuel failure probabilities are small and the computational model is time consuming to evaluate them using traditional Monte Carlo-type approaches. In the paper, we present a multifidelity active learning approach to efficiently estimate small failure probabilities given an expensive computational model. Active learning suggests the next best training set for optimal subsequent predictive performance and multifidelity modeling uses cheaper low-fidelity models to approximate the high-fidelity model output. After presenting the multifidelity active learning approach, we apply it to efficiently predict TRISO failure probability and make comparisons to the reference results.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Subordinated Gaussian Processes for Solar Irradiance

Traditionally the power grid has been a one-way street with power flowing from large transmission-connected generators through the distribution network to consumers. This paradigm is changing with the introduction of distributed renewable energy resources (DERs), and with it, the way the grid is managed. There is currently a dearth of high fidelity solar irradiance datasets available to help grid researchers understand how expansion of DERs could affect future power system operations. Realistic simulations of by-the-second solar irradiances are needed to study how DER variability affects the grid. Irradiance data are highly non-stationary and non-Gaussian, and even modern time series models are challenged by their distributional properties. We develop a subordinated non-Gaussian stochastic model whose simulations realistically capture the distribution and dependence structure in measured irradiance. We illustrate our approach on a fine resolution dataset from Hawaii, where our approach outperforms standard nonlinear time series models.

MATHEMATICS AND COMPUTING,SOLAR ENERGY↗

Ion Current Density Study of the NASA-300M and NASA-457Mv2 Hall Thrusters

NASA Glenn Research Center is developing a Hall thruster in the 15-50 kW range to support future NASA missions. As a part of the process, the performance and plume characteristics of the NASA-300M, a 20-kW Hall thruster, and the NASA-457Mv2, a 50-kW Hall thruster, were evaluated. The collected data will be used to improve the fidelity of the JPL modeling tool, Hall2De, which will then be used to aid the design of the 15-50 kW Hall thruster. This paper gives a detailed overview of the Faraday probe portion of the plume characterization study. The Faraday probe in this study is a near-field probe swept radially at many axial locations downstream of the thruster exit plane. Threshold-based integration limits with threshold values of 1/e, 1/e2, and 1/e3 times the local peak current density are tried for the purpose of ion current integration and divergence angle calculation. The NASA-300M is operated at 7 conditions and the NASA-457Mv2 at 14 conditions. These conditions span discharge voltages of 200 to 500 V and discharge power of 10 to 50 kW. The ion current density profiles of the near-field plume originating from the discharge channel are discovered to strongly resemble Gaussian distributions. A novel analysis approach involving a form of ray tracing is used to determine an effective point of origin for the near-field plume. In the process of performing this analysis, definitive evidence is discovered that showed the near-field plume is bending towards the thruster centerline.

Huang, Wensheng↗

Limits on semiclassical fluctuations in the primordial universe

We place limits on semiclassical fluctuations that might be present in the primordial perturbation spectrum. These can arise if some signatures of pre-inflationary features survive the expansion, or could be created by whatever mechanism ends inflation. We study two possible models for such remnant fluctuations, both of which break the isotropy of CMB on large scales. We first consider a semiclassical fluctuation in one Fourier mode of primordial perturbations. The second scenario we analyze is a semiclassical Gaussian bump somewhere in space. These models are tested with the seven-year WMAP data using a Markov Chain Monte Carlo Bayesian analysis, and we place limits on these fluctuations. The upper bound for the amplitude of a fluctuation in a single Fourier mode is a ≤ 10{sup −4}, while for the Gaussian bump a ≤ 10{sup −3}.

79 ASTRONOMY AND ASTROPHYSICS↗

Identification and correction of temporal and spatial distortions in scanning transmission electron microscopy

Scanning transmission electron microscopy (STEM) has become the technique of choice for quantitative characterization of atomic structure of materials, where the minute displacements of atomic columns from high-symmetry positions can be used to map strain, polarization, octahedra tilts, and other physical and chemical order parameter fields. The latter can be used as inputs into mesoscopic and atomistic models, providing insight into the correlative relationships and generative physics of materials on the atomic level. However, these quantitative applications of STEM necessitate understanding the microscope induced image distortions and developing the pathways to compensate them both as part of a rapid calibration procedure for in situ imaging, and the post-experimental data analysis stage. Here, we explore the spatiotemporal structure of the microscopic distortions in STEM using multivariate analysis of the atomic trajectories in the image stacks. Based on the behavior of principal component analysis (PCA), we develop the Gaussian process (GP)-based regression method for quantification of the distortion function. The limitations of such an approach and possible strategies for implementation as a part of in-line data acquisition in STEM are discussed. Here, the analysis workflow is summarized in a Jupyter notebook that can be used to retrace the analysis and analyze the reader's data.

36 MATERIALS SCIENCE↗

Efficient computer algebra algorithms for polynomial matrices in control design

The theory of polynomial matrices plays a key role in the design and analysis of multi-input multi-output control and communications systems using frequency domain methods. Examples include coprime factorizations of transfer functions, cannonical realizations from matrix fraction descriptions, and the transfer function design of feedback compensators. Typically, such problems abstract in a natural way to the need to solve systems of Diophantine equations or systems of linear equations over polynomials. These and other problems involving polynomial matrices can in turn be reduced to polynomial matrix triangularization procedures, a result which is not surprising given the importance of matrix triangularization techniques in numerical linear algebra. Matrices with entries from a field and Gaussian elimination play a fundamental role in understanding the triangularization process. In the case of polynomial matrices, matrices with entries from a ring for which Gaussian elimination is not defined and triangularization is accomplished by what is quite properly called Euclidean elimination. Unfortunately, the numerical stability and sensitivity issues which accompany floating point approaches to Euclidean elimination are not very well understood. New algorithms are presented which circumvent entirely such numerical issues through the use of exact, symbolic methods in computer algebra. The use of such error-free algorithms guarantees that the results are accurate to within the precision of the model data--the best that can be hoped for. Care must be taken in the design of such algorithms due to the phenomenon of intermediate expressions swell.

Baras, J. S.↗

Peri-Net-Pro: the neural processes with quantified uncertainty for crack patterns

Abstract This paper develops a deep learning tool based on neural processes (NPs) called the Peri-Net-Pro, to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified uncertainties. In particular, image classification and regression studies are conducted by means of convolutional neural networks (CNNs) and NPs. First, the amount and quality of the data are enhanced by using peridynamics to theoretically compensate for the problems of the finite element method (FEM) in generating crack pattern images. Second, case studies are conducted with the prototype microelastic brittle (PMB), linear peridynamic solid (LPS), and viscoelastic solid (VES) models obtained by using the peridynamic theory. The case studies are performed to classify the images by using CNNs and determine the suitability of the PMB, LBS, and VES models. Finally, a regression analysis is performed on the crack pattern images with NPs to predict the crack patterns. The regression analysis results confirm that the variance decreases when the number of epochs increases by using the NPs. The training results gradually improve, and the variance ranges decrease to less than 0.035. The main finding of this study is that the NPs enable accurate predictions, even with missing or insufficient training data. The results demonstrate that if the context points are set to the 10th, 100th, 300th, and 784th, the training information is deliberately omitted for the context points of the 10th, 100th, and 300th, and the predictions are different when the context points are significantly lower. However, the comparison of the results of the 100th and 784th context points shows that the predicted results are similar because of the Gaussian processes in the NPs. Therefore, if the NPs are employed for training, the missing information of the training data can be supplemented to predict the results.

Mathematics↗

Modeling and optimization of photon pair sources based on spontaneous parametric down-conversion

We address the problem of efficient modeling of photon pairs generated in spontaneous parametric down-conversion and coupled into single-mode fibers. It is shown that when the range of relevant transverse wave vectors is restricted by the pump and fiber modes, the computational complexity can be reduced substantially with the help of the paraxial approximation, while retaining the full spectral characteristics of the source. This approach can serve as a basis for efficient numerical calculations or can be combined with analytically tractable approximations of the phase-matching function. We introduce here a cosine-Gaussian approximation of the phase-matching function that works for a broader range of parameters than the Gaussian model used previously. The developed modeling tools are used to evaluate characteristics of the photon pair sources such as the pair production rate and the spectral purity quantifying frequency correlations. Strategies to generate spectrally uncorrelated photons, necessary in multiphoton interference experiments, are analyzed with respect to trade-offs between parameters of the source.

74 ATOMIC AND MOLECULAR PHYSICS↗

Beyond optimization—supervised learning applications in relativistic laser-plasma experiments

We explore the applications of machine learning techniques in relativistic laser-plasma experiments beyond optimization purposes. We predict the beam charge of electrons produced in a laser wakefield accelerator given the laser wavefront change caused by a deformable mirror. Machine learning enables feature analysis beyond merely searching for an optimal beam charge, showing that specific aberrations in the laser wavefront are favored in generating higher beam charges. Supervised learning models allow characterizing the measured data quality as well as recognizing irreproducible data and potential outliers. Furthermore, we also include virtual measurement errors in the experimental data to examine the model robustness under these conditions. This work demonstrates how machine learning methods can benefit data analysis and physics interpretation in a highly nonlinear problem of relativistic laser-plasma interaction.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Gaussian Process Regression Method for Costing SmallSat Bus Capabilities

NASA is responding to the growing interest in, andcapabilities of, small satellites for science applications with an increasingnumber and frequency of Announcements of Opportunityfor small satellite space missions. Estimating the probabilitythat these mission concepts will fit within the small cost capsof these opportunities is largely driven by the probability thatone of the burgeoning number of small satellite providers will beable to meet the payload’s accommodation requirements withinthe budget for the spacecraft. JPL has collected a databasecontaining technical specifications and cost of commerciallyavailable Smallsat buses across various vendors. The primarypurpose of the database is for use in JPL’s Team X architecturestudies to inform cost estimates of a spacecraft bus which fitsthe customer’s technical requirements for their payload andmission. Customer needs are often unique and don’t alignperfectly with an off-the-shelf commercial spacecraft bus, whichmotivates the need to develop a cost model across the continuoustechnical parameter space.Al’s Bus Cost Distribution Estimator (ABCDE) uses Gaussianprocess regression (GPR) to predict commercial Smallsat spacecraftbus cost based on a subset of a customer’s technicalrequirements (payload mass, payload power, delta V, pointingcontrol, and downlink rate). GPR is implemented in ABCDE asa Bayesian method which fits an implied multivariate regressionon the technical parameters and uses kriging to intentionally“overfit” the residuals. Overfitting the residuals allows costestimates to collapse in uncertainty closer to the data pointswhile maintaining larger uncertainty intervals in regions of parameterspace with fewer data records. The data used to fit thismodel is sensitive and represents cost estimates for off-the-shelfcommercial buses. GPR simultaneously protects the sensitivityof the database and uses the sparse nature of the database toaccount for uncertainty in cost in a useful way. For a givenset of customer technical requirements, the tool provides a costestimate distribution, the percentiles of which can be interpretedas a confidence level of finding a commercial bus under a specifiedcost cap. ABCDE dramatically pushes the boundaries ofspacecraft cost estimation models due to its Bayesian methodology(accounting for the maximum uncertainty in the underlyingregression), the mathematically advanced kriging methodology,and the novelty of its application in Team X architecture tradestudies.

Austin, Alex↗

Utilization of machine learning to accelerate colloidal synthesis and discovery

Machine learning techniques are seeing increased usage for predicting new materials with targeted properties. However, widespread adoption of these techniques is hindered by the relatively greater experimental efforts required to test the predictions. Furthermore, because failed synthesis pathways are rarely communicated, it is difficult to find prior datasets that are sufficient for modeling. This work presents a closed-loop machine learning-based strategy for colloidal synthesis of nanoparticles, assuming no prior knowledge of the synthetic process, in order to show that synthetic discovery can be accelerated despite limited data availability.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

HOT GAS LINES IN T TAURI STARS

For Classical T Tauri Stars (CTTSs), the resonance doublets of N V, Si IV, and C IV, as well as the He II 1640 A line, trace hot gas flows and act as diagnostics of the accretion process. In this paper we assemble a large high-resolution, high-sensitivity data set of these lines in CTTSs and Weak T Tauri Stars (WTTSs). The sample comprises 35 stars: 1 Herbig Ae star, 28 CTTSs, and 6 WTTSs. We find that the C IV, Si IV, and N V lines in CTTSs all have similar shapes. We decompose the C IV and He II lines into broad and narrow Gaussian components (BC and NC). The most common (50%) C IV line morphology in CTTSs is that of a low-velocity NC together with a redshifted BC. For CTTSs, a strong BC is the result of the accretion process. The contribution fraction of the NC to the C IV line flux in CTTSs increases with accretion rate, from {approx}20% to up to {approx}80%. The velocity centroids of the BCs and NCs are such that V{sub BC} {approx}> 4 V{sub NC}, consistent with the predictions of the accretion shock model, in at most 12 out of 22 CTTSs. We do not find evidence of the post-shock becoming buried in the stellar photosphere due to the pressure of the accretion flow. The He II CTTSs lines are generally symmetric and narrow, with FWHM and redshifts comparable to those of WTTSs. They are less redshifted than the CTTSs C IV lines, by {approx}10 km s{sup -1}. The amount of flux in the BC of the He II line is small compared to that of the C IV line, and we show that this is consistent with models of the pre-shock column emission. Overall, the observations are consistent with the presence of multiple accretion columns with different densities or with accretion models that predict a slow-moving, low-density region in the periphery of the accretion column. For HN Tau A and RW Aur A, most of the C IV line is blueshifted suggesting that the C IV emission is produced by shocks within outflow jets. In our sample, the Herbig Ae star DX Cha is the only object for which we find a P-Cygni profile in the C IV line, which argues for the presence of a hot (10{sup 5} K) wind. For the overall sample, the Si IV and N V line luminosities are correlated with the C IV line luminosities, although the relationship between Si IV and C IV shows large scatter about a linear relationship and suggests that TW Hya, V4046 Sgr, AA Tau, DF Tau, GM Aur, and V1190 Sco are silicon-poor, while CV Cha, DX Cha, RU Lup, and RW Aur may be silicon-rich.

79 ASTRONOMY AND ASTROPHYSICS↗